diff --git a/.editorconfig b/.editorconfig index 9f73416c90..9d08a1a828 100644 --- a/.editorconfig +++ b/.editorconfig @@ -1,9 +1,9 @@ -root = true - -[*] -charset = utf-8 -indent_style = space -indent_size = 2 -end_of_line = lf -insert_final_newline = true -trim_trailing_whitespace = true +root = true + +[*] +charset = utf-8 +indent_style = space +indent_size = 2 +end_of_line = lf +insert_final_newline = true +trim_trailing_whitespace = true diff --git a/.git-blame-ignore-revs b/.git-blame-ignore-revs new file mode 100644 index 0000000000..80532ea84b --- /dev/null +++ b/.git-blame-ignore-revs @@ -0,0 +1,2 @@ +# ignore #7923 eol change and code formatting +4ac8a388347ff35f34de42c3ef4a2f81f03fb3b1 diff --git a/.gitattributes b/.gitattributes index 9a854faccd..849602b2b6 100644 --- a/.gitattributes +++ b/.gitattributes @@ -1,2 +1,3 @@ +* text=auto eol=lf /.yarn/** linguist-vendored /.yarn/releases/* binary diff --git a/.github/ISSUE_TEMPLATE/#3_others.yml b/.github/ISSUE_TEMPLATE/#3_others.yml index 7b42bcda38..8eec58b381 100644 --- a/.github/ISSUE_TEMPLATE/#3_others.yml +++ b/.github/ISSUE_TEMPLATE/#3_others.yml @@ -73,4 +73,4 @@ body: id: additional attributes: label: 附加信息 - description: 任何能让我们对您的问题有更多了解的信息，包括截图或相关链接 \ No newline at end of file + description: 任何能让我们对您的问题有更多了解的信息，包括截图或相关链接 diff --git a/.github/ISSUE_TEMPLATE/3_others.yml b/.github/ISSUE_TEMPLATE/3_others.yml index dd99048bf7..4d8a383080 100644 --- a/.github/ISSUE_TEMPLATE/3_others.yml +++ b/.github/ISSUE_TEMPLATE/3_others.yml @@ -73,4 +73,4 @@ body: id: additional attributes: label: Additional Information - description: Any other information that could help us better understand your question, including screenshots or relevant links \ No newline at end of file + description: Any other information that could help us better understand your question, including screenshots or relevant links diff --git a/.github/issue-checker.yml b/.github/issue-checker.yml index b126d85477..483e9d966f 100644 --- a/.github/issue-checker.yml +++ b/.github/issue-checker.yml @@ -9,115 +9,115 @@ labels: # skips and removes - name: skip all content: - regexes: "[Ss]kip (?:[Aa]ll |)[Ll]abels?" + regexes: '[Ss]kip (?:[Aa]ll |)[Ll]abels?' - name: remove all content: - regexes: "[Rr]emove (?:[Aa]ll |)[Ll]abels?" + regexes: '[Rr]emove (?:[Aa]ll |)[Ll]abels?' - name: skip kind/bug content: - regexes: "[Ss]kip (?:[Ll]abels? |)(?:`|)kind/bug(?:`|)" + regexes: '[Ss]kip (?:[Ll]abels? |)(?:`|)kind/bug(?:`|)' - name: remove kind/bug content: - regexes: "[Rr]emove (?:[Ll]abels? |)(?:`|)kind/bug(?:`|)" + regexes: '[Rr]emove (?:[Ll]abels? |)(?:`|)kind/bug(?:`|)' - name: skip kind/enhancement content: - regexes: "[Ss]kip (?:[Ll]abels? |)(?:`|)kind/enhancement(?:`|)" + regexes: '[Ss]kip (?:[Ll]abels? |)(?:`|)kind/enhancement(?:`|)' - name: remove kind/enhancement content: - regexes: "[Rr]emove (?:[Ll]abels? |)(?:`|)kind/enhancement(?:`|)" + regexes: '[Rr]emove (?:[Ll]abels? |)(?:`|)kind/enhancement(?:`|)' - name: skip kind/question content: - regexes: "[Ss]kip (?:[Ll]abels? |)(?:`|)kind/question(?:`|)" + regexes: '[Ss]kip (?:[Ll]abels? |)(?:`|)kind/question(?:`|)' - name: remove kind/question content: - regexes: "[Rr]emove (?:[Ll]abels? |)(?:`|)kind/question(?:`|)" + regexes: '[Rr]emove (?:[Ll]abels? |)(?:`|)kind/question(?:`|)' - name: skip area/Connectivity content: - regexes: "[Ss]kip (?:[Ll]abels? |)(?:`|)area/Connectivity(?:`|)" + regexes: '[Ss]kip (?:[Ll]abels? |)(?:`|)area/Connectivity(?:`|)' - name: remove area/Connectivity content: - regexes: "[Rr]emove (?:[Ll]abels? |)(?:`|)area/Connectivity(?:`|)" + regexes: '[Rr]emove (?:[Ll]abels? |)(?:`|)area/Connectivity(?:`|)' - name: skip area/UI/UX content: - regexes: "[Ss]kip (?:[Ll]abels? |)(?:`|)area/UI/UX(?:`|)" + regexes: '[Ss]kip (?:[Ll]abels? |)(?:`|)area/UI/UX(?:`|)' - name: remove area/UI/UX content: - regexes: "[Rr]emove (?:[Ll]abels? |)(?:`|)area/UI/UX(?:`|)" + regexes: '[Rr]emove (?:[Ll]abels? |)(?:`|)area/UI/UX(?:`|)' - name: skip kind/documentation content: - regexes: "[Ss]kip (?:[Ll]abels? |)(?:`|)kind/documentation(?:`|)" + regexes: '[Ss]kip (?:[Ll]abels? |)(?:`|)kind/documentation(?:`|)' - name: remove kind/documentation content: - regexes: "[Rr]emove (?:[Ll]abels? |)(?:`|)kind/documentation(?:`|)" + regexes: '[Rr]emove (?:[Ll]abels? |)(?:`|)kind/documentation(?:`|)' - name: skip client:linux content: - regexes: "[Ss]kip (?:[Ll]abels? |)(?:`|)client:linux(?:`|)" + regexes: '[Ss]kip (?:[Ll]abels? |)(?:`|)client:linux(?:`|)' - name: remove client:linux content: - regexes: "[Rr]emove (?:[Ll]abels? |)(?:`|)client:linux(?:`|)" + regexes: '[Rr]emove (?:[Ll]abels? |)(?:`|)client:linux(?:`|)' - name: skip client:mac content: - regexes: "[Ss]kip (?:[Ll]abels? |)(?:`|)client:mac(?:`|)" + regexes: '[Ss]kip (?:[Ll]abels? |)(?:`|)client:mac(?:`|)' - name: remove client:mac content: - regexes: "[Rr]emove (?:[Ll]abels? |)(?:`|)client:mac(?:`|)" + regexes: '[Rr]emove (?:[Ll]abels? |)(?:`|)client:mac(?:`|)' - name: skip client:win content: - regexes: "[Ss]kip (?:[Ll]abels? |)(?:`|)client:win(?:`|)" + regexes: '[Ss]kip (?:[Ll]abels? |)(?:`|)client:win(?:`|)' - name: remove client:win content: - regexes: "[Rr]emove (?:[Ll]abels? |)(?:`|)client:win(?:`|)" - + regexes: '[Rr]emove (?:[Ll]abels? |)(?:`|)client:win(?:`|)' + - name: skip sig/Assistant content: - regexes: "[Ss]kip (?:[Ll]abels? |)(?:`|)sig/Assistant(?:`|)" + regexes: '[Ss]kip (?:[Ll]abels? |)(?:`|)sig/Assistant(?:`|)' - name: remove sig/Assistant content: - regexes: "[Rr]emove (?:[Ll]abels? |)(?:`|)sig/Assistant(?:`|)" + regexes: '[Rr]emove (?:[Ll]abels? |)(?:`|)sig/Assistant(?:`|)' - name: skip sig/Data content: - regexes: "[Ss]kip (?:[Ll]abels? |)(?:`|)sig/Data(?:`|)" + regexes: '[Ss]kip (?:[Ll]abels? |)(?:`|)sig/Data(?:`|)' - name: remove sig/Data content: - regexes: "[Rr]emove (?:[Ll]abels? |)(?:`|)sig/Data(?:`|)" + regexes: '[Rr]emove (?:[Ll]abels? |)(?:`|)sig/Data(?:`|)' - name: skip sig/MCP content: - regexes: "[Ss]kip (?:[Ll]abels? |)(?:`|)sig/MCP(?:`|)" + regexes: '[Ss]kip (?:[Ll]abels? |)(?:`|)sig/MCP(?:`|)' - name: remove sig/MCP content: - regexes: "[Rr]emove (?:[Ll]abels? |)(?:`|)sig/MCP(?:`|)" + regexes: '[Rr]emove (?:[Ll]abels? |)(?:`|)sig/MCP(?:`|)' - name: skip sig/RAG content: - regexes: "[Ss]kip (?:[Ll]abels? |)(?:`|)sig/RAG(?:`|)" + regexes: '[Ss]kip (?:[Ll]abels? |)(?:`|)sig/RAG(?:`|)' - name: remove sig/RAG content: - regexes: "[Rr]emove (?:[Ll]abels? |)(?:`|)sig/RAG(?:`|)" + regexes: '[Rr]emove (?:[Ll]abels? |)(?:`|)sig/RAG(?:`|)' - name: skip lgtm content: - regexes: "[Ss]kip (?:[Ll]abels? |)(?:`|)lgtm(?:`|)" + regexes: '[Ss]kip (?:[Ll]abels? |)(?:`|)lgtm(?:`|)' - name: remove lgtm content: - regexes: "[Rr]emove (?:[Ll]abels? |)(?:`|)lgtm(?:`|)" + regexes: '[Rr]emove (?:[Ll]abels? |)(?:`|)lgtm(?:`|)' - name: skip License content: - regexes: "[Ss]kip (?:[Ll]abels? |)(?:`|)License(?:`|)" + regexes: '[Ss]kip (?:[Ll]abels? |)(?:`|)License(?:`|)' - name: remove License content: - regexes: "[Rr]emove (?:[Ll]abels? |)(?:`|)License(?:`|)" + regexes: '[Rr]emove (?:[Ll]abels? |)(?:`|)License(?:`|)' # `Dev Team` - name: Dev Team @@ -129,7 +129,7 @@ labels: # Area labels - name: area/Connectivity content: area/Connectivity - regexes: "代理|[Pp]roxy" + regexes: '代理|[Pp]roxy' skip-if: - skip all - skip area/Connectivity @@ -139,7 +139,7 @@ labels: - name: area/UI/UX content: area/UI/UX - regexes: "界面|[Uu][Ii]|重叠|按钮|图标|组件|渲染|菜单|栏目|头像|主题|样式|[Cc][Ss][Ss]" + regexes: '界面|[Uu][Ii]|重叠|按钮|图标|组件|渲染|菜单|栏目|头像|主题|样式|[Cc][Ss][Ss]' skip-if: - skip all - skip area/UI/UX @@ -150,7 +150,7 @@ labels: # Kind labels - name: kind/documentation content: kind/documentation - regexes: "文档|教程|[Dd]oc(s|umentation)|[Rr]eadme" + regexes: '文档|教程|[Dd]oc(s|umentation)|[Rr]eadme' skip-if: - skip all - skip kind/documentation @@ -161,7 +161,7 @@ labels: # Client labels - name: client:linux content: client:linux - regexes: "(?:[Ll]inux|[Uu]buntu|[Dd]ebian)" + regexes: '(?:[Ll]inux|[Uu]buntu|[Dd]ebian)' skip-if: - skip all - skip client:linux @@ -171,7 +171,7 @@ labels: - name: client:mac content: client:mac - regexes: "(?:[Mm]ac|[Mm]acOS|[Oo]SX)" + regexes: '(?:[Mm]ac|[Mm]acOS|[Oo]SX)' skip-if: - skip all - skip client:mac @@ -181,7 +181,7 @@ labels: - name: client:win content: client:win - regexes: "(?:[Ww]in|[Ww]indows)" + regexes: '(?:[Ww]in|[Ww]indows)' skip-if: - skip all - skip client:win @@ -192,7 +192,7 @@ labels: # SIG labels - name: sig/Assistant content: sig/Assistant - regexes: "快捷助手|[Aa]ssistant" + regexes: '快捷助手|[Aa]ssistant' skip-if: - skip all - skip sig/Assistant @@ -202,7 +202,7 @@ labels: - name: sig/Data content: sig/Data - regexes: "[Ww]ebdav|坚果云|备份|同步|数据|Obsidian|Notion|Joplin|思源" + regexes: '[Ww]ebdav|坚果云|备份|同步|数据|Obsidian|Notion|Joplin|思源' skip-if: - skip all - skip sig/Data @@ -212,7 +212,7 @@ labels: - name: sig/MCP content: sig/MCP - regexes: "[Mm][Cc][Pp]" + regexes: '[Mm][Cc][Pp]' skip-if: - skip all - skip sig/MCP @@ -222,7 +222,7 @@ labels: - name: sig/RAG content: sig/RAG - regexes: "知识库|[Rr][Aa][Gg]" + regexes: '知识库|[Rr][Aa][Gg]' skip-if: - skip all - skip sig/RAG @@ -233,7 +233,7 @@ labels: # Other labels - name: lgtm content: lgtm - regexes: "(?:[Ll][Gg][Tt][Mm]|[Ll]ooks [Gg]ood [Tt]o [Mm]e)" + regexes: '(?:[Ll][Gg][Tt][Mm]|[Ll]ooks [Gg]ood [Tt]o [Mm]e)' skip-if: - skip all - skip lgtm @@ -243,7 +243,7 @@ labels: - name: License content: License - regexes: "(?:[Ll]icense|[Cc]opyright|[Mm][Ii][Tt]|[Aa]pache)" + regexes: '(?:[Ll]icense|[Cc]opyright|[Mm][Ii][Tt]|[Aa]pache)' skip-if: - skip all - skip License diff --git a/.github/workflows/dispatch-docs-update.yml b/.github/workflows/dispatch-docs-update.yml new file mode 100644 index 0000000000..b9457faec6 --- /dev/null +++ b/.github/workflows/dispatch-docs-update.yml @@ -0,0 +1,27 @@ +name: Dispatch Docs Update on Release + +on: + release: + types: [released] + +permissions: + contents: write + +jobs: + dispatch-docs-update: + runs-on: ubuntu-latest + steps: + - name: Get Release Tag from Event + id: get-event-tag + shell: bash + run: | + # 从当前 Release 事件中获取 tag_name + echo "tag=${{ github.event.release.tag_name }}" >> $GITHUB_OUTPUT + + - name: Dispatch update-download-version workflow to cherry-studio-docs + uses: peter-evans/repository-dispatch@v3 + with: + token: ${{ secrets.REPO_DISPATCH_TOKEN }} + repository: CherryHQ/cherry-studio-docs + event-type: update-download-version + client-payload: '{"version": "${{ steps.get-event-tag.outputs.tag }}"}' diff --git a/.github/workflows/issue-checker.yml b/.github/workflows/issue-checker.yml index cb768531c0..45da0f6b50 100644 --- a/.github/workflows/issue-checker.yml +++ b/.github/workflows/issue-checker.yml @@ -1,4 +1,4 @@ -name: "Issue Checker" +name: 'Issue Checker' on: issues: @@ -19,7 +19,7 @@ jobs: steps: - uses: MaaAssistantArknights/issue-checker@v1.14 with: - repo-token: "${{ secrets.GITHUB_TOKEN }}" + repo-token: '${{ secrets.GITHUB_TOKEN }}' configuration-path: .github/issue-checker.yml not-before: 2022-08-05T00:00:00Z - include-title: 1 \ No newline at end of file + include-title: 1 diff --git a/.github/workflows/issue-management.yml b/.github/workflows/issue-management.yml index 59faedc04e..89ccc1fa8d 100644 --- a/.github/workflows/issue-management.yml +++ b/.github/workflows/issue-management.yml @@ -1,8 +1,8 @@ -name: "Stale Issue Management" +name: 'Stale Issue Management' on: schedule: - - cron: "0 0 * * *" + - cron: '0 0 * * *' workflow_dispatch: env: @@ -24,18 +24,18 @@ jobs: uses: actions/stale@v9 with: repo-token: ${{ secrets.GITHUB_TOKEN }} - only-labels: "needs-more-info" + only-labels: 'needs-more-info' days-before-stale: ${{ env.daysBeforeStale }} - days-before-close: 0 # Close immediately after stale - stale-issue-label: "inactive" - close-issue-label: "closed:no-response" + days-before-close: 0 # Close immediately after stale + stale-issue-label: 'inactive' + close-issue-label: 'closed:no-response' stale-issue-message: | This issue has been labeled as needing more information and has been inactive for ${{ env.daysBeforeStale }} days. It will be closed now due to lack of additional information. - + 该问题被标记为"需要更多信息"且已经 ${{ env.daysBeforeStale }} 天没有任何活动，将立即关闭。 operations-per-run: 50 - exempt-issue-labels: "pending, Dev Team" + exempt-issue-labels: 'pending, Dev Team' days-before-pr-stale: -1 days-before-pr-close: -1 @@ -45,11 +45,11 @@ jobs: repo-token: ${{ secrets.GITHUB_TOKEN }} days-before-stale: ${{ env.daysBeforeStale }} days-before-close: ${{ env.daysBeforeClose }} - stale-issue-label: "inactive" + stale-issue-label: 'inactive' stale-issue-message: | This issue has been inactive for a prolonged period and will be closed automatically in ${{ env.daysBeforeClose }} days. 该问题已长时间处于闲置状态，${{ env.daysBeforeClose }} 天后将自动关闭。 - exempt-issue-labels: "pending, Dev Team, kind/enhancement" + exempt-issue-labels: 'pending, Dev Team, kind/enhancement' days-before-pr-stale: -1 # Completely disable stalling for PRs days-before-pr-close: -1 # Completely disable closing for PRs diff --git a/.github/workflows/release.yml b/.github/workflows/release.yml index 2d60f3e75c..23d1aeca7e 100644 --- a/.github/workflows/release.yml +++ b/.github/workflows/release.yml @@ -118,38 +118,3 @@ jobs: tag: ${{ steps.get-tag.outputs.tag }} artifacts: 'dist/*.exe,dist/*.zip,dist/*.dmg,dist/*.AppImage,dist/*.snap,dist/*.deb,dist/*.rpm,dist/*.tar.gz,dist/latest*.yml,dist/rc*.yml,dist/*.blockmap' token: ${{ secrets.GITHUB_TOKEN }} - - dispatch-docs-update: - needs: release - if: success() && github.repository == 'CherryHQ/cherry-studio' # 确保所有构建成功且在主仓库中运行 - runs-on: ubuntu-latest - steps: - - name: Get release tag - id: get-tag - shell: bash - run: | - if [ "${{ github.event_name }}" = "workflow_dispatch" ]; then - echo "tag=${{ github.event.inputs.tag }}" >> $GITHUB_OUTPUT - else - echo "tag=${GITHUB_REF#refs/tags/}" >> $GITHUB_OUTPUT - fi - - - name: Check if tag is pre-release - id: check-tag - shell: bash - run: | - TAG="${{ steps.get-tag.outputs.tag }}" - if [[ "$TAG" == *"rc"* || "$TAG" == *"pre-release"* ]]; then - echo "is_pre_release=true" >> $GITHUB_OUTPUT - else - echo "is_pre_release=false" >> $GITHUB_OUTPUT - fi - - - name: Dispatch update-download-version workflow to cherry-studio-docs - if: steps.check-tag.outputs.is_pre_release == 'false' - uses: peter-evans/repository-dispatch@v3 - with: - token: ${{ secrets.REPO_DISPATCH_TOKEN }} - repository: CherryHQ/cherry-studio-docs - event-type: update-download-version - client-payload: '{"version": "${{ steps.get-tag.outputs.tag }}"}' diff --git a/.yarn/patches/@google-genai-npm-1.0.1-e26f0f9af7.patch b/.yarn/patches/@google-genai-npm-1.0.1-e26f0f9af7.patch index 49fcd73ad2..330b3b9a42 100644 --- a/.yarn/patches/@google-genai-npm-1.0.1-e26f0f9af7.patch +++ b/.yarn/patches/@google-genai-npm-1.0.1-e26f0f9af7.patch @@ -75,6397 +75,6397 @@ index c27ff1fc034001e751c6a995476013145c3ea42f..00000000000000000000000000000000 --- a/dist/web/web.d.ts +++ /dev/null @@ -1,6394 +0,0 @@ --import { Client } from '@modelcontextprotocol/sdk/client/index.js'; --import { GoogleAuthOptions } from 'google-auth-library'; -- --/** Marks the end of user activity. -- -- This can only be sent if automatic (i.e. server-side) activity detection is -- disabled. -- */ --export declare interface ActivityEnd { --} -- --/** The different ways of handling user activity. */ --export declare enum ActivityHandling { -- /** -- * If unspecified, the default behavior is `START_OF_ACTIVITY_INTERRUPTS`. -- */ -- ACTIVITY_HANDLING_UNSPECIFIED = "ACTIVITY_HANDLING_UNSPECIFIED", -- /** -- * If true, start of activity will interrupt the model's response (also called "barge in"). The model's current response will be cut-off in the moment of the interruption. This is the default behavior. -- */ -- START_OF_ACTIVITY_INTERRUPTS = "START_OF_ACTIVITY_INTERRUPTS", -- /** -- * The model's response will not be interrupted. -- */ -- NO_INTERRUPTION = "NO_INTERRUPTION" --} -- --/** Marks the start of user activity. -- -- This can only be sent if automatic (i.e. server-side) activity detection is -- disabled. -- */ --export declare interface ActivityStart { --} -- --/** Optional. Adapter size for tuning. */ --export declare enum AdapterSize { -- /** -- * Adapter size is unspecified. -- */ -- ADAPTER_SIZE_UNSPECIFIED = "ADAPTER_SIZE_UNSPECIFIED", -- /** -- * Adapter size 1. -- */ -- ADAPTER_SIZE_ONE = "ADAPTER_SIZE_ONE", -- /** -- * Adapter size 2. -- */ -- ADAPTER_SIZE_TWO = "ADAPTER_SIZE_TWO", -- /** -- * Adapter size 4. -- */ -- ADAPTER_SIZE_FOUR = "ADAPTER_SIZE_FOUR", -- /** -- * Adapter size 8. -- */ -- ADAPTER_SIZE_EIGHT = "ADAPTER_SIZE_EIGHT", -- /** -- * Adapter size 16. -- */ -- ADAPTER_SIZE_SIXTEEN = "ADAPTER_SIZE_SIXTEEN", -- /** -- * Adapter size 32. -- */ -- ADAPTER_SIZE_THIRTY_TWO = "ADAPTER_SIZE_THIRTY_TWO" --} -- --/** -- * The ApiClient class is used to send requests to the Gemini API or Vertex AI -- * endpoints. -- */ --declare class ApiClient { -- readonly clientOptions: ApiClientInitOptions; -- constructor(opts: ApiClientInitOptions); -- /** -- * Determines the base URL for Vertex AI based on project and location. -- * Uses the global endpoint if location is 'global' or if project/location -- * are not specified (implying API key usage). -- * @private -- */ -- private baseUrlFromProjectLocation; -- /** -- * Normalizes authentication parameters for Vertex AI. -- * If project and location are provided, API key is cleared. -- * If project and location are not provided (implying API key usage), -- * project and location are cleared. -- * @private -- */ -- private normalizeAuthParameters; -- isVertexAI(): boolean; -- getProject(): string | undefined; -- getLocation(): string | undefined; -- getApiVersion(): string; -- getBaseUrl(): string; -- getRequestUrl(): string; -- getHeaders(): Record; -- private getRequestUrlInternal; -- getBaseResourcePath(): string; -- getApiKey(): string | undefined; -- getWebsocketBaseUrl(): string; -- setBaseUrl(url: string): void; -- private constructUrl; -- private shouldPrependVertexProjectPath; -- request(request: HttpRequest): Promise; -- private patchHttpOptions; -- requestStream(request: HttpRequest): Promise>; -- private includeExtraHttpOptionsToRequestInit; -- private unaryApiCall; -- private streamApiCall; -- processStreamResponse(response: Response): AsyncGenerator; -- private apiCall; -- getDefaultHeaders(): Record; -- private getHeadersInternal; -- /** -- * Uploads a file asynchronously using Gemini API only, this is not supported -- * in Vertex AI. -- * -- * @param file The string path to the file to be uploaded or a Blob object. -- * @param config Optional parameters specified in the `UploadFileConfig` -- * interface. @see {@link UploadFileConfig} -- * @return A promise that resolves to a `File` object. -- * @throws An error if called on a Vertex AI client. -- * @throws An error if the `mimeType` is not provided and can not be inferred, -- */ -- uploadFile(file: string | Blob, config?: UploadFileConfig): Promise; -- /** -- * Downloads a file asynchronously to the specified path. -- * -- * @params params - The parameters for the download request, see {@link -- * DownloadFileParameters} -- */ -- downloadFile(params: DownloadFileParameters): Promise; -- private fetchUploadUrl; --} -- --/** -- * Options for initializing the ApiClient. The ApiClient uses the parameters -- * for authentication purposes as well as to infer if SDK should send the -- * request to Vertex AI or Gemini API. -- */ --declare interface ApiClientInitOptions { -- /** -- * The object used for adding authentication headers to API requests. -- */ -- auth: Auth; -- /** -- * The uploader to use for uploading files. This field is required for -- * creating a client, will be set through the Node_client or Web_client. -- */ -- uploader: Uploader; -- /** -- * Optional. The downloader to use for downloading files. This field is -- * required for creating a client, will be set through the Node_client or -- * Web_client. -- */ -- downloader: Downloader; -- /** -- * Optional. The Google Cloud project ID for Vertex AI users. -- * It is not the numeric project name. -- * If not provided, SDK will try to resolve it from runtime environment. -- */ -- project?: string; -- /** -- * Optional. The Google Cloud project location for Vertex AI users. -- * If not provided, SDK will try to resolve it from runtime environment. -- */ -- location?: string; -- /** -- * The API Key. This is required for Gemini API users. -- */ -- apiKey?: string; -- /** -- * Optional. Set to true if you intend to call Vertex AI endpoints. -- * If unset, default SDK behavior is to call Gemini API. -- */ -- vertexai?: boolean; -- /** -- * Optional. The API version for the endpoint. -- * If unset, SDK will choose a default api version. -- */ -- apiVersion?: string; -- /** -- * Optional. A set of customizable configuration for HTTP requests. -- */ -- httpOptions?: HttpOptions; -- /** -- * Optional. An extra string to append at the end of the User-Agent header. -- * -- * This can be used to e.g specify the runtime and its version. -- */ -- userAgentExtra?: string; --} -- --/** Config for authentication with API key. */ --export declare interface ApiKeyConfig { -- /** The API key to be used in the request directly. */ -- apiKeyString?: string; --} -- --/** Representation of an audio chunk. */ --export declare interface AudioChunk { -- /** Raw byets of audio data. */ -- data?: string; -- /** MIME type of the audio chunk. */ -- mimeType?: string; -- /** Prompts and config used for generating this audio chunk. */ -- sourceMetadata?: LiveMusicSourceMetadata; --} -- --/** The audio transcription configuration in Setup. */ --export declare interface AudioTranscriptionConfig { --} -- --/** -- * @license -- * Copyright 2025 Google LLC -- * SPDX-License-Identifier: Apache-2.0 -- */ --/** -- * The Auth interface is used to authenticate with the API service. -- */ --declare interface Auth { -- /** -- * Sets the headers needed to authenticate with the API service. -- * -- * @param headers - The Headers object that will be updated with the authentication headers. -- */ -- addAuthHeaders(headers: Headers): Promise; --} -- --/** Auth configuration to run the extension. */ --export declare interface AuthConfig { -- /** Config for API key auth. */ -- apiKeyConfig?: ApiKeyConfig; -- /** Type of auth scheme. */ -- authType?: AuthType; -- /** Config for Google Service Account auth. */ -- googleServiceAccountConfig?: AuthConfigGoogleServiceAccountConfig; -- /** Config for HTTP Basic auth. */ -- httpBasicAuthConfig?: AuthConfigHttpBasicAuthConfig; -- /** Config for user oauth. */ -- oauthConfig?: AuthConfigOauthConfig; -- /** Config for user OIDC auth. */ -- oidcConfig?: AuthConfigOidcConfig; --} -- --/** Config for Google Service Account Authentication. */ --export declare interface AuthConfigGoogleServiceAccountConfig { -- /** Optional. The service account that the extension execution service runs as. - If the service account is specified, the `iam.serviceAccounts.getAccessToken` permission should be granted to Vertex AI Extension Service Agent (https://cloud.google.com/vertex-ai/docs/general/access-control#service-agents) on the specified service account. - If not specified, the Vertex AI Extension Service Agent will be used to execute the Extension. */ -- serviceAccount?: string; --} -- --/** Config for HTTP Basic Authentication. */ --export declare interface AuthConfigHttpBasicAuthConfig { -- /** Required. The name of the SecretManager secret version resource storing the base64 encoded credentials. Format: `projects/{project}/secrets/{secrete}/versions/{version}` - If specified, the `secretmanager.versions.access` permission should be granted to Vertex AI Extension Service Agent (https://cloud.google.com/vertex-ai/docs/general/access-control#service-agents) on the specified resource. */ -- credentialSecret?: string; --} -- --/** Config for user oauth. */ --export declare interface AuthConfigOauthConfig { -- /** Access token for extension endpoint. Only used to propagate token from [[ExecuteExtensionRequest.runtime_auth_config]] at request time. */ -- accessToken?: string; -- /** The service account used to generate access tokens for executing the Extension. - If the service account is specified, the `iam.serviceAccounts.getAccessToken` permission should be granted to Vertex AI Extension Service Agent (https://cloud.google.com/vertex-ai/docs/general/access-control#service-agents) on the provided service account. */ -- serviceAccount?: string; --} -- --/** Config for user OIDC auth. */ --export declare interface AuthConfigOidcConfig { -- /** OpenID Connect formatted ID token for extension endpoint. Only used to propagate token from [[ExecuteExtensionRequest.runtime_auth_config]] at request time. */ -- idToken?: string; -- /** The service account used to generate an OpenID Connect (OIDC)-compatible JWT token signed by the Google OIDC Provider (accounts.google.com) for extension endpoint (https://cloud.google.com/iam/docs/create-short-lived-credentials-direct#sa-credentials-oidc). - The audience for the token will be set to the URL in the server url defined in the OpenApi spec. - If the service account is provided, the service account should grant `iam.serviceAccounts.getOpenIdToken` permission to Vertex AI Extension Service Agent (https://cloud.google.com/vertex-ai/docs/general/access-control#service-agents). */ -- serviceAccount?: string; --} -- --/** Type of auth scheme. */ --export declare enum AuthType { -- AUTH_TYPE_UNSPECIFIED = "AUTH_TYPE_UNSPECIFIED", -- /** -- * No Auth. -- */ -- NO_AUTH = "NO_AUTH", -- /** -- * API Key Auth. -- */ -- API_KEY_AUTH = "API_KEY_AUTH", -- /** -- * HTTP Basic Auth. -- */ -- HTTP_BASIC_AUTH = "HTTP_BASIC_AUTH", -- /** -- * Google Service Account Auth. -- */ -- GOOGLE_SERVICE_ACCOUNT_AUTH = "GOOGLE_SERVICE_ACCOUNT_AUTH", -- /** -- * OAuth auth. -- */ -- OAUTH = "OAUTH", -- /** -- * OpenID Connect (OIDC) Auth. -- */ -- OIDC_AUTH = "OIDC_AUTH" --} -- --/** Configures automatic detection of activity. */ --export declare interface AutomaticActivityDetection { -- /** If enabled, detected voice and text input count as activity. If disabled, the client must send activity signals. */ -- disabled?: boolean; -- /** Determines how likely speech is to be detected. */ -- startOfSpeechSensitivity?: StartSensitivity; -- /** Determines how likely detected speech is ended. */ -- endOfSpeechSensitivity?: EndSensitivity; -- /** The required duration of detected speech before start-of-speech is committed. The lower this value the more sensitive the start-of-speech detection is and the shorter speech can be recognized. However, this also increases the probability of false positives. */ -- prefixPaddingMs?: number; -- /** The required duration of detected non-speech (e.g. silence) before end-of-speech is committed. The larger this value, the longer speech gaps can be without interrupting the user's activity but this will increase the model's latency. */ -- silenceDurationMs?: number; --} -- --/** The configuration for automatic function calling. */ --export declare interface AutomaticFunctionCallingConfig { -- /** Whether to disable automatic function calling. -- If not set or set to False, will enable automatic function calling. -- If set to True, will disable automatic function calling. -- */ -- disable?: boolean; -- /** If automatic function calling is enabled, -- maximum number of remote calls for automatic function calling. -- This number should be a positive integer. -- If not set, SDK will set maximum number of remote calls to 10. -- */ -- maximumRemoteCalls?: number; -- /** If automatic function calling is enabled, -- whether to ignore call history to the response. -- If not set, SDK will set ignore_call_history to false, -- and will append the call history to -- GenerateContentResponse.automatic_function_calling_history. -- */ -- ignoreCallHistory?: boolean; --} -- --/** -- * @license -- * Copyright 2025 Google LLC -- * SPDX-License-Identifier: Apache-2.0 -- */ --declare class BaseModule { --} -- --/** -- * Parameters for setting the base URLs for the Gemini API and Vertex AI API. -- */ --export declare interface BaseUrlParameters { -- geminiUrl?: string; -- vertexUrl?: string; --} -- --/** Defines the function behavior. Defaults to `BLOCKING`. */ --export declare enum Behavior { -- /** -- * This value is unused. -- */ -- UNSPECIFIED = "UNSPECIFIED", -- /** -- * If set, the system will wait to receive the function response before continuing the conversation. -- */ -- BLOCKING = "BLOCKING", -- /** -- * If set, the system will not wait to receive the function response. Instead, it will attempt to handle function responses as they become available while maintaining the conversation between the user and the model. -- */ -- NON_BLOCKING = "NON_BLOCKING" --} -- --/** Content blob. */ --declare interface Blob_2 { -- /** Optional. Display name of the blob. Used to provide a label or filename to distinguish blobs. This field is not currently used in the Gemini GenerateContent calls. */ -- displayName?: string; -- /** Required. Raw bytes. */ -- data?: string; -- /** Required. The IANA standard MIME type of the source data. */ -- mimeType?: string; --} --export { Blob_2 as Blob } -- --export declare type BlobImageUnion = Blob_2; -- --/** Output only. Blocked reason. */ --export declare enum BlockedReason { -- /** -- * Unspecified blocked reason. -- */ -- BLOCKED_REASON_UNSPECIFIED = "BLOCKED_REASON_UNSPECIFIED", -- /** -- * Candidates blocked due to safety. -- */ -- SAFETY = "SAFETY", -- /** -- * Candidates blocked due to other reason. -- */ -- OTHER = "OTHER", -- /** -- * Candidates blocked due to the terms which are included from the terminology blocklist. -- */ -- BLOCKLIST = "BLOCKLIST", -- /** -- * Candidates blocked due to prohibited content. -- */ -- PROHIBITED_CONTENT = "PROHIBITED_CONTENT" --} -- --/** A resource used in LLM queries for users to explicitly specify what to cache. */ --export declare interface CachedContent { -- /** The server-generated resource name of the cached content. */ -- name?: string; -- /** The user-generated meaningful display name of the cached content. */ -- displayName?: string; -- /** The name of the publisher model to use for cached content. */ -- model?: string; -- /** Creation time of the cache entry. */ -- createTime?: string; -- /** When the cache entry was last updated in UTC time. */ -- updateTime?: string; -- /** Expiration time of the cached content. */ -- expireTime?: string; -- /** Metadata on the usage of the cached content. */ -- usageMetadata?: CachedContentUsageMetadata; --} -- --/** Metadata on the usage of the cached content. */ --export declare interface CachedContentUsageMetadata { -- /** Duration of audio in seconds. */ -- audioDurationSeconds?: number; -- /** Number of images. */ -- imageCount?: number; -- /** Number of text characters. */ -- textCount?: number; -- /** Total number of tokens that the cached content consumes. */ -- totalTokenCount?: number; -- /** Duration of video in seconds. */ -- videoDurationSeconds?: number; --} -- --export declare class Caches extends BaseModule { -- private readonly apiClient; -- constructor(apiClient: ApiClient); -- /** -- * Lists cached content configurations. -- * -- * @param params - The parameters for the list request. -- * @return The paginated results of the list of cached contents. -- * -- * @example -- * ```ts -- * const cachedContents = await ai.caches.list({config: {'pageSize': 2}}); -- * for (const cachedContent of cachedContents) { -- * console.log(cachedContent); -- * } -- * ``` -- */ -- list: (params?: types.ListCachedContentsParameters) => Promise>; -- /** -- * Creates a cached contents resource. -- * -- * @remarks -- * Context caching is only supported for specific models. See [Gemini -- * Developer API reference](https://ai.google.dev/gemini-api/docs/caching?lang=node/context-cac) -- * and [Vertex AI reference](https://cloud.google.com/vertex-ai/generative-ai/docs/context-cache/context-cache-overview#supported_models) -- * for more information. -- * -- * @param params - The parameters for the create request. -- * @return The created cached content. -- * -- * @example -- * ```ts -- * const contents = ...; // Initialize the content to cache. -- * const response = await ai.caches.create({ -- * model: 'gemini-2.0-flash-001', -- * config: { -- * 'contents': contents, -- * 'displayName': 'test cache', -- * 'systemInstruction': 'What is the sum of the two pdfs?', -- * 'ttl': '86400s', -- * } -- * }); -- * ``` -- */ -- create(params: types.CreateCachedContentParameters): Promise; -- /** -- * Gets cached content configurations. -- * -- * @param params - The parameters for the get request. -- * @return The cached content. -- * -- * @example -- * ```ts -- * await ai.caches.get({name: '...'}); // The server-generated resource name. -- * ``` -- */ -- get(params: types.GetCachedContentParameters): Promise; -- /** -- * Deletes cached content. -- * -- * @param params - The parameters for the delete request. -- * @return The empty response returned by the API. -- * -- * @example -- * ```ts -- * await ai.caches.delete({name: '...'}); // The server-generated resource name. -- * ``` -- */ -- delete(params: types.DeleteCachedContentParameters): Promise; -- /** -- * Updates cached content configurations. -- * -- * @param params - The parameters for the update request. -- * @return The updated cached content. -- * -- * @example -- * ```ts -- * const response = await ai.caches.update({ -- * name: '...', // The server-generated resource name. -- * config: {'ttl': '7600s'} -- * }); -- * ``` -- */ -- update(params: types.UpdateCachedContentParameters): Promise; -- private listInternal; --} -- --/** -- * CallableTool is an invokable tool that can be executed with external -- * application (e.g., via Model Context Protocol) or local functions with -- * function calling. -- */ --export declare interface CallableTool { -- /** -- * Returns tool that can be called by Gemini. -- */ -- tool(): Promise; -- /** -- * Executes the callable tool with the given function call arguments and -- * returns the response parts from the tool execution. -- */ -- callTool(functionCalls: FunctionCall[]): Promise; --} -- --/** -- * CallableToolConfig is the configuration for a callable tool. -- */ --export declare interface CallableToolConfig { -- /** -- * Specifies the model's behavior after invoking this tool. -- */ -- behavior?: Behavior; --} -- --/** A response candidate generated from the model. */ --export declare interface Candidate { -- /** Contains the multi-part content of the response. -- */ -- content?: Content; -- /** Source attribution of the generated content. -- */ -- citationMetadata?: CitationMetadata; -- /** Describes the reason the model stopped generating tokens. -- */ -- finishMessage?: string; -- /** Number of tokens for this candidate. -- */ -- tokenCount?: number; -- /** The reason why the model stopped generating tokens. -- If empty, the model has not stopped generating the tokens. -- */ -- finishReason?: FinishReason; -- /** Metadata related to url context retrieval tool. */ -- urlContextMetadata?: UrlContextMetadata; -- /** Output only. Average log probability score of the candidate. */ -- avgLogprobs?: number; -- /** Output only. Metadata specifies sources used to ground generated content. */ -- groundingMetadata?: GroundingMetadata; -- /** Output only. Index of the candidate. */ -- index?: number; -- /** Output only. Log-likelihood scores for the response tokens and top tokens */ -- logprobsResult?: LogprobsResult; -- /** Output only. List of ratings for the safety of a response candidate. There is at most one rating per category. */ -- safetyRatings?: SafetyRating[]; --} -- --/** -- * Chat session that enables sending messages to the model with previous -- * conversation context. -- * -- * @remarks -- * The session maintains all the turns between user and model. -- */ --export declare class Chat { -- private readonly apiClient; -- private readonly modelsModule; -- private readonly model; -- private readonly config; -- private history; -- private sendPromise; -- constructor(apiClient: ApiClient, modelsModule: Models, model: string, config?: types.GenerateContentConfig, history?: types.Content[]); -- /** -- * Sends a message to the model and returns the response. -- * -- * @remarks -- * This method will wait for the previous message to be processed before -- * sending the next message. -- * -- * @see {@link Chat#sendMessageStream} for streaming method. -- * @param params - parameters for sending messages within a chat session. -- * @returns The model's response. -- * -- * @example -- * ```ts -- * const chat = ai.chats.create({model: 'gemini-2.0-flash'}); -- * const response = await chat.sendMessage({ -- * message: 'Why is the sky blue?' -- * }); -- * console.log(response.text); -- * ``` -- */ -- sendMessage(params: types.SendMessageParameters): Promise; -- /** -- * Sends a message to the model and returns the response in chunks. -- * -- * @remarks -- * This method will wait for the previous message to be processed before -- * sending the next message. -- * -- * @see {@link Chat#sendMessage} for non-streaming method. -- * @param params - parameters for sending the message. -- * @return The model's response. -- * -- * @example -- * ```ts -- * const chat = ai.chats.create({model: 'gemini-2.0-flash'}); -- * const response = await chat.sendMessageStream({ -- * message: 'Why is the sky blue?' -- * }); -- * for await (const chunk of response) { -- * console.log(chunk.text); -- * } -- * ``` -- */ -- sendMessageStream(params: types.SendMessageParameters): Promise>; -- /** -- * Returns the chat history. -- * -- * @remarks -- * The history is a list of contents alternating between user and model. -- * -- * There are two types of history: -- * - The `curated history` contains only the valid turns between user and -- * model, which will be included in the subsequent requests sent to the model. -- * - The `comprehensive history` contains all turns, including invalid or -- * empty model outputs, providing a complete record of the history. -- * -- * The history is updated after receiving the response from the model, -- * for streaming response, it means receiving the last chunk of the response. -- * -- * The `comprehensive history` is returned by default. To get the `curated -- * history`, set the `curated` parameter to `true`. -- * -- * @param curated - whether to return the curated history or the comprehensive -- * history. -- * @return History contents alternating between user and model for the entire -- * chat session. -- */ -- getHistory(curated?: boolean): types.Content[]; -- private processStreamResponse; -- private recordHistory; --} -- --/** -- * A utility class to create a chat session. -- */ --export declare class Chats { -- private readonly modelsModule; -- private readonly apiClient; -- constructor(modelsModule: Models, apiClient: ApiClient); -- /** -- * Creates a new chat session. -- * -- * @remarks -- * The config in the params will be used for all requests within the chat -- * session unless overridden by a per-request `config` in -- * @see {@link types.SendMessageParameters#config}. -- * -- * @param params - Parameters for creating a chat session. -- * @returns A new chat session. -- * -- * @example -- * ```ts -- * const chat = ai.chats.create({ -- * model: 'gemini-2.0-flash' -- * config: { -- * temperature: 0.5, -- * maxOutputTokens: 1024, -- * } -- * }); -- * ``` -- */ -- create(params: types.CreateChatParameters): Chat; --} -- --/** Describes the machine learning model version checkpoint. */ --export declare interface Checkpoint { -- /** The ID of the checkpoint. -- */ -- checkpointId?: string; -- /** The epoch of the checkpoint. -- */ -- epoch?: string; -- /** The step of the checkpoint. -- */ -- step?: string; --} -- --/** Source attributions for content. */ --export declare interface Citation { -- /** Output only. End index into the content. */ -- endIndex?: number; -- /** Output only. License of the attribution. */ -- license?: string; -- /** Output only. Publication date of the attribution. */ -- publicationDate?: GoogleTypeDate; -- /** Output only. Start index into the content. */ -- startIndex?: number; -- /** Output only. Title of the attribution. */ -- title?: string; -- /** Output only. Url reference of the attribution. */ -- uri?: string; --} -- --/** Citation information when the model quotes another source. */ --export declare interface CitationMetadata { -- /** Contains citation information when the model directly quotes, at -- length, from another source. Can include traditional websites and code -- repositories. -- */ -- citations?: Citation[]; --} -- --/** Result of executing the [ExecutableCode]. Always follows a `part` containing the [ExecutableCode]. */ --export declare interface CodeExecutionResult { -- /** Required. Outcome of the code execution. */ -- outcome?: Outcome; -- /** Optional. Contains stdout when code execution is successful, stderr or other description otherwise. */ -- output?: string; --} -- --/** Optional parameters for computing tokens. */ --export declare interface ComputeTokensConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; --} -- --/** Parameters for computing tokens. */ --export declare interface ComputeTokensParameters { -- /** ID of the model to use. For a list of models, see `Google models -- `_. */ -- model: string; -- /** Input content. */ -- contents: ContentListUnion; -- /** Optional parameters for the request. -- */ -- config?: ComputeTokensConfig; --} -- --/** Response for computing tokens. */ --export declare class ComputeTokensResponse { -- /** Lists of tokens info from the input. A ComputeTokensRequest could have multiple instances with a prompt in each instance. We also need to return lists of tokens info for the request with multiple instances. */ -- tokensInfo?: TokensInfo[]; --} -- --/** Contains the multi-part content of a message. */ --export declare interface Content { -- /** List of parts that constitute a single message. Each part may have -- a different IANA MIME type. */ -- parts?: Part[]; -- /** Optional. The producer of the content. Must be either 'user' or -- 'model'. Useful to set for multi-turn conversations, otherwise can be -- empty. If role is not specified, SDK will determine the role. */ -- role?: string; --} -- --/** The embedding generated from an input content. */ --export declare interface ContentEmbedding { -- /** A list of floats representing an embedding. -- */ -- values?: number[]; -- /** Vertex API only. Statistics of the input text associated with this -- embedding. -- */ -- statistics?: ContentEmbeddingStatistics; --} -- --/** Statistics of the input text associated with the result of content embedding. */ --export declare interface ContentEmbeddingStatistics { -- /** Vertex API only. If the input text was truncated due to having -- a length longer than the allowed maximum input. -- */ -- truncated?: boolean; -- /** Vertex API only. Number of tokens of the input text. -- */ -- tokenCount?: number; --} -- --export declare type ContentListUnion = Content | Content[] | PartUnion | PartUnion[]; -- --export declare type ContentUnion = Content | PartUnion[] | PartUnion; -- --/** Enables context window compression -- mechanism managing model context window so it does not exceed given length. */ --export declare interface ContextWindowCompressionConfig { -- /** Number of tokens (before running turn) that triggers context window compression mechanism. */ -- triggerTokens?: string; -- /** Sliding window compression mechanism. */ -- slidingWindow?: SlidingWindow; --} -- --/** Configuration for a Control reference image. */ --export declare interface ControlReferenceConfig { -- /** The type of control reference image to use. */ -- controlType?: ControlReferenceType; -- /** Defaults to False. When set to True, the control image will be -- computed by the model based on the control type. When set to False, -- the control image must be provided by the user. */ -- enableControlImageComputation?: boolean; --} -- --/** A control reference image. -- -- The image of the control reference image is either a control image provided -- by the user, or a regular image which the backend will use to generate a -- control image of. In the case of the latter, the -- enable_control_image_computation field in the config should be set to True. -- -- A control image is an image that represents a sketch image of areas for the -- model to fill in based on the prompt. -- */ --export declare class ControlReferenceImage { -- /** The reference image for the editing operation. */ -- referenceImage?: Image_2; -- /** The id of the reference image. */ -- referenceId?: number; -- /** The type of the reference image. Only set by the SDK. */ -- referenceType?: string; -- /** Configuration for the control reference image. */ -- config?: ControlReferenceConfig; -- /** Internal method to convert to ReferenceImageAPIInternal. */ -- toReferenceImageAPI(): any; --} -- --/** Enum representing the control type of a control reference image. */ --export declare enum ControlReferenceType { -- CONTROL_TYPE_DEFAULT = "CONTROL_TYPE_DEFAULT", -- CONTROL_TYPE_CANNY = "CONTROL_TYPE_CANNY", -- CONTROL_TYPE_SCRIBBLE = "CONTROL_TYPE_SCRIBBLE", -- CONTROL_TYPE_FACE_MESH = "CONTROL_TYPE_FACE_MESH" --} -- --/** Config for the count_tokens method. */ --export declare interface CountTokensConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; -- /** Instructions for the model to steer it toward better performance. -- */ -- systemInstruction?: ContentUnion; -- /** Code that enables the system to interact with external systems to -- perform an action outside of the knowledge and scope of the model. -- */ -- tools?: Tool[]; -- /** Configuration that the model uses to generate the response. Not -- supported by the Gemini Developer API. -- */ -- generationConfig?: GenerationConfig; --} -- --/** Parameters for counting tokens. */ --export declare interface CountTokensParameters { -- /** ID of the model to use. For a list of models, see `Google models -- `_. */ -- model: string; -- /** Input content. */ -- contents: ContentListUnion; -- /** Configuration for counting tokens. */ -- config?: CountTokensConfig; --} -- --/** Response for counting tokens. */ --export declare class CountTokensResponse { -- /** Total number of tokens. */ -- totalTokens?: number; -- /** Number of tokens in the cached part of the prompt (the cached content). */ -- cachedContentTokenCount?: number; --} -- --/** Optional parameters. */ --export declare interface CreateAuthTokenConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; -- /** An optional time after which, when using the resulting token, -- messages in Live API sessions will be rejected. (Gemini may -- preemptively close the session after this time.) -- -- If not set then this defaults to 30 minutes in the future. If set, this -- value must be less than 20 hours in the future. */ -- expireTime?: string; -- /** The time after which new Live API sessions using the token -- resulting from this request will be rejected. -- -- If not set this defaults to 60 seconds in the future. If set, this value -- must be less than 20 hours in the future. */ -- newSessionExpireTime?: string; -- /** The number of times the token can be used. If this value is zero -- then no limit is applied. Default is 1. Resuming a Live API session does -- not count as a use. */ -- uses?: number; -- /** Configuration specific to Live API connections created using this token. */ -- liveEphemeralParameters?: LiveEphemeralParameters; -- /** Additional fields to lock in the effective LiveConnectParameters. */ -- lockAdditionalFields?: string[]; --} -- --/** Optional configuration for cached content creation. */ --export declare interface CreateCachedContentConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; -- /** The TTL for this resource. The expiration time is computed: now + TTL. It is a duration string, with up to nine fractional digits, terminated by 's'. Example: "3.5s". */ -- ttl?: string; -- /** Timestamp of when this resource is considered expired. Uses RFC 3339 format, Example: 2014-10-02T15:01:23Z. */ -- expireTime?: string; -- /** The user-generated meaningful display name of the cached content. -- */ -- displayName?: string; -- /** The content to cache. -- */ -- contents?: ContentListUnion; -- /** Developer set system instruction. -- */ -- systemInstruction?: ContentUnion; -- /** A list of `Tools` the model may use to generate the next response. -- */ -- tools?: Tool[]; -- /** Configuration for the tools to use. This config is shared for all tools. -- */ -- toolConfig?: ToolConfig; -- /** The Cloud KMS resource identifier of the customer managed -- encryption key used to protect a resource. -- The key needs to be in the same region as where the compute resource is -- created. See -- https://cloud.google.com/vertex-ai/docs/general/cmek for more -- details. If this is set, then all created CachedContent objects -- will be encrypted with the provided encryption key. -- Allowed formats: projects/{project}/locations/{location}/keyRings/{key_ring}/cryptoKeys/{crypto_key} -- */ -- kmsKeyName?: string; --} -- --/** Parameters for caches.create method. */ --export declare interface CreateCachedContentParameters { -- /** ID of the model to use. Example: gemini-2.0-flash */ -- model: string; -- /** Configuration that contains optional parameters. -- */ -- config?: CreateCachedContentConfig; --} -- --/** Parameters for initializing a new chat session. -- -- These parameters are used when creating a chat session with the -- `chats.create()` method. -- */ --export declare interface CreateChatParameters { -- /** The name of the model to use for the chat session. -- -- For example: 'gemini-2.0-flash', 'gemini-2.0-flash-lite', etc. See Gemini API -- docs to find the available models. -- */ -- model: string; -- /** Config for the entire chat session. -- -- This config applies to all requests within the session -- unless overridden by a per-request `config` in `SendMessageParameters`. -- */ -- config?: GenerateContentConfig; -- /** The initial conversation history for the chat session. -- -- This allows you to start the chat with a pre-existing history. The history -- must be a list of `Content` alternating between 'user' and 'model' roles. -- It should start with a 'user' message. -- */ -- history?: Content[]; --} -- --/** Used to override the default configuration. */ --export declare interface CreateFileConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; --} -- --/** Generates the parameters for the private _create method. */ --export declare interface CreateFileParameters { -- /** The file to be uploaded. -- mime_type: (Required) The MIME type of the file. Must be provided. -- name: (Optional) The name of the file in the destination (e.g. -- 'files/sample-image'). -- display_name: (Optional) The display name of the file. -- */ -- file: File_2; -- /** Used to override the default configuration. */ -- config?: CreateFileConfig; --} -- --/** Response for the create file method. */ --export declare class CreateFileResponse { -- /** Used to retain the full HTTP response. */ -- sdkHttpResponse?: HttpResponse; --} -- --/** -- * Creates a `Content` object with a model role from a `PartListUnion` object or `string`. -- */ --export declare function createModelContent(partOrString: PartListUnion | string): Content; -- --/** -- * Creates a `Part` object from a `base64` encoded `string`. -- */ --export declare function createPartFromBase64(data: string, mimeType: string): Part; -- --/** -- * Creates a `Part` object from the `outcome` and `output` of a `CodeExecutionResult` object. -- */ --export declare function createPartFromCodeExecutionResult(outcome: Outcome, output: string): Part; -- --/** -- * Creates a `Part` object from the `code` and `language` of an `ExecutableCode` object. -- */ --export declare function createPartFromExecutableCode(code: string, language: Language): Part; -- --/** -- * Creates a `Part` object from a `FunctionCall` object. -- */ --export declare function createPartFromFunctionCall(name: string, args: Record): Part; -- --/** -- * Creates a `Part` object from a `FunctionResponse` object. -- */ --export declare function createPartFromFunctionResponse(id: string, name: string, response: Record): Part; -- --/** -- * Creates a `Part` object from a `text` string. -- */ --export declare function createPartFromText(text: string): Part; -- --/** -- * Creates a `Part` object from a `URI` string. -- */ --export declare function createPartFromUri(uri: string, mimeType: string): Part; -- --/** Supervised fine-tuning job creation request - optional fields. */ --export declare interface CreateTuningJobConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; -- /** Cloud Storage path to file containing training dataset for tuning. The dataset must be formatted as a JSONL file. */ -- validationDataset?: TuningValidationDataset; -- /** The display name of the tuned Model. The name can be up to 128 characters long and can consist of any UTF-8 characters. */ -- tunedModelDisplayName?: string; -- /** The description of the TuningJob */ -- description?: string; -- /** Number of complete passes the model makes over the entire training dataset during training. */ -- epochCount?: number; -- /** Multiplier for adjusting the default learning rate. */ -- learningRateMultiplier?: number; -- /** If set to true, disable intermediate checkpoints for SFT and only the last checkpoint will be exported. Otherwise, enable intermediate checkpoints for SFT. */ -- exportLastCheckpointOnly?: boolean; -- /** Adapter size for tuning. */ -- adapterSize?: AdapterSize; -- /** The batch size hyperparameter for tuning. If not set, a default of 4 or 16 will be used based on the number of training examples. */ -- batchSize?: number; -- /** The learning rate hyperparameter for tuning. If not set, a default of 0.001 or 0.0002 will be calculated based on the number of training examples. */ -- learningRate?: number; --} -- --/** Supervised fine-tuning job creation parameters - optional fields. */ --export declare interface CreateTuningJobParameters { -- /** The base model that is being tuned, e.g., "gemini-1.0-pro-002". */ -- baseModel: string; -- /** Cloud Storage path to file containing training dataset for tuning. The dataset must be formatted as a JSONL file. */ -- trainingDataset: TuningDataset; -- /** Configuration for the tuning job. */ -- config?: CreateTuningJobConfig; --} -- --/** -- * Creates a `Content` object with a user role from a `PartListUnion` object or `string`. -- */ --export declare function createUserContent(partOrString: PartListUnion | string): Content; -- --/** Distribution computed over a tuning dataset. */ --export declare interface DatasetDistribution { -- /** Output only. Defines the histogram bucket. */ -- buckets?: DatasetDistributionDistributionBucket[]; -- /** Output only. The maximum of the population values. */ -- max?: number; -- /** Output only. The arithmetic mean of the values in the population. */ -- mean?: number; -- /** Output only. The median of the values in the population. */ -- median?: number; -- /** Output only. The minimum of the population values. */ -- min?: number; -- /** Output only. The 5th percentile of the values in the population. */ -- p5?: number; -- /** Output only. The 95th percentile of the values in the population. */ -- p95?: number; -- /** Output only. Sum of a given population of values. */ -- sum?: number; --} -- --/** Dataset bucket used to create a histogram for the distribution given a population of values. */ --export declare interface DatasetDistributionDistributionBucket { -- /** Output only. Number of values in the bucket. */ -- count?: string; -- /** Output only. Left bound of the bucket. */ -- left?: number; -- /** Output only. Right bound of the bucket. */ -- right?: number; --} -- --/** Statistics computed over a tuning dataset. */ --export declare interface DatasetStats { -- /** Output only. Number of billable characters in the tuning dataset. */ -- totalBillableCharacterCount?: string; -- /** Output only. Number of tuning characters in the tuning dataset. */ -- totalTuningCharacterCount?: string; -- /** Output only. Number of examples in the tuning dataset. */ -- tuningDatasetExampleCount?: string; -- /** Output only. Number of tuning steps for this Tuning Job. */ -- tuningStepCount?: string; -- /** Output only. Sample user messages in the training dataset uri. */ -- userDatasetExamples?: Content[]; -- /** Output only. Dataset distributions for the user input tokens. */ -- userInputTokenDistribution?: DatasetDistribution; -- /** Output only. Dataset distributions for the messages per example. */ -- userMessagePerExampleDistribution?: DatasetDistribution; -- /** Output only. Dataset distributions for the user output tokens. */ -- userOutputTokenDistribution?: DatasetDistribution; --} -- --/** Optional parameters for caches.delete method. */ --export declare interface DeleteCachedContentConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; --} -- --/** Parameters for caches.delete method. */ --export declare interface DeleteCachedContentParameters { -- /** The server-generated resource name of the cached content. -- */ -- name: string; -- /** Optional parameters for the request. -- */ -- config?: DeleteCachedContentConfig; --} -- --/** Empty response for caches.delete method. */ --export declare class DeleteCachedContentResponse { --} -- --/** Used to override the default configuration. */ --export declare interface DeleteFileConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; --} -- --/** Generates the parameters for the get method. */ --export declare interface DeleteFileParameters { -- /** The name identifier for the file to be deleted. */ -- name: string; -- /** Used to override the default configuration. */ -- config?: DeleteFileConfig; --} -- --/** Response for the delete file method. */ --export declare class DeleteFileResponse { --} -- --/** Configuration for deleting a tuned model. */ --export declare interface DeleteModelConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; --} -- --/** Parameters for deleting a tuned model. */ --export declare interface DeleteModelParameters { -- model: string; -- /** Optional parameters for the request. */ -- config?: DeleteModelConfig; --} -- --export declare class DeleteModelResponse { --} -- --/** Statistics computed for datasets used for distillation. */ --export declare interface DistillationDataStats { -- /** Output only. Statistics computed for the training dataset. */ -- trainingDatasetStats?: DatasetStats; --} -- --/** Hyperparameters for Distillation. */ --export declare interface DistillationHyperParameters { -- /** Optional. Adapter size for distillation. */ -- adapterSize?: AdapterSize; -- /** Optional. Number of complete passes the model makes over the entire training dataset during training. */ -- epochCount?: string; -- /** Optional. Multiplier for adjusting the default learning rate. */ -- learningRateMultiplier?: number; --} -- --/** Tuning Spec for Distillation. */ --export declare interface DistillationSpec { -- /** The base teacher model that is being distilled, e.g., "gemini-1.0-pro-002". */ -- baseTeacherModel?: string; -- /** Optional. Hyperparameters for Distillation. */ -- hyperParameters?: DistillationHyperParameters; -- /** Required. A path in a Cloud Storage bucket, which will be treated as the root output directory of the distillation pipeline. It is used by the system to generate the paths of output artifacts. */ -- pipelineRootDirectory?: string; -- /** The student model that is being tuned, e.g., "google/gemma-2b-1.1-it". */ -- studentModel?: string; -- /** Required. Cloud Storage path to file containing training dataset for tuning. The dataset must be formatted as a JSONL file. */ -- trainingDatasetUri?: string; -- /** The resource name of the Tuned teacher model. Format: `projects/{project}/locations/{location}/models/{model}`. */ -- tunedTeacherModelSource?: string; -- /** Optional. Cloud Storage path to file containing validation dataset for tuning. The dataset must be formatted as a JSONL file. */ -- validationDatasetUri?: string; --} -- --export declare type DownloadableFileUnion = string | File_2 | GeneratedVideo | Video; -- --declare interface Downloader { -- /** -- * Downloads a file to the given location. -- * -- * @param params The parameters for downloading the file. -- * @param apiClient The ApiClient to use for uploading. -- * @return A Promises that resolves when the download is complete. -- */ -- download(params: DownloadFileParameters, apiClient: ApiClient): Promise; --} -- --/** Used to override the default configuration. */ --export declare interface DownloadFileConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; --} -- --/** Parameters used to download a file. */ --export declare interface DownloadFileParameters { -- /** The file to download. It can be a file name, a file object or a generated video. */ -- file: DownloadableFileUnion; -- /** Location where the file should be downloaded to. */ -- downloadPath: string; -- /** Configuration to for the download operation. */ -- config?: DownloadFileConfig; --} -- --/** Describes the options to customize dynamic retrieval. */ --export declare interface DynamicRetrievalConfig { -- /** The mode of the predictor to be used in dynamic retrieval. */ -- mode?: DynamicRetrievalConfigMode; -- /** Optional. The threshold to be used in dynamic retrieval. If not set, a system default value is used. */ -- dynamicThreshold?: number; --} -- --/** Config for the dynamic retrieval config mode. */ --export declare enum DynamicRetrievalConfigMode { -- /** -- * Always trigger retrieval. -- */ -- MODE_UNSPECIFIED = "MODE_UNSPECIFIED", -- /** -- * Run retrieval only when system decides it is necessary. -- */ -- MODE_DYNAMIC = "MODE_DYNAMIC" --} -- --/** Configuration for editing an image. */ --export declare interface EditImageConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; -- /** Cloud Storage URI used to store the generated images. -- */ -- outputGcsUri?: string; -- /** Description of what to discourage in the generated images. -- */ -- negativePrompt?: string; -- /** Number of images to generate. -- */ -- numberOfImages?: number; -- /** Aspect ratio of the generated images. -- */ -- aspectRatio?: string; -- /** Controls how much the model adheres to the text prompt. Large -- values increase output and prompt alignment, but may compromise image -- quality. -- */ -- guidanceScale?: number; -- /** Random seed for image generation. This is not available when -- ``add_watermark`` is set to true. -- */ -- seed?: number; -- /** Filter level for safety filtering. -- */ -- safetyFilterLevel?: SafetyFilterLevel; -- /** Allows generation of people by the model. -- */ -- personGeneration?: PersonGeneration; -- /** Whether to report the safety scores of each generated image and -- the positive prompt in the response. -- */ -- includeSafetyAttributes?: boolean; -- /** Whether to include the Responsible AI filter reason if the image -- is filtered out of the response. -- */ -- includeRaiReason?: boolean; -- /** Language of the text in the prompt. -- */ -- language?: ImagePromptLanguage; -- /** MIME type of the generated image. -- */ -- outputMimeType?: string; -- /** Compression quality of the generated image (for ``image/jpeg`` -- only). -- */ -- outputCompressionQuality?: number; -- /** Describes the editing mode for the request. */ -- editMode?: EditMode; -- /** The number of sampling steps. A higher value has better image -- quality, while a lower value has better latency. */ -- baseSteps?: number; --} -- --/** Parameters for the request to edit an image. */ --export declare interface EditImageParameters { -- /** The model to use. */ -- model: string; -- /** A text description of the edit to apply to the image. */ -- prompt: string; -- /** The reference images for Imagen 3 editing. */ -- referenceImages: ReferenceImage[]; -- /** Configuration for editing. */ -- config?: EditImageConfig; --} -- --/** Response for the request to edit an image. */ --export declare class EditImageResponse { -- /** Generated images. */ -- generatedImages?: GeneratedImage[]; --} -- --/** Enum representing the Imagen 3 Edit mode. */ --export declare enum EditMode { -- EDIT_MODE_DEFAULT = "EDIT_MODE_DEFAULT", -- EDIT_MODE_INPAINT_REMOVAL = "EDIT_MODE_INPAINT_REMOVAL", -- EDIT_MODE_INPAINT_INSERTION = "EDIT_MODE_INPAINT_INSERTION", -- EDIT_MODE_OUTPAINT = "EDIT_MODE_OUTPAINT", -- EDIT_MODE_CONTROLLED_EDITING = "EDIT_MODE_CONTROLLED_EDITING", -- EDIT_MODE_STYLE = "EDIT_MODE_STYLE", -- EDIT_MODE_BGSWAP = "EDIT_MODE_BGSWAP", -- EDIT_MODE_PRODUCT_IMAGE = "EDIT_MODE_PRODUCT_IMAGE" --} -- --/** Optional parameters for the embed_content method. */ --export declare interface EmbedContentConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; -- /** Type of task for which the embedding will be used. -- */ -- taskType?: string; -- /** Title for the text. Only applicable when TaskType is -- `RETRIEVAL_DOCUMENT`. -- */ -- title?: string; -- /** Reduced dimension for the output embedding. If set, -- excessive values in the output embedding are truncated from the end. -- Supported by newer models since 2024 only. You cannot set this value if -- using the earlier model (`models/embedding-001`). -- */ -- outputDimensionality?: number; -- /** Vertex API only. The MIME type of the input. -- */ -- mimeType?: string; -- /** Vertex API only. Whether to silently truncate inputs longer than -- the max sequence length. If this option is set to false, oversized inputs -- will lead to an INVALID_ARGUMENT error, similar to other text APIs. -- */ -- autoTruncate?: boolean; --} -- --/** Request-level metadata for the Vertex Embed Content API. */ --export declare interface EmbedContentMetadata { -- /** Vertex API only. The total number of billable characters included -- in the request. -- */ -- billableCharacterCount?: number; --} -- --/** Parameters for the embed_content method. */ --export declare interface EmbedContentParameters { -- /** ID of the model to use. For a list of models, see `Google models -- `_. */ -- model: string; -- /** The content to embed. Only the `parts.text` fields will be counted. -- */ -- contents: ContentListUnion; -- /** Configuration that contains optional parameters. -- */ -- config?: EmbedContentConfig; --} -- --/** Response for the embed_content method. */ --export declare class EmbedContentResponse { -- /** The embeddings for each request, in the same order as provided in -- the batch request. -- */ -- embeddings?: ContentEmbedding[]; -- /** Vertex API only. Metadata about the request. -- */ -- metadata?: EmbedContentMetadata; --} -- --/** Represents a customer-managed encryption key spec that can be applied to a top-level resource. */ --export declare interface EncryptionSpec { -- /** Required. The Cloud KMS resource identifier of the customer managed encryption key used to protect a resource. Has the form: `projects/my-project/locations/my-region/keyRings/my-kr/cryptoKeys/my-key`. The key needs to be in the same region as where the compute resource is created. */ -- kmsKeyName?: string; --} -- --/** An endpoint where you deploy models. */ --export declare interface Endpoint { -- /** Resource name of the endpoint. */ -- name?: string; -- /** ID of the model that's deployed to the endpoint. */ -- deployedModelId?: string; --} -- --/** End of speech sensitivity. */ --export declare enum EndSensitivity { -- /** -- * The default is END_SENSITIVITY_LOW. -- */ -- END_SENSITIVITY_UNSPECIFIED = "END_SENSITIVITY_UNSPECIFIED", -- /** -- * Automatic detection ends speech more often. -- */ -- END_SENSITIVITY_HIGH = "END_SENSITIVITY_HIGH", -- /** -- * Automatic detection ends speech less often. -- */ -- END_SENSITIVITY_LOW = "END_SENSITIVITY_LOW" --} -- --/** Tool to search public web data, powered by Vertex AI Search and Sec4 compliance. */ --export declare interface EnterpriseWebSearch { --} -- --/** Code generated by the model that is meant to be executed, and the result returned to the model. Generated when using the [FunctionDeclaration] tool and [FunctionCallingConfig] mode is set to [Mode.CODE]. */ --export declare interface ExecutableCode { -- /** Required. The code to be executed. */ -- code?: string; -- /** Required. Programming language of the `code`. */ -- language?: Language; --} -- --/** Options for feature selection preference. */ --export declare enum FeatureSelectionPreference { -- FEATURE_SELECTION_PREFERENCE_UNSPECIFIED = "FEATURE_SELECTION_PREFERENCE_UNSPECIFIED", -- PRIORITIZE_QUALITY = "PRIORITIZE_QUALITY", -- BALANCED = "BALANCED", -- PRIORITIZE_COST = "PRIORITIZE_COST" --} -- --export declare interface FetchPredictOperationConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; --} -- --/** Parameters for the fetchPredictOperation method. */ --export declare interface FetchPredictOperationParameters { -- /** The server-assigned name for the operation. */ -- operationName: string; -- resourceName: string; -- /** Used to override the default configuration. */ -- config?: FetchPredictOperationConfig; --} -- --/** A file uploaded to the API. */ --declare interface File_2 { -- /** The `File` resource name. The ID (name excluding the "files/" prefix) can contain up to 40 characters that are lowercase alphanumeric or dashes (-). The ID cannot start or end with a dash. If the name is empty on create, a unique name will be generated. Example: `files/123-456` */ -- name?: string; -- /** Optional. The human-readable display name for the `File`. The display name must be no more than 512 characters in length, including spaces. Example: 'Welcome Image' */ -- displayName?: string; -- /** Output only. MIME type of the file. */ -- mimeType?: string; -- /** Output only. Size of the file in bytes. */ -- sizeBytes?: string; -- /** Output only. The timestamp of when the `File` was created. */ -- createTime?: string; -- /** Output only. The timestamp of when the `File` will be deleted. Only set if the `File` is scheduled to expire. */ -- expirationTime?: string; -- /** Output only. The timestamp of when the `File` was last updated. */ -- updateTime?: string; -- /** Output only. SHA-256 hash of the uploaded bytes. The hash value is encoded in base64 format. */ -- sha256Hash?: string; -- /** Output only. The URI of the `File`. */ -- uri?: string; -- /** Output only. The URI of the `File`, only set for downloadable (generated) files. */ -- downloadUri?: string; -- /** Output only. Processing state of the File. */ -- state?: FileState; -- /** Output only. The source of the `File`. */ -- source?: FileSource; -- /** Output only. Metadata for a video. */ -- videoMetadata?: Record; -- /** Output only. Error status if File processing failed. */ -- error?: FileStatus; --} --export { File_2 as File } -- --/** URI based data. */ --export declare interface FileData { -- /** Required. URI. */ -- fileUri?: string; -- /** Required. The IANA standard MIME type of the source data. */ -- mimeType?: string; --} -- --export declare class Files extends BaseModule { -- private readonly apiClient; -- constructor(apiClient: ApiClient); -- /** -- * Lists all current project files from the service. -- * -- * @param params - The parameters for the list request -- * @return The paginated results of the list of files -- * -- * @example -- * The following code prints the names of all files from the service, the -- * size of each page is 10. -- * -- * ```ts -- * const listResponse = await ai.files.list({config: {'pageSize': 10}}); -- * for await (const file of listResponse) { -- * console.log(file.name); -- * } -- * ``` -- */ -- list: (params?: types.ListFilesParameters) => Promise>; -- /** -- * Uploads a file asynchronously to the Gemini API. -- * This method is not available in Vertex AI. -- * Supported upload sources: -- * - Node.js: File path (string) or Blob object. -- * - Browser: Blob object (e.g., File). -- * -- * @remarks -- * The `mimeType` can be specified in the `config` parameter. If omitted: -- * - For file path (string) inputs, the `mimeType` will be inferred from the -- * file extension. -- * - For Blob object inputs, the `mimeType` will be set to the Blob's `type` -- * property. -- * Somex eamples for file extension to mimeType mapping: -- * .txt -> text/plain -- * .json -> application/json -- * .jpg -> image/jpeg -- * .png -> image/png -- * .mp3 -> audio/mpeg -- * .mp4 -> video/mp4 -- * -- * This section can contain multiple paragraphs and code examples. -- * -- * @param params - Optional parameters specified in the -- * `types.UploadFileParameters` interface. -- * @see {@link types.UploadFileParameters#config} for the optional -- * config in the parameters. -- * @return A promise that resolves to a `types.File` object. -- * @throws An error if called on a Vertex AI client. -- * @throws An error if the `mimeType` is not provided and can not be inferred, -- * the `mimeType` can be provided in the `params.config` parameter. -- * @throws An error occurs if a suitable upload location cannot be established. -- * -- * @example -- * The following code uploads a file to Gemini API. -- * -- * ```ts -- * const file = await ai.files.upload({file: 'file.txt', config: { -- * mimeType: 'text/plain', -- * }}); -- * console.log(file.name); -- * ``` -- */ -- upload(params: types.UploadFileParameters): Promise; -- /** -- * Downloads a remotely stored file asynchronously to a location specified in -- * the `params` object. This method only works on Node environment, to -- * download files in the browser, use a browser compliant method like an -- * tag. -- * -- * @param params - The parameters for the download request. -- * -- * @example -- * The following code downloads an example file named "files/mehozpxf877d" as -- * "file.txt". -- * -- * ```ts -- * await ai.files.download({file: file.name, downloadPath: 'file.txt'}); -- * ``` -- */ -- download(params: types.DownloadFileParameters): Promise; -- private listInternal; -- private createInternal; -- /** -- * Retrieves the file information from the service. -- * -- * @param params - The parameters for the get request -- * @return The Promise that resolves to the types.File object requested. -- * -- * @example -- * ```ts -- * const config: GetFileParameters = { -- * name: fileName, -- * }; -- * file = await ai.files.get(config); -- * console.log(file.name); -- * ``` -- */ -- get(params: types.GetFileParameters): Promise; -- /** -- * Deletes a remotely stored file. -- * -- * @param params - The parameters for the delete request. -- * @return The DeleteFileResponse, the response for the delete method. -- * -- * @example -- * The following code deletes an example file named "files/mehozpxf877d". -- * -- * ```ts -- * await ai.files.delete({name: file.name}); -- * ``` -- */ -- delete(params: types.DeleteFileParameters): Promise; --} -- --/** Source of the File. */ --export declare enum FileSource { -- SOURCE_UNSPECIFIED = "SOURCE_UNSPECIFIED", -- UPLOADED = "UPLOADED", -- GENERATED = "GENERATED" --} -- --/** -- * Represents the size and mimeType of a file. The information is used to -- * request the upload URL from the https://generativelanguage.googleapis.com/upload/v1beta/files endpoint. -- * This interface defines the structure for constructing and executing HTTP -- * requests. -- */ --declare interface FileStat { -- /** -- * The size of the file in bytes. -- */ -- size: number; -- /** -- * The MIME type of the file. -- */ -- type: string | undefined; --} -- --/** State for the lifecycle of a File. */ --export declare enum FileState { -- STATE_UNSPECIFIED = "STATE_UNSPECIFIED", -- PROCESSING = "PROCESSING", -- ACTIVE = "ACTIVE", -- FAILED = "FAILED" --} -- --/** Status of a File that uses a common error model. */ --export declare interface FileStatus { -- /** A list of messages that carry the error details. There is a common set of message types for APIs to use. */ -- details?: Record[]; -- /** A list of messages that carry the error details. There is a common set of message types for APIs to use. */ -- message?: string; -- /** The status code. 0 for OK, 1 for CANCELLED */ -- code?: number; --} -- --/** Output only. The reason why the model stopped generating tokens. -- -- If empty, the model has not stopped generating the tokens. -- */ --export declare enum FinishReason { -- /** -- * The finish reason is unspecified. -- */ -- FINISH_REASON_UNSPECIFIED = "FINISH_REASON_UNSPECIFIED", -- /** -- * Token generation reached a natural stopping point or a configured stop sequence. -- */ -- STOP = "STOP", -- /** -- * Token generation reached the configured maximum output tokens. -- */ -- MAX_TOKENS = "MAX_TOKENS", -- /** -- * Token generation stopped because the content potentially contains safety violations. NOTE: When streaming, [content][] is empty if content filters blocks the output. -- */ -- SAFETY = "SAFETY", -- /** -- * The token generation stopped because of potential recitation. -- */ -- RECITATION = "RECITATION", -- /** -- * The token generation stopped because of using an unsupported language. -- */ -- LANGUAGE = "LANGUAGE", -- /** -- * All other reasons that stopped the token generation. -- */ -- OTHER = "OTHER", -- /** -- * Token generation stopped because the content contains forbidden terms. -- */ -- BLOCKLIST = "BLOCKLIST", -- /** -- * Token generation stopped for potentially containing prohibited content. -- */ -- PROHIBITED_CONTENT = "PROHIBITED_CONTENT", -- /** -- * Token generation stopped because the content potentially contains Sensitive Personally Identifiable Information (SPII). -- */ -- SPII = "SPII", -- /** -- * The function call generated by the model is invalid. -- */ -- MALFORMED_FUNCTION_CALL = "MALFORMED_FUNCTION_CALL", -- /** -- * Token generation stopped because generated images have safety violations. -- */ -- IMAGE_SAFETY = "IMAGE_SAFETY" --} -- --/** A function call. */ --export declare interface FunctionCall { -- /** The unique id of the function call. If populated, the client to execute the -- `function_call` and return the response with the matching `id`. */ -- id?: string; -- /** Optional. Required. The function parameters and values in JSON object format. See [FunctionDeclaration.parameters] for parameter details. */ -- args?: Record; -- /** Required. The name of the function to call. Matches [FunctionDeclaration.name]. */ -- name?: string; --} -- --/** Function calling config. */ --export declare interface FunctionCallingConfig { -- /** Optional. Function calling mode. */ -- mode?: FunctionCallingConfigMode; -- /** Optional. Function names to call. Only set when the Mode is ANY. Function names should match [FunctionDeclaration.name]. With mode set to ANY, model will predict a function call from the set of function names provided. */ -- allowedFunctionNames?: string[]; --} -- --/** Config for the function calling config mode. */ --export declare enum FunctionCallingConfigMode { -- /** -- * The function calling config mode is unspecified. Should not be used. -- */ -- MODE_UNSPECIFIED = "MODE_UNSPECIFIED", -- /** -- * Default model behavior, model decides to predict either function calls or natural language response. -- */ -- AUTO = "AUTO", -- /** -- * Model is constrained to always predicting function calls only. If "allowed_function_names" are set, the predicted function calls will be limited to any one of "allowed_function_names", else the predicted function calls will be any one of the provided "function_declarations". -- */ -- ANY = "ANY", -- /** -- * Model will not predict any function calls. Model behavior is same as when not passing any function declarations. -- */ -- NONE = "NONE" --} -- --/** Defines a function that the model can generate JSON inputs for. -- -- The inputs are based on `OpenAPI 3.0 specifications -- `_. -- */ --export declare interface FunctionDeclaration { -- /** Defines the function behavior. */ -- behavior?: Behavior; -- /** Optional. Description and purpose of the function. Model uses it to decide how and whether to call the function. */ -- description?: string; -- /** Required. The name of the function to call. Must start with a letter or an underscore. Must be a-z, A-Z, 0-9, or contain underscores, dots and dashes, with a maximum length of 64. */ -- name?: string; -- /** Optional. Describes the parameters to this function in JSON Schema Object format. Reflects the Open API 3.03 Parameter Object. string Key: the name of the parameter. Parameter names are case sensitive. Schema Value: the Schema defining the type used for the parameter. For function with no parameters, this can be left unset. Parameter names must start with a letter or an underscore and must only contain chars a-z, A-Z, 0-9, or underscores with a maximum length of 64. Example with 1 required and 1 optional parameter: type: OBJECT properties: param1: type: STRING param2: type: INTEGER required: - param1 */ -- parameters?: Schema; -- /** Optional. Describes the output from this function in JSON Schema format. Reflects the Open API 3.03 Response Object. The Schema defines the type used for the response value of the function. */ -- response?: Schema; --} -- --/** A function response. */ --export declare class FunctionResponse { -- /** Signals that function call continues, and more responses will be returned, turning the function call into a generator. Is only applicable to NON_BLOCKING function calls (see FunctionDeclaration.behavior for details), ignored otherwise. If false, the default, future responses will not be considered. Is only applicable to NON_BLOCKING function calls, is ignored otherwise. If set to false, future responses will not be considered. It is allowed to return empty `response` with `will_continue=False` to signal that the function call is finished. */ -- willContinue?: boolean; -- /** Specifies how the response should be scheduled in the conversation. Only applicable to NON_BLOCKING function calls, is ignored otherwise. Defaults to WHEN_IDLE. */ -- scheduling?: FunctionResponseScheduling; -- /** Optional. The id of the function call this response is for. Populated by the client to match the corresponding function call `id`. */ -- id?: string; -- /** Required. The name of the function to call. Matches [FunctionDeclaration.name] and [FunctionCall.name]. */ -- name?: string; -- /** Required. The function response in JSON object format. Use "output" key to specify function output and "error" key to specify error details (if any). If "output" and "error" keys are not specified, then whole "response" is treated as function output. */ -- response?: Record; --} -- --/** Specifies how the response should be scheduled in the conversation. */ --export declare enum FunctionResponseScheduling { -- /** -- * This value is unused. -- */ -- SCHEDULING_UNSPECIFIED = "SCHEDULING_UNSPECIFIED", -- /** -- * Only add the result to the conversation context, do not interrupt or trigger generation. -- */ -- SILENT = "SILENT", -- /** -- * Add the result to the conversation context, and prompt to generate output without interrupting ongoing generation. -- */ -- WHEN_IDLE = "WHEN_IDLE", -- /** -- * Add the result to the conversation context, interrupt ongoing generation and prompt to generate output. -- */ -- INTERRUPT = "INTERRUPT" --} -- --/** Optional model configuration parameters. -- -- For more information, see `Content generation parameters -- `_. -- */ --export declare interface GenerateContentConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; -- /** Instructions for the model to steer it toward better performance. -- For example, "Answer as concisely as possible" or "Don't use technical -- terms in your response". -- */ -- systemInstruction?: ContentUnion; -- /** Value that controls the degree of randomness in token selection. -- Lower temperatures are good for prompts that require a less open-ended or -- creative response, while higher temperatures can lead to more diverse or -- creative results. -- */ -- temperature?: number; -- /** Tokens are selected from the most to least probable until the sum -- of their probabilities equals this value. Use a lower value for less -- random responses and a higher value for more random responses. -- */ -- topP?: number; -- /** For each token selection step, the ``top_k`` tokens with the -- highest probabilities are sampled. Then tokens are further filtered based -- on ``top_p`` with the final token selected using temperature sampling. Use -- a lower number for less random responses and a higher number for more -- random responses. -- */ -- topK?: number; -- /** Number of response variations to return. -- */ -- candidateCount?: number; -- /** Maximum number of tokens that can be generated in the response. -- */ -- maxOutputTokens?: number; -- /** List of strings that tells the model to stop generating text if one -- of the strings is encountered in the response. -- */ -- stopSequences?: string[]; -- /** Whether to return the log probabilities of the tokens that were -- chosen by the model at each step. -- */ -- responseLogprobs?: boolean; -- /** Number of top candidate tokens to return the log probabilities for -- at each generation step. -- */ -- logprobs?: number; -- /** Positive values penalize tokens that already appear in the -- generated text, increasing the probability of generating more diverse -- content. -- */ -- presencePenalty?: number; -- /** Positive values penalize tokens that repeatedly appear in the -- generated text, increasing the probability of generating more diverse -- content. -- */ -- frequencyPenalty?: number; -- /** When ``seed`` is fixed to a specific number, the model makes a best -- effort to provide the same response for repeated requests. By default, a -- random number is used. -- */ -- seed?: number; -- /** Output response mimetype of the generated candidate text. -- Supported mimetype: -- - `text/plain`: (default) Text output. -- - `application/json`: JSON response in the candidates. -- The model needs to be prompted to output the appropriate response type, -- otherwise the behavior is undefined. -- This is a preview feature. -- */ -- responseMimeType?: string; -- /** The `Schema` object allows the definition of input and output data types. -- These types can be objects, but also primitives and arrays. -- Represents a select subset of an [OpenAPI 3.0 schema -- object](https://spec.openapis.org/oas/v3.0.3#schema). -- If set, a compatible response_mime_type must also be set. -- Compatible mimetypes: `application/json`: Schema for JSON response. -- */ -- responseSchema?: SchemaUnion; -- /** Configuration for model router requests. -- */ -- routingConfig?: GenerationConfigRoutingConfig; -- /** Configuration for model selection. -- */ -- modelSelectionConfig?: ModelSelectionConfig; -- /** Safety settings in the request to block unsafe content in the -- response. -- */ -- safetySettings?: SafetySetting[]; -- /** Code that enables the system to interact with external systems to -- perform an action outside of the knowledge and scope of the model. -- */ -- tools?: ToolListUnion; -- /** Associates model output to a specific function call. -- */ -- toolConfig?: ToolConfig; -- /** Labels with user-defined metadata to break down billed charges. */ -- labels?: Record; -- /** Resource name of a context cache that can be used in subsequent -- requests. -- */ -- cachedContent?: string; -- /** The requested modalities of the response. Represents the set of -- modalities that the model can return. -- */ -- responseModalities?: string[]; -- /** If specified, the media resolution specified will be used. -- */ -- mediaResolution?: MediaResolution; -- /** The speech generation configuration. -- */ -- speechConfig?: SpeechConfigUnion; -- /** If enabled, audio timestamp will be included in the request to the -- model. -- */ -- audioTimestamp?: boolean; -- /** The configuration for automatic function calling. -- */ -- automaticFunctionCalling?: AutomaticFunctionCallingConfig; -- /** The thinking features configuration. -- */ -- thinkingConfig?: ThinkingConfig; --} -- --/** Config for models.generate_content parameters. */ --export declare interface GenerateContentParameters { -- /** ID of the model to use. For a list of models, see `Google models -- `_. */ -- model: string; -- /** Content of the request. -- */ -- contents: ContentListUnion; -- /** Configuration that contains optional model parameters. -- */ -- config?: GenerateContentConfig; --} -- --/** Response message for PredictionService.GenerateContent. */ --export declare class GenerateContentResponse { -- /** Response variations returned by the model. -- */ -- candidates?: Candidate[]; -- /** Timestamp when the request is made to the server. -- */ -- createTime?: string; -- /** Identifier for each response. -- */ -- responseId?: string; -- /** The history of automatic function calling. -- */ -- automaticFunctionCallingHistory?: Content[]; -- /** Output only. The model version used to generate the response. */ -- modelVersion?: string; -- /** Output only. Content filter results for a prompt sent in the request. Note: Sent only in the first stream chunk. Only happens when no candidates were generated due to content violations. */ -- promptFeedback?: GenerateContentResponsePromptFeedback; -- /** Usage metadata about the response(s). */ -- usageMetadata?: GenerateContentResponseUsageMetadata; -- /** -- * Returns the concatenation of all text parts from the first candidate in the response. -- * -- * @remarks -- * If there are multiple candidates in the response, the text from the first -- * one will be returned. -- * If there are non-text parts in the response, the concatenation of all text -- * parts will be returned, and a warning will be logged. -- * If there are thought parts in the response, the concatenation of all text -- * parts excluding the thought parts will be returned. -- * -- * @example -- * ```ts -- * const response = await ai.models.generateContent({ -- * model: 'gemini-2.0-flash', -- * contents: -- * 'Why is the sky blue?', -- * }); -- * -- * console.debug(response.text); -- * ``` -- */ -- get text(): string | undefined; -- /** -- * Returns the concatenation of all inline data parts from the first candidate -- * in the response. -- * -- * @remarks -- * If there are multiple candidates in the response, the inline data from the -- * first one will be returned. If there are non-inline data parts in the -- * response, the concatenation of all inline data parts will be returned, and -- * a warning will be logged. -- */ -- get data(): string | undefined; -- /** -- * Returns the function calls from the first candidate in the response. -- * -- * @remarks -- * If there are multiple candidates in the response, the function calls from -- * the first one will be returned. -- * If there are no function calls in the response, undefined will be returned. -- * -- * @example -- * ```ts -- * const controlLightFunctionDeclaration: FunctionDeclaration = { -- * name: 'controlLight', -- * parameters: { -- * type: Type.OBJECT, -- * description: 'Set the brightness and color temperature of a room light.', -- * properties: { -- * brightness: { -- * type: Type.NUMBER, -- * description: -- * 'Light level from 0 to 100. Zero is off and 100 is full brightness.', -- * }, -- * colorTemperature: { -- * type: Type.STRING, -- * description: -- * 'Color temperature of the light fixture which can be `daylight`, `cool` or `warm`.', -- * }, -- * }, -- * required: ['brightness', 'colorTemperature'], -- * }; -- * const response = await ai.models.generateContent({ -- * model: 'gemini-2.0-flash', -- * contents: 'Dim the lights so the room feels cozy and warm.', -- * config: { -- * tools: [{functionDeclarations: [controlLightFunctionDeclaration]}], -- * toolConfig: { -- * functionCallingConfig: { -- * mode: FunctionCallingConfigMode.ANY, -- * allowedFunctionNames: ['controlLight'], -- * }, -- * }, -- * }, -- * }); -- * console.debug(JSON.stringify(response.functionCalls)); -- * ``` -- */ -- get functionCalls(): FunctionCall[] | undefined; -- /** -- * Returns the first executable code from the first candidate in the response. -- * -- * @remarks -- * If there are multiple candidates in the response, the executable code from -- * the first one will be returned. -- * If there are no executable code in the response, undefined will be -- * returned. -- * -- * @example -- * ```ts -- * const response = await ai.models.generateContent({ -- * model: 'gemini-2.0-flash', -- * contents: -- * 'What is the sum of the first 50 prime numbers? Generate and run code for the calculation, and make sure you get all 50.' -- * config: { -- * tools: [{codeExecution: {}}], -- * }, -- * }); -- * -- * console.debug(response.executableCode); -- * ``` -- */ -- get executableCode(): string | undefined; -- /** -- * Returns the first code execution result from the first candidate in the response. -- * -- * @remarks -- * If there are multiple candidates in the response, the code execution result from -- * the first one will be returned. -- * If there are no code execution result in the response, undefined will be returned. -- * -- * @example -- * ```ts -- * const response = await ai.models.generateContent({ -- * model: 'gemini-2.0-flash', -- * contents: -- * 'What is the sum of the first 50 prime numbers? Generate and run code for the calculation, and make sure you get all 50.' -- * config: { -- * tools: [{codeExecution: {}}], -- * }, -- * }); -- * -- * console.debug(response.codeExecutionResult); -- * ``` -- */ -- get codeExecutionResult(): string | undefined; --} -- --/** Content filter results for a prompt sent in the request. */ --export declare class GenerateContentResponsePromptFeedback { -- /** Output only. Blocked reason. */ -- blockReason?: BlockedReason; -- /** Output only. A readable block reason message. */ -- blockReasonMessage?: string; -- /** Output only. Safety ratings. */ -- safetyRatings?: SafetyRating[]; --} -- --/** Usage metadata about response(s). */ --export declare class GenerateContentResponseUsageMetadata { -- /** Output only. List of modalities of the cached content in the request input. */ -- cacheTokensDetails?: ModalityTokenCount[]; -- /** Output only. Number of tokens in the cached part in the input (the cached content). */ -- cachedContentTokenCount?: number; -- /** Number of tokens in the response(s). */ -- candidatesTokenCount?: number; -- /** Output only. List of modalities that were returned in the response. */ -- candidatesTokensDetails?: ModalityTokenCount[]; -- /** Number of tokens in the request. When `cached_content` is set, this is still the total effective prompt size meaning this includes the number of tokens in the cached content. */ -- promptTokenCount?: number; -- /** Output only. List of modalities that were processed in the request input. */ -- promptTokensDetails?: ModalityTokenCount[]; -- /** Output only. Number of tokens present in thoughts output. */ -- thoughtsTokenCount?: number; -- /** Output only. Number of tokens present in tool-use prompt(s). */ -- toolUsePromptTokenCount?: number; -- /** Output only. List of modalities that were processed for tool-use request inputs. */ -- toolUsePromptTokensDetails?: ModalityTokenCount[]; -- /** Total token count for prompt, response candidates, and tool-use prompts (if present). */ -- totalTokenCount?: number; -- /** Output only. Traffic type. This shows whether a request consumes Pay-As-You-Go or Provisioned Throughput quota. */ -- trafficType?: TrafficType; --} -- --/** An output image. */ --export declare interface GeneratedImage { -- /** The output image data. -- */ -- image?: Image_2; -- /** Responsible AI filter reason if the image is filtered out of the -- response. -- */ -- raiFilteredReason?: string; -- /** Safety attributes of the image. Lists of RAI categories and their -- scores of each content. -- */ -- safetyAttributes?: SafetyAttributes; -- /** The rewritten prompt used for the image generation if the prompt -- enhancer is enabled. -- */ -- enhancedPrompt?: string; --} -- --/** A generated video. */ --export declare interface GeneratedVideo { -- /** The output video */ -- video?: Video; --} -- --/** The config for generating an images. */ --export declare interface GenerateImagesConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; -- /** Cloud Storage URI used to store the generated images. -- */ -- outputGcsUri?: string; -- /** Description of what to discourage in the generated images. -- */ -- negativePrompt?: string; -- /** Number of images to generate. -- */ -- numberOfImages?: number; -- /** Aspect ratio of the generated images. -- */ -- aspectRatio?: string; -- /** Controls how much the model adheres to the text prompt. Large -- values increase output and prompt alignment, but may compromise image -- quality. -- */ -- guidanceScale?: number; -- /** Random seed for image generation. This is not available when -- ``add_watermark`` is set to true. -- */ -- seed?: number; -- /** Filter level for safety filtering. -- */ -- safetyFilterLevel?: SafetyFilterLevel; -- /** Allows generation of people by the model. -- */ -- personGeneration?: PersonGeneration; -- /** Whether to report the safety scores of each generated image and -- the positive prompt in the response. -- */ -- includeSafetyAttributes?: boolean; -- /** Whether to include the Responsible AI filter reason if the image -- is filtered out of the response. -- */ -- includeRaiReason?: boolean; -- /** Language of the text in the prompt. -- */ -- language?: ImagePromptLanguage; -- /** MIME type of the generated image. -- */ -- outputMimeType?: string; -- /** Compression quality of the generated image (for ``image/jpeg`` -- only). -- */ -- outputCompressionQuality?: number; -- /** Whether to add a watermark to the generated images. -- */ -- addWatermark?: boolean; -- /** Whether to use the prompt rewriting logic. -- */ -- enhancePrompt?: boolean; --} -- --/** The parameters for generating images. */ --export declare interface GenerateImagesParameters { -- /** ID of the model to use. For a list of models, see `Google models -- `_. */ -- model: string; -- /** Text prompt that typically describes the images to output. -- */ -- prompt: string; -- /** Configuration for generating images. -- */ -- config?: GenerateImagesConfig; --} -- --/** The output images response. */ --export declare class GenerateImagesResponse { -- /** List of generated images. -- */ -- generatedImages?: GeneratedImage[]; -- /** Safety attributes of the positive prompt. Only populated if -- ``include_safety_attributes`` is set to True. -- */ -- positivePromptSafetyAttributes?: SafetyAttributes; --} -- --/** Configuration for generating videos. */ --export declare interface GenerateVideosConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; -- /** Number of output videos. */ -- numberOfVideos?: number; -- /** The gcs bucket where to save the generated videos. */ -- outputGcsUri?: string; -- /** Frames per second for video generation. */ -- fps?: number; -- /** Duration of the clip for video generation in seconds. */ -- durationSeconds?: number; -- /** The RNG seed. If RNG seed is exactly same for each request with unchanged inputs, the prediction results will be consistent. Otherwise, a random RNG seed will be used each time to produce a different result. */ -- seed?: number; -- /** The aspect ratio for the generated video. 16:9 (landscape) and 9:16 (portrait) are supported. */ -- aspectRatio?: string; -- /** The resolution for the generated video. 1280x720, 1920x1080 are supported. */ -- resolution?: string; -- /** Whether allow to generate person videos, and restrict to specific ages. Supported values are: dont_allow, allow_adult. */ -- personGeneration?: string; -- /** The pubsub topic where to publish the video generation progress. */ -- pubsubTopic?: string; -- /** Optional field in addition to the text content. Negative prompts can be explicitly stated here to help generate the video. */ -- negativePrompt?: string; -- /** Whether to use the prompt rewriting logic. */ -- enhancePrompt?: boolean; --} -- --/** A video generation operation. */ --export declare interface GenerateVideosOperation { -- /** The server-assigned name, which is only unique within the same service that originally returns it. If you use the default HTTP mapping, the `name` should be a resource name ending with `operations/{unique_id}`. */ -- name?: string; -- /** Service-specific metadata associated with the operation. It typically contains progress information and common metadata such as create time. Some services might not provide such metadata. Any method that returns a long-running operation should document the metadata type, if any. */ -- metadata?: Record; -- /** If the value is `false`, it means the operation is still in progress. If `true`, the operation is completed, and either `error` or `response` is available. */ -- done?: boolean; -- /** The error result of the operation in case of failure or cancellation. */ -- error?: Record; -- /** The generated videos. */ -- response?: GenerateVideosResponse; --} -- --/** Class that represents the parameters for generating an image. */ --export declare interface GenerateVideosParameters { -- /** ID of the model to use. For a list of models, see `Google models -- `_. */ -- model: string; -- /** The text prompt for generating the videos. Optional for image to video use cases. */ -- prompt?: string; -- /** The input image for generating the videos. -- Optional if prompt is provided. */ -- image?: Image_2; -- /** Configuration for generating videos. */ -- config?: GenerateVideosConfig; --} -- --/** Response with generated videos. */ --export declare class GenerateVideosResponse { -- /** List of the generated videos */ -- generatedVideos?: GeneratedVideo[]; -- /** Returns if any videos were filtered due to RAI policies. */ -- raiMediaFilteredCount?: number; -- /** Returns rai failure reasons if any. */ -- raiMediaFilteredReasons?: string[]; --} -- --/** Generation config. */ --export declare interface GenerationConfig { -- /** Optional. If enabled, audio timestamp will be included in the request to the model. */ -- audioTimestamp?: boolean; -- /** Optional. Number of candidates to generate. */ -- candidateCount?: number; -- /** Optional. Frequency penalties. */ -- frequencyPenalty?: number; -- /** Optional. Logit probabilities. */ -- logprobs?: number; -- /** Optional. The maximum number of output tokens to generate per message. */ -- maxOutputTokens?: number; -- /** Optional. If specified, the media resolution specified will be used. */ -- mediaResolution?: MediaResolution; -- /** Optional. Positive penalties. */ -- presencePenalty?: number; -- /** Optional. If true, export the logprobs results in response. */ -- responseLogprobs?: boolean; -- /** Optional. Output response mimetype of the generated candidate text. Supported mimetype: - `text/plain`: (default) Text output. - `application/json`: JSON response in the candidates. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. This is a preview feature. */ -- responseMimeType?: string; -- /** Optional. The `Schema` object allows the definition of input and output data types. These types can be objects, but also primitives and arrays. Represents a select subset of an [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema). If set, a compatible response_mime_type must also be set. Compatible mimetypes: `application/json`: Schema for JSON response. */ -- responseSchema?: Schema; -- /** Optional. Routing configuration. */ -- routingConfig?: GenerationConfigRoutingConfig; -- /** Optional. Seed. */ -- seed?: number; -- /** Optional. Stop sequences. */ -- stopSequences?: string[]; -- /** Optional. Controls the randomness of predictions. */ -- temperature?: number; -- /** Optional. If specified, top-k sampling will be used. */ -- topK?: number; -- /** Optional. If specified, nucleus sampling will be used. */ -- topP?: number; --} -- --/** The configuration for routing the request to a specific model. */ --export declare interface GenerationConfigRoutingConfig { -- /** Automated routing. */ -- autoMode?: GenerationConfigRoutingConfigAutoRoutingMode; -- /** Manual routing. */ -- manualMode?: GenerationConfigRoutingConfigManualRoutingMode; --} -- --/** When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. */ --export declare interface GenerationConfigRoutingConfigAutoRoutingMode { -- /** The model routing preference. */ -- modelRoutingPreference?: 'UNKNOWN' | 'PRIORITIZE_QUALITY' | 'BALANCED' | 'PRIORITIZE_COST'; --} -- --/** When manual routing is set, the specified model will be used directly. */ --export declare interface GenerationConfigRoutingConfigManualRoutingMode { -- /** The model name to use. Only the public LLM models are accepted. e.g. 'gemini-1.5-pro-001'. */ -- modelName?: string; --} -- --/** Optional parameters for caches.get method. */ --export declare interface GetCachedContentConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; --} -- --/** Parameters for caches.get method. */ --export declare interface GetCachedContentParameters { -- /** The server-generated resource name of the cached content. -- */ -- name: string; -- /** Optional parameters for the request. -- */ -- config?: GetCachedContentConfig; --} -- --/** Used to override the default configuration. */ --export declare interface GetFileConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; --} -- --/** Generates the parameters for the get method. */ --export declare interface GetFileParameters { -- /** The name identifier for the file to retrieve. */ -- name: string; -- /** Used to override the default configuration. */ -- config?: GetFileConfig; --} -- --/** Optional parameters for models.get method. */ --export declare interface GetModelConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; --} -- --export declare interface GetModelParameters { -- model: string; -- /** Optional parameters for the request. */ -- config?: GetModelConfig; --} -- --export declare interface GetOperationConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; --} -- --/** Parameters for the GET method. */ --export declare interface GetOperationParameters { -- /** The server-assigned name for the operation. */ -- operationName: string; -- /** Used to override the default configuration. */ -- config?: GetOperationConfig; --} -- --/** Optional parameters for tunings.get method. */ --export declare interface GetTuningJobConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; --} -- --/** Parameters for the get method. */ --export declare interface GetTuningJobParameters { -- name: string; -- /** Optional parameters for the request. */ -- config?: GetTuningJobConfig; --} -- --/** -- * The Google GenAI SDK. -- * -- * @remarks -- * Provides access to the GenAI features through either the {@link -- * https://cloud.google.com/vertex-ai/docs/reference/rest | Gemini API} or -- * the {@link https://cloud.google.com/vertex-ai/docs/reference/rest | Vertex AI -- * API}. -- * -- * The {@link GoogleGenAIOptions.vertexai} value determines which of the API -- * services to use. -- * -- * When using the Gemini API, a {@link GoogleGenAIOptions.apiKey} must also be -- * set. When using Vertex AI, currently only {@link GoogleGenAIOptions.apiKey} -- * is supported via Express mode. {@link GoogleGenAIOptions.project} and {@link -- * GoogleGenAIOptions.location} should not be set. -- * -- * @example -- * Initializing the SDK for using the Gemini API: -- * ```ts -- * import {GoogleGenAI} from '@google/genai'; -- * const ai = new GoogleGenAI({apiKey: 'GEMINI_API_KEY'}); -- * ``` -- * -- * @example -- * Initializing the SDK for using the Vertex AI API: -- * ```ts -- * import {GoogleGenAI} from '@google/genai'; -- * const ai = new GoogleGenAI({ -- * vertexai: true, -- * project: 'PROJECT_ID', -- * location: 'PROJECT_LOCATION' -- * }); -- * ``` -- * -- */ --export declare class GoogleGenAI { -- protected readonly apiClient: ApiClient; -- private readonly apiKey?; -- readonly vertexai: boolean; -- private readonly apiVersion?; -- readonly models: Models; -- readonly live: Live; -- readonly chats: Chats; -- readonly caches: Caches; -- readonly files: Files; -- readonly operations: Operations; -- readonly tunings: Tunings; -- constructor(options: GoogleGenAIOptions); --} -- --/** -- * Google Gen AI SDK's configuration options. -- * -- * See {@link GoogleGenAI} for usage samples. -- */ --export declare interface GoogleGenAIOptions { -- /** -- * Optional. Determines whether to use the Vertex AI or the Gemini API. -- * -- * @remarks -- * When true, the {@link https://cloud.google.com/vertex-ai/docs/reference/rest | Vertex AI API} will used. -- * When false, the {@link https://ai.google.dev/api | Gemini API} will be used. -- * -- * If unset, default SDK behavior is to use the Gemini API service. -- */ -- vertexai?: boolean; -- /** -- * Optional. The Google Cloud project ID for Vertex AI clients. -- * -- * Find your project ID: https://cloud.google.com/resource-manager/docs/creating-managing-projects#identifying_projects -- * -- * @remarks -- * Only supported on Node runtimes, ignored on browser runtimes. -- */ -- project?: string; -- /** -- * Optional. The Google Cloud project {@link https://cloud.google.com/vertex-ai/generative-ai/docs/learn/locations | location} for Vertex AI clients. -- * -- * @remarks -- * Only supported on Node runtimes, ignored on browser runtimes. -- * -- */ -- location?: string; -- /** -- * The API Key, required for Gemini API clients. -- * -- * @remarks -- * Required on browser runtimes. -- */ -- apiKey?: string; -- /** -- * Optional. The API version to use. -- * -- * @remarks -- * If unset, the default API version will be used. -- */ -- apiVersion?: string; -- /** -- * Optional. Authentication options defined by the by google-auth-library for Vertex AI clients. -- * -- * @remarks -- * @see {@link https://github.com/googleapis/google-auth-library-nodejs/blob/v9.15.0/src/auth/googleauth.ts | GoogleAuthOptions interface in google-auth-library-nodejs}. -- * -- * Only supported on Node runtimes, ignored on browser runtimes. -- * -- */ -- googleAuthOptions?: GoogleAuthOptions; -- /** -- * Optional. A set of customizable configuration for HTTP requests. -- */ -- httpOptions?: HttpOptions; --} -- --/** Tool to support Google Maps in Model. */ --export declare interface GoogleMaps { -- /** Optional. Auth config for the Google Maps tool. */ -- authConfig?: AuthConfig; --} -- --/** The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). */ --export declare interface GoogleRpcStatus { -- /** The status code, which should be an enum value of google.rpc.Code. */ -- code?: number; -- /** A list of messages that carry the error details. There is a common set of message types for APIs to use. */ -- details?: Record[]; -- /** A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. */ -- message?: string; --} -- --/** Tool to support Google Search in Model. Powered by Google. */ --export declare interface GoogleSearch { -- /** Optional. Filter search results to a specific time range. -- If customers set a start time, they must set an end time (and vice versa). -- */ -- timeRangeFilter?: Interval; --} -- --/** Tool to retrieve public web data for grounding, powered by Google. */ --export declare interface GoogleSearchRetrieval { -- /** Specifies the dynamic retrieval configuration for the given source. */ -- dynamicRetrievalConfig?: DynamicRetrievalConfig; --} -- --/** Represents a whole or partial calendar date, such as a birthday. The time of day and time zone are either specified elsewhere or are insignificant. The date is relative to the Gregorian Calendar. This can represent one of the following: * A full date, with non-zero year, month, and day values. * A month and day, with a zero year (for example, an anniversary). * A year on its own, with a zero month and a zero day. * A year and month, with a zero day (for example, a credit card expiration date). Related types: * google.type.TimeOfDay * google.type.DateTime * google.protobuf.Timestamp */ --export declare interface GoogleTypeDate { -- /** Day of a month. Must be from 1 to 31 and valid for the year and month, or 0 to specify a year by itself or a year and month where the day isn't significant. */ -- day?: number; -- /** Month of a year. Must be from 1 to 12, or 0 to specify a year without a month and day. */ -- month?: number; -- /** Year of the date. Must be from 1 to 9999, or 0 to specify a date without a year. */ -- year?: number; --} -- --/** Grounding chunk. */ --export declare interface GroundingChunk { -- /** Grounding chunk from context retrieved by the retrieval tools. */ -- retrievedContext?: GroundingChunkRetrievedContext; -- /** Grounding chunk from the web. */ -- web?: GroundingChunkWeb; --} -- --/** Chunk from context retrieved by the retrieval tools. */ --export declare interface GroundingChunkRetrievedContext { -- /** Text of the attribution. */ -- text?: string; -- /** Title of the attribution. */ -- title?: string; -- /** URI reference of the attribution. */ -- uri?: string; --} -- --/** Chunk from the web. */ --export declare interface GroundingChunkWeb { -- /** Domain of the (original) URI. */ -- domain?: string; -- /** Title of the chunk. */ -- title?: string; -- /** URI reference of the chunk. */ -- uri?: string; --} -- --/** Metadata returned to client when grounding is enabled. */ --export declare interface GroundingMetadata { -- /** List of supporting references retrieved from specified grounding source. */ -- groundingChunks?: GroundingChunk[]; -- /** Optional. List of grounding support. */ -- groundingSupports?: GroundingSupport[]; -- /** Optional. Output only. Retrieval metadata. */ -- retrievalMetadata?: RetrievalMetadata; -- /** Optional. Queries executed by the retrieval tools. */ -- retrievalQueries?: string[]; -- /** Optional. Google search entry for the following-up web searches. */ -- searchEntryPoint?: SearchEntryPoint; -- /** Optional. Web search queries for the following-up web search. */ -- webSearchQueries?: string[]; --} -- --/** Grounding support. */ --export declare interface GroundingSupport { -- /** Confidence score of the support references. Ranges from 0 to 1. 1 is the most confident. This list must have the same size as the grounding_chunk_indices. */ -- confidenceScores?: number[]; -- /** A list of indices (into 'grounding_chunk') specifying the citations associated with the claim. For instance [1,3,4] means that grounding_chunk[1], grounding_chunk[3], grounding_chunk[4] are the retrieved content attributed to the claim. */ -- groundingChunkIndices?: number[]; -- /** Segment of the content this support belongs to. */ -- segment?: Segment; --} -- --/** Optional. Specify if the threshold is used for probability or severity score. If not specified, the threshold is used for probability score. */ --export declare enum HarmBlockMethod { -- /** -- * The harm block method is unspecified. -- */ -- HARM_BLOCK_METHOD_UNSPECIFIED = "HARM_BLOCK_METHOD_UNSPECIFIED", -- /** -- * The harm block method uses both probability and severity scores. -- */ -- SEVERITY = "SEVERITY", -- /** -- * The harm block method uses the probability score. -- */ -- PROBABILITY = "PROBABILITY" --} -- --/** Required. The harm block threshold. */ --export declare enum HarmBlockThreshold { -- /** -- * Unspecified harm block threshold. -- */ -- HARM_BLOCK_THRESHOLD_UNSPECIFIED = "HARM_BLOCK_THRESHOLD_UNSPECIFIED", -- /** -- * Block low threshold and above (i.e. block more). -- */ -- BLOCK_LOW_AND_ABOVE = "BLOCK_LOW_AND_ABOVE", -- /** -- * Block medium threshold and above. -- */ -- BLOCK_MEDIUM_AND_ABOVE = "BLOCK_MEDIUM_AND_ABOVE", -- /** -- * Block only high threshold (i.e. block less). -- */ -- BLOCK_ONLY_HIGH = "BLOCK_ONLY_HIGH", -- /** -- * Block none. -- */ -- BLOCK_NONE = "BLOCK_NONE", -- /** -- * Turn off the safety filter. -- */ -- OFF = "OFF" --} -- --/** Required. Harm category. */ --export declare enum HarmCategory { -- /** -- * The harm category is unspecified. -- */ -- HARM_CATEGORY_UNSPECIFIED = "HARM_CATEGORY_UNSPECIFIED", -- /** -- * The harm category is hate speech. -- */ -- HARM_CATEGORY_HATE_SPEECH = "HARM_CATEGORY_HATE_SPEECH", -- /** -- * The harm category is dangerous content. -- */ -- HARM_CATEGORY_DANGEROUS_CONTENT = "HARM_CATEGORY_DANGEROUS_CONTENT", -- /** -- * The harm category is harassment. -- */ -- HARM_CATEGORY_HARASSMENT = "HARM_CATEGORY_HARASSMENT", -- /** -- * The harm category is sexually explicit content. -- */ -- HARM_CATEGORY_SEXUALLY_EXPLICIT = "HARM_CATEGORY_SEXUALLY_EXPLICIT", -- /** -- * The harm category is civic integrity. -- */ -- HARM_CATEGORY_CIVIC_INTEGRITY = "HARM_CATEGORY_CIVIC_INTEGRITY" --} -- --/** Output only. Harm probability levels in the content. */ --export declare enum HarmProbability { -- /** -- * Harm probability unspecified. -- */ -- HARM_PROBABILITY_UNSPECIFIED = "HARM_PROBABILITY_UNSPECIFIED", -- /** -- * Negligible level of harm. -- */ -- NEGLIGIBLE = "NEGLIGIBLE", -- /** -- * Low level of harm. -- */ -- LOW = "LOW", -- /** -- * Medium level of harm. -- */ -- MEDIUM = "MEDIUM", -- /** -- * High level of harm. -- */ -- HIGH = "HIGH" --} -- --/** Output only. Harm severity levels in the content. */ --export declare enum HarmSeverity { -- /** -- * Harm severity unspecified. -- */ -- HARM_SEVERITY_UNSPECIFIED = "HARM_SEVERITY_UNSPECIFIED", -- /** -- * Negligible level of harm severity. -- */ -- HARM_SEVERITY_NEGLIGIBLE = "HARM_SEVERITY_NEGLIGIBLE", -- /** -- * Low level of harm severity. -- */ -- HARM_SEVERITY_LOW = "HARM_SEVERITY_LOW", -- /** -- * Medium level of harm severity. -- */ -- HARM_SEVERITY_MEDIUM = "HARM_SEVERITY_MEDIUM", -- /** -- * High level of harm severity. -- */ -- HARM_SEVERITY_HIGH = "HARM_SEVERITY_HIGH" --} -- --/** HTTP options to be used in each of the requests. */ --export declare interface HttpOptions { -- /** The base URL for the AI platform service endpoint. */ -- baseUrl?: string; -- /** Specifies the version of the API to use. */ -- apiVersion?: string; -- /** Additional HTTP headers to be sent with the request. */ -- headers?: Record; -- /** Timeout for the request in milliseconds. */ -- timeout?: number; --} -- --/** -- * Represents the necessary information to send a request to an API endpoint. -- * This interface defines the structure for constructing and executing HTTP -- * requests. -- */ --declare interface HttpRequest { -- /** -- * URL path from the modules, this path is appended to the base API URL to -- * form the complete request URL. -- * -- * If you wish to set full URL, use httpOptions.baseUrl instead. Example to -- * set full URL in the request: -- * -- * const request: HttpRequest = { -- * path: '', -- * httpOptions: { -- * baseUrl: 'https://', -- * apiVersion: '', -- * }, -- * httpMethod: 'GET', -- * }; -- * -- * The result URL will be: https:// -- * -- */ -- path: string; -- /** -- * Optional query parameters to be appended to the request URL. -- */ -- queryParams?: Record; -- /** -- * Optional request body in json string or Blob format, GET request doesn't -- * need a request body. -- */ -- body?: string | Blob; -- /** -- * The HTTP method to be used for the request. -- */ -- httpMethod: 'GET' | 'POST' | 'PATCH' | 'DELETE'; -- /** -- * Optional set of customizable configuration for HTTP requests. -- */ -- httpOptions?: HttpOptions; -- /** -- * Optional abort signal which can be used to cancel the request. -- */ -- abortSignal?: AbortSignal; --} -- --/** A wrapper class for the http response. */ --export declare class HttpResponse { -- /** Used to retain the processed HTTP headers in the response. */ -- headers?: Record; -- /** -- * The original http response. -- */ -- responseInternal: Response; -- constructor(response: Response); -- json(): Promise; --} -- --/** An image. */ --declare interface Image_2 { -- /** The Cloud Storage URI of the image. ``Image`` can contain a value -- for this field or the ``image_bytes`` field but not both. -- */ -- gcsUri?: string; -- /** The image bytes data. ``Image`` can contain a value for this field -- or the ``gcs_uri`` field but not both. -- */ -- imageBytes?: string; -- /** The MIME type of the image. */ -- mimeType?: string; --} --export { Image_2 as Image } -- --/** Enum that specifies the language of the text in the prompt. */ --export declare enum ImagePromptLanguage { -- auto = "auto", -- en = "en", -- ja = "ja", -- ko = "ko", -- hi = "hi" --} -- --/** Represents a time interval, encoded as a start time (inclusive) and an end time (exclusive). -- -- The start time must be less than or equal to the end time. -- When the start equals the end time, the interval is an empty interval. -- (matches no time) -- When both start and end are unspecified, the interval matches any time. -- */ --export declare interface Interval { -- /** The start time of the interval. */ -- startTime?: string; -- /** The end time of the interval. */ -- endTime?: string; --} -- --/** Output only. The detailed state of the job. */ --export declare enum JobState { -- /** -- * The job state is unspecified. -- */ -- JOB_STATE_UNSPECIFIED = "JOB_STATE_UNSPECIFIED", -- /** -- * The job has been just created or resumed and processing has not yet begun. -- */ -- JOB_STATE_QUEUED = "JOB_STATE_QUEUED", -- /** -- * The service is preparing to run the job. -- */ -- JOB_STATE_PENDING = "JOB_STATE_PENDING", -- /** -- * The job is in progress. -- */ -- JOB_STATE_RUNNING = "JOB_STATE_RUNNING", -- /** -- * The job completed successfully. -- */ -- JOB_STATE_SUCCEEDED = "JOB_STATE_SUCCEEDED", -- /** -- * The job failed. -- */ -- JOB_STATE_FAILED = "JOB_STATE_FAILED", -- /** -- * The job is being cancelled. From this state the job may only go to either `JOB_STATE_SUCCEEDED`, `JOB_STATE_FAILED` or `JOB_STATE_CANCELLED`. -- */ -- JOB_STATE_CANCELLING = "JOB_STATE_CANCELLING", -- /** -- * The job has been cancelled. -- */ -- JOB_STATE_CANCELLED = "JOB_STATE_CANCELLED", -- /** -- * The job has been stopped, and can be resumed. -- */ -- JOB_STATE_PAUSED = "JOB_STATE_PAUSED", -- /** -- * The job has expired. -- */ -- JOB_STATE_EXPIRED = "JOB_STATE_EXPIRED", -- /** -- * The job is being updated. Only jobs in the `RUNNING` state can be updated. After updating, the job goes back to the `RUNNING` state. -- */ -- JOB_STATE_UPDATING = "JOB_STATE_UPDATING", -- /** -- * The job is partially succeeded, some results may be missing due to errors. -- */ -- JOB_STATE_PARTIALLY_SUCCEEDED = "JOB_STATE_PARTIALLY_SUCCEEDED" --} -- --/** Required. Programming language of the `code`. */ --export declare enum Language { -- /** -- * Unspecified language. This value should not be used. -- */ -- LANGUAGE_UNSPECIFIED = "LANGUAGE_UNSPECIFIED", -- /** -- * Python >= 3.10, with numpy and simpy available. -- */ -- PYTHON = "PYTHON" --} -- --/** An object that represents a latitude/longitude pair. -- -- This is expressed as a pair of doubles to represent degrees latitude and -- degrees longitude. Unless specified otherwise, this object must conform to the -- -- WGS84 standard. Values must be within normalized ranges. -- */ --export declare interface LatLng { -- /** The latitude in degrees. It must be in the range [-90.0, +90.0]. */ -- latitude?: number; -- /** The longitude in degrees. It must be in the range [-180.0, +180.0] */ -- longitude?: number; --} -- --/** Config for caches.list method. */ --export declare interface ListCachedContentsConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; -- pageSize?: number; -- pageToken?: string; --} -- --/** Parameters for caches.list method. */ --export declare interface ListCachedContentsParameters { -- /** Configuration that contains optional parameters. -- */ -- config?: ListCachedContentsConfig; --} -- --export declare class ListCachedContentsResponse { -- nextPageToken?: string; -- /** List of cached contents. -- */ -- cachedContents?: CachedContent[]; --} -- --/** Used to override the default configuration. */ --export declare interface ListFilesConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; -- pageSize?: number; -- pageToken?: string; --} -- --/** Generates the parameters for the list method. */ --export declare interface ListFilesParameters { -- /** Used to override the default configuration. */ -- config?: ListFilesConfig; --} -- --/** Response for the list files method. */ --export declare class ListFilesResponse { -- /** A token to retrieve next page of results. */ -- nextPageToken?: string; -- /** The list of files. */ -- files?: File_2[]; --} -- --export declare interface ListModelsConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; -- pageSize?: number; -- pageToken?: string; -- filter?: string; -- /** Set true to list base models, false to list tuned models. */ -- queryBase?: boolean; --} -- --export declare interface ListModelsParameters { -- config?: ListModelsConfig; --} -- --export declare class ListModelsResponse { -- nextPageToken?: string; -- models?: Model[]; --} -- --/** Configuration for the list tuning jobs method. */ --export declare interface ListTuningJobsConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; -- pageSize?: number; -- pageToken?: string; -- filter?: string; --} -- --/** Parameters for the list tuning jobs method. */ --export declare interface ListTuningJobsParameters { -- config?: ListTuningJobsConfig; --} -- --/** Response for the list tuning jobs method. */ --export declare class ListTuningJobsResponse { -- /** A token to retrieve the next page of results. Pass to ListTuningJobsRequest.page_token to obtain that page. */ -- nextPageToken?: string; -- /** List of TuningJobs in the requested page. */ -- tuningJobs?: TuningJob[]; --} -- --/** -- Live class encapsulates the configuration for live interaction with the -- Generative Language API. It embeds ApiClient for general API settings. -- -- @experimental -- */ --export declare class Live { -- private readonly apiClient; -- private readonly auth; -- private readonly webSocketFactory; -- readonly music: LiveMusic; -- constructor(apiClient: ApiClient, auth: Auth, webSocketFactory: WebSocketFactory); -- /** -- Establishes a connection to the specified model with the given -- configuration and returns a Session object representing that connection. -- -- @experimental Built-in MCP support is an experimental feature, may change in -- future versions. -- -- @remarks -- -- @param params - The parameters for establishing a connection to the model. -- @return A live session. -- -- @example -- ```ts -- let model: string; -- if (GOOGLE_GENAI_USE_VERTEXAI) { -- model = 'gemini-2.0-flash-live-preview-04-09'; -- } else { -- model = 'gemini-2.0-flash-live-001'; -- } -- const session = await ai.live.connect({ -- model: model, -- config: { -- responseModalities: [Modality.AUDIO], -- }, -- callbacks: { -- onopen: () => { -- console.log('Connected to the socket.'); -- }, -- onmessage: (e: MessageEvent) => { -- console.log('Received message from the server: %s\n', debug(e.data)); -- }, -- onerror: (e: ErrorEvent) => { -- console.log('Error occurred: %s\n', debug(e.error)); -- }, -- onclose: (e: CloseEvent) => { -- console.log('Connection closed.'); -- }, -- }, -- }); -- ``` -- */ -- connect(params: types.LiveConnectParameters): Promise; -- private isCallableTool; --} -- --/** Callbacks for the live API. */ --export declare interface LiveCallbacks { -- /** -- * Called when the websocket connection is established. -- */ -- onopen?: (() => void) | null; -- /** -- * Called when a message is received from the server. -- */ -- onmessage: (e: LiveServerMessage) => void; -- /** -- * Called when an error occurs. -- */ -- onerror?: ((e: ErrorEvent) => void) | null; -- /** -- * Called when the websocket connection is closed. -- */ -- onclose?: ((e: CloseEvent) => void) | null; --} -- --/** Incremental update of the current conversation delivered from the client. -- -- All the content here will unconditionally be appended to the conversation -- history and used as part of the prompt to the model to generate content. -- -- A message here will interrupt any current model generation. -- */ --export declare interface LiveClientContent { -- /** The content appended to the current conversation with the model. -- -- For single-turn queries, this is a single instance. For multi-turn -- queries, this is a repeated field that contains conversation history and -- latest request. -- */ -- turns?: Content[]; -- /** If true, indicates that the server content generation should start with -- the currently accumulated prompt. Otherwise, the server will await -- additional messages before starting generation. */ -- turnComplete?: boolean; --} -- --/** Messages sent by the client in the API call. */ --export declare interface LiveClientMessage { -- /** Message to be sent by the system when connecting to the API. SDK users should not send this message. */ -- setup?: LiveClientSetup; -- /** Incremental update of the current conversation delivered from the client. */ -- clientContent?: LiveClientContent; -- /** User input that is sent in real time. */ -- realtimeInput?: LiveClientRealtimeInput; -- /** Response to a `ToolCallMessage` received from the server. */ -- toolResponse?: LiveClientToolResponse; --} -- --/** User input that is sent in real time. -- -- This is different from `LiveClientContent` in a few ways: -- -- - Can be sent continuously without interruption to model generation. -- - If there is a need to mix data interleaved across the -- `LiveClientContent` and the `LiveClientRealtimeInput`, server attempts to -- optimize for best response, but there are no guarantees. -- - End of turn is not explicitly specified, but is rather derived from user -- activity (for example, end of speech). -- - Even before the end of turn, the data is processed incrementally -- to optimize for a fast start of the response from the model. -- - Is always assumed to be the user's input (cannot be used to populate -- conversation history). -- */ --export declare interface LiveClientRealtimeInput { -- /** Inlined bytes data for media input. */ -- mediaChunks?: Blob_2[]; -- /** The realtime audio input stream. */ -- audio?: Blob_2; -- /** -- Indicates that the audio stream has ended, e.g. because the microphone was -- turned off. -- -- This should only be sent when automatic activity detection is enabled -- (which is the default). -- -- The client can reopen the stream by sending an audio message. -- */ -- audioStreamEnd?: boolean; -- /** The realtime video input stream. */ -- video?: Blob_2; -- /** The realtime text input stream. */ -- text?: string; -- /** Marks the start of user activity. */ -- activityStart?: ActivityStart; -- /** Marks the end of user activity. */ -- activityEnd?: ActivityEnd; --} -- --/** Message contains configuration that will apply for the duration of the streaming session. */ --export declare interface LiveClientSetup { -- /** -- The fully qualified name of the publisher model or tuned model endpoint to -- use. -- */ -- model?: string; -- /** The generation configuration for the session. -- Note: only a subset of fields are supported. -- */ -- generationConfig?: GenerationConfig; -- /** The user provided system instructions for the model. -- Note: only text should be used in parts and content in each part will be -- in a separate paragraph. */ -- systemInstruction?: ContentUnion; -- /** A list of `Tools` the model may use to generate the next response. -- -- A `Tool` is a piece of code that enables the system to interact with -- external systems to perform an action, or set of actions, outside of -- knowledge and scope of the model. */ -- tools?: ToolListUnion; -- /** Configures the realtime input behavior in BidiGenerateContent. */ -- realtimeInputConfig?: RealtimeInputConfig; -- /** Configures session resumption mechanism. -- -- If included server will send SessionResumptionUpdate messages. */ -- sessionResumption?: SessionResumptionConfig; -- /** Configures context window compression mechanism. -- -- If included, server will compress context window to fit into given length. */ -- contextWindowCompression?: ContextWindowCompressionConfig; -- /** The transcription of the input aligns with the input audio language. -- */ -- inputAudioTranscription?: AudioTranscriptionConfig; -- /** The transcription of the output aligns with the language code -- specified for the output audio. -- */ -- outputAudioTranscription?: AudioTranscriptionConfig; -- /** Configures the proactivity of the model. This allows the model to respond proactively to -- the input and to ignore irrelevant input. */ -- proactivity?: ProactivityConfig; --} -- --/** Client generated response to a `ToolCall` received from the server. -- -- Individual `FunctionResponse` objects are matched to the respective -- `FunctionCall` objects by the `id` field. -- -- Note that in the unary and server-streaming GenerateContent APIs function -- calling happens by exchanging the `Content` parts, while in the bidi -- GenerateContent APIs function calling happens over this dedicated set of -- messages. -- */ --export declare class LiveClientToolResponse { -- /** The response to the function calls. */ -- functionResponses?: FunctionResponse[]; --} -- --/** Session config for the API connection. */ --export declare interface LiveConnectConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; -- /** The generation configuration for the session. */ -- generationConfig?: GenerationConfig; -- /** The requested modalities of the response. Represents the set of -- modalities that the model can return. Defaults to AUDIO if not specified. -- */ -- responseModalities?: Modality[]; -- /** Value that controls the degree of randomness in token selection. -- Lower temperatures are good for prompts that require a less open-ended or -- creative response, while higher temperatures can lead to more diverse or -- creative results. -- */ -- temperature?: number; -- /** Tokens are selected from the most to least probable until the sum -- of their probabilities equals this value. Use a lower value for less -- random responses and a higher value for more random responses. -- */ -- topP?: number; -- /** For each token selection step, the ``top_k`` tokens with the -- highest probabilities are sampled. Then tokens are further filtered based -- on ``top_p`` with the final token selected using temperature sampling. Use -- a lower number for less random responses and a higher number for more -- random responses. -- */ -- topK?: number; -- /** Maximum number of tokens that can be generated in the response. -- */ -- maxOutputTokens?: number; -- /** If specified, the media resolution specified will be used. -- */ -- mediaResolution?: MediaResolution; -- /** When ``seed`` is fixed to a specific number, the model makes a best -- effort to provide the same response for repeated requests. By default, a -- random number is used. -- */ -- seed?: number; -- /** The speech generation configuration. -- */ -- speechConfig?: SpeechConfig; -- /** If enabled, the model will detect emotions and adapt its responses accordingly. */ -- enableAffectiveDialog?: boolean; -- /** The user provided system instructions for the model. -- Note: only text should be used in parts and content in each part will be -- in a separate paragraph. */ -- systemInstruction?: ContentUnion; -- /** A list of `Tools` the model may use to generate the next response. -- -- A `Tool` is a piece of code that enables the system to interact with -- external systems to perform an action, or set of actions, outside of -- knowledge and scope of the model. */ -- tools?: ToolListUnion; -- /** Configures session resumption mechanism. -- -- If included the server will send SessionResumptionUpdate messages. */ -- sessionResumption?: SessionResumptionConfig; -- /** The transcription of the input aligns with the input audio language. -- */ -- inputAudioTranscription?: AudioTranscriptionConfig; -- /** The transcription of the output aligns with the language code -- specified for the output audio. -- */ -- outputAudioTranscription?: AudioTranscriptionConfig; -- /** Configures the realtime input behavior in BidiGenerateContent. */ -- realtimeInputConfig?: RealtimeInputConfig; -- /** Configures context window compression mechanism. -- -- If included, server will compress context window to fit into given length. */ -- contextWindowCompression?: ContextWindowCompressionConfig; -- /** Configures the proactivity of the model. This allows the model to respond proactively to -- the input and to ignore irrelevant input. */ -- proactivity?: ProactivityConfig; --} -- --/** Parameters for connecting to the live API. */ --export declare interface LiveConnectParameters { -- /** ID of the model to use. For a list of models, see `Google models -- `_. */ -- model: string; -- /** callbacks */ -- callbacks: LiveCallbacks; -- /** Optional configuration parameters for the request. -- */ -- config?: LiveConnectConfig; --} -- --/** Config for LiveEphemeralParameters for Auth Token creation. */ --export declare interface LiveEphemeralParameters { -- /** ID of the model to configure in the ephemeral token for Live API. -- For a list of models, see `Gemini models -- `. */ -- model?: string; -- /** Configuration specific to Live API connections created using this token. */ -- config?: LiveConnectConfig; --} -- --/** -- LiveMusic class encapsulates the configuration for live music -- generation via Lyria Live models. -- -- @experimental -- */ --declare class LiveMusic { -- private readonly apiClient; -- private readonly auth; -- private readonly webSocketFactory; -- constructor(apiClient: ApiClient, auth: Auth, webSocketFactory: WebSocketFactory); -- /** -- Establishes a connection to the specified model and returns a -- LiveMusicSession object representing that connection. -- -- @experimental -- -- @remarks -- -- @param params - The parameters for establishing a connection to the model. -- @return A live session. -- -- @example -- ```ts -- let model = 'models/lyria-realtime-exp'; -- const session = await ai.live.music.connect({ -- model: model, -- callbacks: { -- onmessage: (e: MessageEvent) => { -- console.log('Received message from the server: %s\n', debug(e.data)); -- }, -- onerror: (e: ErrorEvent) => { -- console.log('Error occurred: %s\n', debug(e.error)); -- }, -- onclose: (e: CloseEvent) => { -- console.log('Connection closed.'); -- }, -- }, -- }); -- ``` -- */ -- connect(params: types.LiveMusicConnectParameters): Promise; --} -- --/** Callbacks for the realtime music API. */ --export declare interface LiveMusicCallbacks { -- /** -- * Called when a message is received from the server. -- */ -- onmessage: (e: LiveMusicServerMessage) => void; -- /** -- * Called when an error occurs. -- */ -- onerror?: ((e: ErrorEvent) => void) | null; -- /** -- * Called when the websocket connection is closed. -- */ -- onclose?: ((e: CloseEvent) => void) | null; --} -- --/** User input to start or steer the music. */ --export declare interface LiveMusicClientContent { -- /** Weighted prompts as the model input. */ -- weightedPrompts?: WeightedPrompt[]; --} -- --/** Messages sent by the client in the LiveMusicClientMessage call. */ --export declare interface LiveMusicClientMessage { -- /** Message to be sent in the first (and only in the first) `LiveMusicClientMessage`. -- Clients should wait for a `LiveMusicSetupComplete` message before -- sending any additional messages. */ -- setup?: LiveMusicClientSetup; -- /** User input to influence music generation. */ -- clientContent?: LiveMusicClientContent; -- /** Configuration for music generation. */ -- musicGenerationConfig?: LiveMusicGenerationConfig; -- /** Playback control signal for the music generation. */ -- playbackControl?: LiveMusicPlaybackControl; --} -- --/** Message to be sent by the system when connecting to the API. */ --export declare interface LiveMusicClientSetup { -- /** The model's resource name. Format: `models/{model}`. */ -- model?: string; --} -- --/** Parameters for connecting to the live API. */ --export declare interface LiveMusicConnectParameters { -- /** The model's resource name. */ -- model: string; -- /** Callbacks invoked on server events. */ -- callbacks: LiveMusicCallbacks; --} -- --/** A prompt that was filtered with the reason. */ --export declare interface LiveMusicFilteredPrompt { -- /** The text prompt that was filtered. */ -- text?: string; -- /** The reason the prompt was filtered. */ -- filteredReason?: string; --} -- --/** Configuration for music generation. */ --export declare interface LiveMusicGenerationConfig { -- /** Controls the variance in audio generation. Higher values produce -- higher variance. Range is [0.0, 3.0]. */ -- temperature?: number; -- /** Controls how the model selects tokens for output. Samples the topK -- tokens with the highest probabilities. Range is [1, 1000]. */ -- topK?: number; -- /** Seeds audio generation. If not set, the request uses a randomly -- generated seed. */ -- seed?: number; -- /** Controls how closely the model follows prompts. -- Higher guidance follows more closely, but will make transitions more -- abrupt. Range is [0.0, 6.0]. */ -- guidance?: number; -- /** Beats per minute. Range is [60, 200]. */ -- bpm?: number; -- /** Density of sounds. Range is [0.0, 1.0]. */ -- density?: number; -- /** Brightness of the music. Range is [0.0, 1.0]. */ -- brightness?: number; -- /** Scale of the generated music. */ -- scale?: Scale; -- /** Whether the audio output should contain bass. */ -- muteBass?: boolean; -- /** Whether the audio output should contain drums. */ -- muteDrums?: boolean; -- /** Whether the audio output should contain only bass and drums. */ -- onlyBassAndDrums?: boolean; -- /** The mode of music generation. Default mode is QUALITY. */ -- musicGenerationMode?: MusicGenerationMode; --} -- --/** The playback control signal to apply to the music generation. */ --export declare enum LiveMusicPlaybackControl { -- /** -- * This value is unused. -- */ -- PLAYBACK_CONTROL_UNSPECIFIED = "PLAYBACK_CONTROL_UNSPECIFIED", -- /** -- * Start generating the music. -- */ -- PLAY = "PLAY", -- /** -- * Hold the music generation. Use PLAY to resume from the current position. -- */ -- PAUSE = "PAUSE", -- /** -- * Stop the music generation and reset the context (prompts retained). -- Use PLAY to restart the music generation. -- */ -- STOP = "STOP", -- /** -- * Reset the context of the music generation without stopping it. -- Retains the current prompts and config. -- */ -- RESET_CONTEXT = "RESET_CONTEXT" --} -- --/** Server update generated by the model in response to client messages. -- -- Content is generated as quickly as possible, and not in real time. -- Clients may choose to buffer and play it out in real time. -- */ --export declare interface LiveMusicServerContent { -- /** The audio chunks that the model has generated. */ -- audioChunks?: AudioChunk[]; --} -- --/** Response message for the LiveMusicClientMessage call. */ --export declare class LiveMusicServerMessage { -- /** Message sent in response to a `LiveMusicClientSetup` message from the client. -- Clients should wait for this message before sending any additional messages. */ -- setupComplete?: LiveMusicServerSetupComplete; -- /** Content generated by the model in response to client messages. */ -- serverContent?: LiveMusicServerContent; -- /** A prompt that was filtered with the reason. */ -- filteredPrompt?: LiveMusicFilteredPrompt; -- /** -- * Returns the first audio chunk from the server content, if present. -- * -- * @remarks -- * If there are no audio chunks in the response, undefined will be returned. -- */ -- get audioChunk(): AudioChunk | undefined; --} -- --/** Sent in response to a `LiveMusicClientSetup` message from the client. */ --export declare interface LiveMusicServerSetupComplete { --} -- --/** -- Represents a connection to the API. -- -- @experimental -- */ --export declare class LiveMusicSession { -- readonly conn: WebSocket_2; -- private readonly apiClient; -- constructor(conn: WebSocket_2, apiClient: ApiClient); -- /** -- Sets inputs to steer music generation. Updates the session's current -- weighted prompts. -- -- @param params - Contains one property, `weightedPrompts`. -- -- - `weightedPrompts` to send to the model; weights are normalized to -- sum to 1.0. -- -- @experimental -- */ -- setWeightedPrompts(params: types.LiveMusicSetWeightedPromptsParameters): Promise; -- /** -- Sets a configuration to the model. Updates the session's current -- music generation config. -- -- @param params - Contains one property, `musicGenerationConfig`. -- -- - `musicGenerationConfig` to set in the model. Passing an empty or -- undefined config to the model will reset the config to defaults. -- -- @experimental -- */ -- setMusicGenerationConfig(params: types.LiveMusicSetConfigParameters): Promise; -- private sendPlaybackControl; -- /** -- * Start the music stream. -- * -- * @experimental -- */ -- play(): void; -- /** -- * Temporarily halt the music stream. Use `play` to resume from the current -- * position. -- * -- * @experimental -- */ -- pause(): void; -- /** -- * Stop the music stream and reset the state. Retains the current prompts -- * and config. -- * -- * @experimental -- */ -- stop(): void; -- /** -- * Resets the context of the music generation without stopping it. -- * Retains the current prompts and config. -- * -- * @experimental -- */ -- resetContext(): void; -- /** -- Terminates the WebSocket connection. -- -- @experimental -- */ -- close(): void; --} -- --/** Parameters for setting config for the live music API. */ --export declare interface LiveMusicSetConfigParameters { -- /** Configuration for music generation. */ -- musicGenerationConfig: LiveMusicGenerationConfig; --} -- --/** Parameters for setting weighted prompts for the live music API. */ --export declare interface LiveMusicSetWeightedPromptsParameters { -- /** A map of text prompts to weights to use for the generation request. */ -- weightedPrompts: WeightedPrompt[]; --} -- --/** Prompts and config used for generating this audio chunk. */ --export declare interface LiveMusicSourceMetadata { -- /** Weighted prompts for generating this audio chunk. */ -- clientContent?: LiveMusicClientContent; -- /** Music generation config for generating this audio chunk. */ -- musicGenerationConfig?: LiveMusicGenerationConfig; --} -- --/** Parameters for sending client content to the live API. */ --export declare interface LiveSendClientContentParameters { -- /** Client content to send to the session. */ -- turns?: ContentListUnion; -- /** If true, indicates that the server content generation should start with -- the currently accumulated prompt. Otherwise, the server will await -- additional messages before starting generation. */ -- turnComplete?: boolean; --} -- --/** Parameters for sending realtime input to the live API. */ --export declare interface LiveSendRealtimeInputParameters { -- /** Realtime input to send to the session. */ -- media?: BlobImageUnion; -- /** The realtime audio input stream. */ -- audio?: Blob_2; -- /** -- Indicates that the audio stream has ended, e.g. because the microphone was -- turned off. -- -- This should only be sent when automatic activity detection is enabled -- (which is the default). -- -- The client can reopen the stream by sending an audio message. -- */ -- audioStreamEnd?: boolean; -- /** The realtime video input stream. */ -- video?: BlobImageUnion; -- /** The realtime text input stream. */ -- text?: string; -- /** Marks the start of user activity. */ -- activityStart?: ActivityStart; -- /** Marks the end of user activity. */ -- activityEnd?: ActivityEnd; --} -- --/** Parameters for sending tool responses to the live API. */ --export declare class LiveSendToolResponseParameters { -- /** Tool responses to send to the session. */ -- functionResponses: FunctionResponse[] | FunctionResponse; --} -- --/** Incremental server update generated by the model in response to client messages. -- -- Content is generated as quickly as possible, and not in real time. Clients -- may choose to buffer and play it out in real time. -- */ --export declare interface LiveServerContent { -- /** The content that the model has generated as part of the current conversation with the user. */ -- modelTurn?: Content; -- /** If true, indicates that the model is done generating. Generation will only start in response to additional client messages. Can be set alongside `content`, indicating that the `content` is the last in the turn. */ -- turnComplete?: boolean; -- /** If true, indicates that a client message has interrupted current model generation. If the client is playing out the content in realtime, this is a good signal to stop and empty the current queue. */ -- interrupted?: boolean; -- /** Metadata returned to client when grounding is enabled. */ -- groundingMetadata?: GroundingMetadata; -- /** If true, indicates that the model is done generating. When model is -- interrupted while generating there will be no generation_complete message -- in interrupted turn, it will go through interrupted > turn_complete. -- When model assumes realtime playback there will be delay between -- generation_complete and turn_complete that is caused by model -- waiting for playback to finish. If true, indicates that the model -- has finished generating all content. This is a signal to the client -- that it can stop sending messages. */ -- generationComplete?: boolean; -- /** Input transcription. The transcription is independent to the model -- turn which means it doesn’t imply any ordering between transcription and -- model turn. */ -- inputTranscription?: Transcription; -- /** Output transcription. The transcription is independent to the model -- turn which means it doesn’t imply any ordering between transcription and -- model turn. -- */ -- outputTranscription?: Transcription; -- /** Metadata related to url context retrieval tool. */ -- urlContextMetadata?: UrlContextMetadata; --} -- --/** Server will not be able to service client soon. */ --export declare interface LiveServerGoAway { -- /** The remaining time before the connection will be terminated as ABORTED. The minimal time returned here is specified differently together with the rate limits for a given model. */ -- timeLeft?: string; --} -- --/** Response message for API call. */ --export declare class LiveServerMessage { -- /** Sent in response to a `LiveClientSetup` message from the client. */ -- setupComplete?: LiveServerSetupComplete; -- /** Content generated by the model in response to client messages. */ -- serverContent?: LiveServerContent; -- /** Request for the client to execute the `function_calls` and return the responses with the matching `id`s. */ -- toolCall?: LiveServerToolCall; -- /** Notification for the client that a previously issued `ToolCallMessage` with the specified `id`s should have been not executed and should be cancelled. */ -- toolCallCancellation?: LiveServerToolCallCancellation; -- /** Usage metadata about model response(s). */ -- usageMetadata?: UsageMetadata; -- /** Server will disconnect soon. */ -- goAway?: LiveServerGoAway; -- /** Update of the session resumption state. */ -- sessionResumptionUpdate?: LiveServerSessionResumptionUpdate; -- /** -- * Returns the concatenation of all text parts from the server content if present. -- * -- * @remarks -- * If there are non-text parts in the response, the concatenation of all text -- * parts will be returned, and a warning will be logged. -- */ -- get text(): string | undefined; -- /** -- * Returns the concatenation of all inline data parts from the server content if present. -- * -- * @remarks -- * If there are non-inline data parts in the -- * response, the concatenation of all inline data parts will be returned, and -- * a warning will be logged. -- */ -- get data(): string | undefined; --} -- --/** Update of the session resumption state. -- -- Only sent if `session_resumption` was set in the connection config. -- */ --export declare interface LiveServerSessionResumptionUpdate { -- /** New handle that represents state that can be resumed. Empty if `resumable`=false. */ -- newHandle?: string; -- /** True if session can be resumed at this point. It might be not possible to resume session at some points. In that case we send update empty new_handle and resumable=false. Example of such case could be model executing function calls or just generating. Resuming session (using previous session token) in such state will result in some data loss. */ -- resumable?: boolean; -- /** Index of last message sent by client that is included in state represented by this SessionResumptionToken. Only sent when `SessionResumptionConfig.transparent` is set. -- -- Presence of this index allows users to transparently reconnect and avoid issue of losing some part of realtime audio input/video. If client wishes to temporarily disconnect (for example as result of receiving GoAway) they can do it without losing state by buffering messages sent since last `SessionResmumptionTokenUpdate`. This field will enable them to limit buffering (avoid keeping all requests in RAM). -- -- Note: This should not be used for when resuming a session at some time later -- in those cases partial audio and video frames arelikely not needed. */ -- lastConsumedClientMessageIndex?: string; --} -- --export declare interface LiveServerSetupComplete { --} -- --/** Request for the client to execute the `function_calls` and return the responses with the matching `id`s. */ --export declare interface LiveServerToolCall { -- /** The function call to be executed. */ -- functionCalls?: FunctionCall[]; --} -- --/** Notification for the client that a previously issued `ToolCallMessage` with the specified `id`s should have been not executed and should be cancelled. -- -- If there were side-effects to those tool calls, clients may attempt to undo -- the tool calls. This message occurs only in cases where the clients interrupt -- server turns. -- */ --export declare interface LiveServerToolCallCancellation { -- /** The ids of the tool calls to be cancelled. */ -- ids?: string[]; --} -- --/** Logprobs Result */ --export declare interface LogprobsResult { -- /** Length = total number of decoding steps. The chosen candidates may or may not be in top_candidates. */ -- chosenCandidates?: LogprobsResultCandidate[]; -- /** Length = total number of decoding steps. */ -- topCandidates?: LogprobsResultTopCandidates[]; --} -- --/** Candidate for the logprobs token and score. */ --export declare interface LogprobsResultCandidate { -- /** The candidate's log probability. */ -- logProbability?: number; -- /** The candidate's token string value. */ -- token?: string; -- /** The candidate's token id value. */ -- tokenId?: number; --} -- --/** Candidates with top log probabilities at each decoding step. */ --export declare interface LogprobsResultTopCandidates { -- /** Sorted by log probability in descending order. */ -- candidates?: LogprobsResultCandidate[]; --} -- --/** Configuration for a Mask reference image. */ --export declare interface MaskReferenceConfig { -- /** Prompts the model to generate a mask instead of you needing to -- provide one (unless MASK_MODE_USER_PROVIDED is used). */ -- maskMode?: MaskReferenceMode; -- /** A list of up to 5 class ids to use for semantic segmentation. -- Automatically creates an image mask based on specific objects. */ -- segmentationClasses?: number[]; -- /** Dilation percentage of the mask provided. -- Float between 0 and 1. */ -- maskDilation?: number; --} -- --/** A mask reference image. -- -- This encapsulates either a mask image provided by the user and configs for -- the user provided mask, or only config parameters for the model to generate -- a mask. -- -- A mask image is an image whose non-zero values indicate where to edit the base -- image. If the user provides a mask image, the mask must be in the same -- dimensions as the raw image. -- */ --export declare class MaskReferenceImage { -- /** The reference image for the editing operation. */ -- referenceImage?: Image_2; -- /** The id of the reference image. */ -- referenceId?: number; -- /** The type of the reference image. Only set by the SDK. */ -- referenceType?: string; -- /** Configuration for the mask reference image. */ -- config?: MaskReferenceConfig; -- /** Internal method to convert to ReferenceImageAPIInternal. */ -- toReferenceImageAPI(): any; --} -- --/** Enum representing the mask mode of a mask reference image. */ --export declare enum MaskReferenceMode { -- MASK_MODE_DEFAULT = "MASK_MODE_DEFAULT", -- MASK_MODE_USER_PROVIDED = "MASK_MODE_USER_PROVIDED", -- MASK_MODE_BACKGROUND = "MASK_MODE_BACKGROUND", -- MASK_MODE_FOREGROUND = "MASK_MODE_FOREGROUND", -- MASK_MODE_SEMANTIC = "MASK_MODE_SEMANTIC" --} -- --/** -- * Creates a McpCallableTool from MCP clients and an optional config. -- * -- * The callable tool can invoke the MCP clients with given function call -- * arguments. (often for automatic function calling). -- * Use the config to modify tool parameters such as behavior. -- * -- * @experimental Built-in MCP support is an experimental feature, may change in future -- * versions. -- */ --export declare function mcpToTool(...args: [...Client[], CallableToolConfig | Client]): CallableTool; -- --/** Server content modalities. */ --export declare enum MediaModality { -- /** -- * The modality is unspecified. -- */ -- MODALITY_UNSPECIFIED = "MODALITY_UNSPECIFIED", -- /** -- * Plain text. -- */ -- TEXT = "TEXT", -- /** -- * Images. -- */ -- IMAGE = "IMAGE", -- /** -- * Video. -- */ -- VIDEO = "VIDEO", -- /** -- * Audio. -- */ -- AUDIO = "AUDIO", -- /** -- * Document, e.g. PDF. -- */ -- DOCUMENT = "DOCUMENT" --} -- --/** The media resolution to use. */ --export declare enum MediaResolution { -- /** -- * Media resolution has not been set -- */ -- MEDIA_RESOLUTION_UNSPECIFIED = "MEDIA_RESOLUTION_UNSPECIFIED", -- /** -- * Media resolution set to low (64 tokens). -- */ -- MEDIA_RESOLUTION_LOW = "MEDIA_RESOLUTION_LOW", -- /** -- * Media resolution set to medium (256 tokens). -- */ -- MEDIA_RESOLUTION_MEDIUM = "MEDIA_RESOLUTION_MEDIUM", -- /** -- * Media resolution set to high (zoomed reframing with 256 tokens). -- */ -- MEDIA_RESOLUTION_HIGH = "MEDIA_RESOLUTION_HIGH" --} -- --/** Server content modalities. */ --export declare enum Modality { -- /** -- * The modality is unspecified. -- */ -- MODALITY_UNSPECIFIED = "MODALITY_UNSPECIFIED", -- /** -- * Indicates the model should return text -- */ -- TEXT = "TEXT", -- /** -- * Indicates the model should return images. -- */ -- IMAGE = "IMAGE", -- /** -- * Indicates the model should return images. -- */ -- AUDIO = "AUDIO" --} -- --/** Represents token counting info for a single modality. */ --export declare interface ModalityTokenCount { -- /** The modality associated with this token count. */ -- modality?: MediaModality; -- /** Number of tokens. */ -- tokenCount?: number; --} -- --/** The mode of the predictor to be used in dynamic retrieval. */ --export declare enum Mode { -- /** -- * Always trigger retrieval. -- */ -- MODE_UNSPECIFIED = "MODE_UNSPECIFIED", -- /** -- * Run retrieval only when system decides it is necessary. -- */ -- MODE_DYNAMIC = "MODE_DYNAMIC" --} -- --/** A trained machine learning model. */ --export declare interface Model { -- /** Resource name of the model. */ -- name?: string; -- /** Display name of the model. */ -- displayName?: string; -- /** Description of the model. */ -- description?: string; -- /** Version ID of the model. A new version is committed when a new -- model version is uploaded or trained under an existing model ID. The -- version ID is an auto-incrementing decimal number in string -- representation. */ -- version?: string; -- /** List of deployed models created from this base model. Note that a -- model could have been deployed to endpoints in different locations. */ -- endpoints?: Endpoint[]; -- /** Labels with user-defined metadata to organize your models. */ -- labels?: Record; -- /** Information about the tuned model from the base model. */ -- tunedModelInfo?: TunedModelInfo; -- /** The maximum number of input tokens that the model can handle. */ -- inputTokenLimit?: number; -- /** The maximum number of output tokens that the model can generate. */ -- outputTokenLimit?: number; -- /** List of actions that are supported by the model. */ -- supportedActions?: string[]; -- /** The default checkpoint id of a model version. -- */ -- defaultCheckpointId?: string; -- /** The checkpoints of the model. */ -- checkpoints?: Checkpoint[]; --} -- --export declare class Models extends BaseModule { -- private readonly apiClient; -- constructor(apiClient: ApiClient); -- /** -- * Makes an API request to generate content with a given model. -- * -- * For the `model` parameter, supported formats for Vertex AI API include: -- * - The Gemini model ID, for example: 'gemini-2.0-flash' -- * - The full resource name starts with 'projects/', for example: -- * 'projects/my-project-id/locations/us-central1/publishers/google/models/gemini-2.0-flash' -- * - The partial resource name with 'publishers/', for example: -- * 'publishers/google/models/gemini-2.0-flash' or -- * 'publishers/meta/models/llama-3.1-405b-instruct-maas' -- * - `/` separated publisher and model name, for example: -- * 'google/gemini-2.0-flash' or 'meta/llama-3.1-405b-instruct-maas' -- * -- * For the `model` parameter, supported formats for Gemini API include: -- * - The Gemini model ID, for example: 'gemini-2.0-flash' -- * - The model name starts with 'models/', for example: -- * 'models/gemini-2.0-flash' -- * - For tuned models, the model name starts with 'tunedModels/', -- * for example: -- * 'tunedModels/1234567890123456789' -- * -- * Some models support multimodal input and output. -- * -- * @param params - The parameters for generating content. -- * @return The response from generating content. -- * -- * @example -- * ```ts -- * const response = await ai.models.generateContent({ -- * model: 'gemini-2.0-flash', -- * contents: 'why is the sky blue?', -- * config: { -- * candidateCount: 2, -- * } -- * }); -- * console.log(response); -- * ``` -- */ -- generateContent: (params: types.GenerateContentParameters) => Promise; -- /** -- * Makes an API request to generate content with a given model and yields the -- * response in chunks. -- * -- * For the `model` parameter, supported formats for Vertex AI API include: -- * - The Gemini model ID, for example: 'gemini-2.0-flash' -- * - The full resource name starts with 'projects/', for example: -- * 'projects/my-project-id/locations/us-central1/publishers/google/models/gemini-2.0-flash' -- * - The partial resource name with 'publishers/', for example: -- * 'publishers/google/models/gemini-2.0-flash' or -- * 'publishers/meta/models/llama-3.1-405b-instruct-maas' -- * - `/` separated publisher and model name, for example: -- * 'google/gemini-2.0-flash' or 'meta/llama-3.1-405b-instruct-maas' -- * -- * For the `model` parameter, supported formats for Gemini API include: -- * - The Gemini model ID, for example: 'gemini-2.0-flash' -- * - The model name starts with 'models/', for example: -- * 'models/gemini-2.0-flash' -- * - For tuned models, the model name starts with 'tunedModels/', -- * for example: -- * 'tunedModels/1234567890123456789' -- * -- * Some models support multimodal input and output. -- * -- * @param params - The parameters for generating content with streaming response. -- * @return The response from generating content. -- * -- * @example -- * ```ts -- * const response = await ai.models.generateContentStream({ -- * model: 'gemini-2.0-flash', -- * contents: 'why is the sky blue?', -- * config: { -- * maxOutputTokens: 200, -- * } -- * }); -- * for await (const chunk of response) { -- * console.log(chunk); -- * } -- * ``` -- */ -- generateContentStream: (params: types.GenerateContentParameters) => Promise>; -- /** -- * Transforms the CallableTools in the parameters to be simply Tools, it -- * copies the params into a new object and replaces the tools, it does not -- * modify the original params. Also sets the MCP usage header if there are -- * MCP tools in the parameters. -- */ -- private processParamsForMcpUsage; -- private initAfcToolsMap; -- private processAfcStream; -- /** -- * Generates an image based on a text description and configuration. -- * -- * @param params - The parameters for generating images. -- * @return The response from the API. -- * -- * @example -- * ```ts -- * const response = await client.models.generateImages({ -- * model: 'imagen-3.0-generate-002', -- * prompt: 'Robot holding a red skateboard', -- * config: { -- * numberOfImages: 1, -- * includeRaiReason: true, -- * }, -- * }); -- * console.log(response?.generatedImages?.[0]?.image?.imageBytes); -- * ``` -- */ -- generateImages: (params: types.GenerateImagesParameters) => Promise; -- list: (params?: types.ListModelsParameters) => Promise>; -- /** -- * Edits an image based on a prompt, list of reference images, and configuration. -- * -- * @param params - The parameters for editing an image. -- * @return The response from the API. -- * -- * @example -- * ```ts -- * const response = await client.models.editImage({ -- * model: 'imagen-3.0-capability-001', -- * prompt: 'Generate an image containing a mug with the product logo [1] visible on the side of the mug.', -- * referenceImages: [subjectReferenceImage] -- * config: { -- * numberOfImages: 1, -- * includeRaiReason: true, -- * }, -- * }); -- * console.log(response?.generatedImages?.[0]?.image?.imageBytes); -- * ``` -- */ -- editImage: (params: types.EditImageParameters) => Promise; -- /** -- * Upscales an image based on an image, upscale factor, and configuration. -- * Only supported in Vertex AI currently. -- * -- * @param params - The parameters for upscaling an image. -- * @return The response from the API. -- * -- * @example -- * ```ts -- * const response = await client.models.upscaleImage({ -- * model: 'imagen-3.0-generate-002', -- * image: image, -- * upscaleFactor: 'x2', -- * config: { -- * includeRaiReason: true, -- * }, -- * }); -- * console.log(response?.generatedImages?.[0]?.image?.imageBytes); -- * ``` -- */ -- upscaleImage: (params: types.UpscaleImageParameters) => Promise; -- private generateContentInternal; -- private generateContentStreamInternal; -- /** -- * Calculates embeddings for the given contents. Only text is supported. -- * -- * @param params - The parameters for embedding contents. -- * @return The response from the API. -- * -- * @example -- * ```ts -- * const response = await ai.models.embedContent({ -- * model: 'text-embedding-004', -- * contents: [ -- * 'What is your name?', -- * 'What is your favorite color?', -- * ], -- * config: { -- * outputDimensionality: 64, -- * }, -- * }); -- * console.log(response); -- * ``` -- */ -- embedContent(params: types.EmbedContentParameters): Promise; -- /** -- * Generates an image based on a text description and configuration. -- * -- * @param params - The parameters for generating images. -- * @return The response from the API. -- * -- * @example -- * ```ts -- * const response = await ai.models.generateImages({ -- * model: 'imagen-3.0-generate-002', -- * prompt: 'Robot holding a red skateboard', -- * config: { -- * numberOfImages: 1, -- * includeRaiReason: true, -- * }, -- * }); -- * console.log(response?.generatedImages?.[0]?.image?.imageBytes); -- * ``` -- */ -- private generateImagesInternal; -- private editImageInternal; -- private upscaleImageInternal; -- /** -- * Fetches information about a model by name. -- * -- * @example -- * ```ts -- * const modelInfo = await ai.models.get({model: 'gemini-2.0-flash'}); -- * ``` -- */ -- get(params: types.GetModelParameters): Promise; -- private listInternal; -- /** -- * Updates a tuned model by its name. -- * -- * @param params - The parameters for updating the model. -- * @return The response from the API. -- * -- * @example -- * ```ts -- * const response = await ai.models.update({ -- * model: 'tuned-model-name', -- * config: { -- * displayName: 'New display name', -- * description: 'New description', -- * }, -- * }); -- * ``` -- */ -- update(params: types.UpdateModelParameters): Promise; -- /** -- * Deletes a tuned model by its name. -- * -- * @param params - The parameters for deleting the model. -- * @return The response from the API. -- * -- * @example -- * ```ts -- * const response = await ai.models.delete({model: 'tuned-model-name'}); -- * ``` -- */ -- delete(params: types.DeleteModelParameters): Promise; -- /** -- * Counts the number of tokens in the given contents. Multimodal input is -- * supported for Gemini models. -- * -- * @param params - The parameters for counting tokens. -- * @return The response from the API. -- * -- * @example -- * ```ts -- * const response = await ai.models.countTokens({ -- * model: 'gemini-2.0-flash', -- * contents: 'The quick brown fox jumps over the lazy dog.' -- * }); -- * console.log(response); -- * ``` -- */ -- countTokens(params: types.CountTokensParameters): Promise; -- /** -- * Given a list of contents, returns a corresponding TokensInfo containing -- * the list of tokens and list of token ids. -- * -- * This method is not supported by the Gemini Developer API. -- * -- * @param params - The parameters for computing tokens. -- * @return The response from the API. -- * -- * @example -- * ```ts -- * const response = await ai.models.computeTokens({ -- * model: 'gemini-2.0-flash', -- * contents: 'What is your name?' -- * }); -- * console.log(response); -- * ``` -- */ -- computeTokens(params: types.ComputeTokensParameters): Promise; -- /** -- * Generates videos based on a text description and configuration. -- * -- * @param params - The parameters for generating videos. -- * @return A Promise which allows you to track the progress and eventually retrieve the generated videos using the operations.get method. -- * -- * @example -- * ```ts -- * const operation = await ai.models.generateVideos({ -- * model: 'veo-2.0-generate-001', -- * prompt: 'A neon hologram of a cat driving at top speed', -- * config: { -- * numberOfVideos: 1 -- * }); -- * -- * while (!operation.done) { -- * await new Promise(resolve => setTimeout(resolve, 10000)); -- * operation = await ai.operations.getVideosOperation({operation: operation}); -- * } -- * -- * console.log(operation.response?.generatedVideos?.[0]?.video?.uri); -- * ``` -- */ -- generateVideos(params: types.GenerateVideosParameters): Promise; --} -- --/** Config for model selection. */ --export declare interface ModelSelectionConfig { -- /** Options for feature selection preference. */ -- featureSelectionPreference?: FeatureSelectionPreference; --} -- --/** The configuration for the multi-speaker setup. */ --export declare interface MultiSpeakerVoiceConfig { -- /** The configuration for the speaker to use. */ -- speakerVoiceConfigs?: SpeakerVoiceConfig[]; --} -- --/** The mode of music generation. */ --export declare enum MusicGenerationMode { -- /** -- * This value is unused. -- */ -- MUSIC_GENERATION_MODE_UNSPECIFIED = "MUSIC_GENERATION_MODE_UNSPECIFIED", -- /** -- * Steer text prompts to regions of latent space with higher quality -- music. -- */ -- QUALITY = "QUALITY", -- /** -- * Steer text prompts to regions of latent space with a larger diversity -- of music. -- */ -- DIVERSITY = "DIVERSITY" --} -- --/** A long-running operation. */ --export declare interface Operation { -- /** The server-assigned name, which is only unique within the same service that originally returns it. If you use the default HTTP mapping, the `name` should be a resource name ending with `operations/{unique_id}`. */ -- name?: string; -- /** Service-specific metadata associated with the operation. It typically contains progress information and common metadata such as create time. Some services might not provide such metadata. Any method that returns a long-running operation should document the metadata type, if any. */ -- metadata?: Record; -- /** If the value is `false`, it means the operation is still in progress. If `true`, the operation is completed, and either `error` or `response` is available. */ -- done?: boolean; -- /** The error result of the operation in case of failure or cancellation. */ -- error?: Record; --} -- --/** Parameters for the get method of the operations module. */ --export declare interface OperationGetParameters { -- /** The operation to be retrieved. */ -- operation: GenerateVideosOperation; -- /** Used to override the default configuration. */ -- config?: GetOperationConfig; --} -- --export declare class Operations extends BaseModule { -- private readonly apiClient; -- constructor(apiClient: ApiClient); -- /** -- * Gets the status of a long-running operation. -- * -- * @param parameters The parameters for the get operation request. -- * @return The updated Operation object, with the latest status or result. -- */ -- getVideosOperation(parameters: types.OperationGetParameters): Promise; -- private getVideosOperationInternal; -- private fetchPredictVideosOperationInternal; --} -- --/** -- * @license -- * Copyright 2025 Google LLC -- * SPDX-License-Identifier: Apache-2.0 -- */ --/** Required. Outcome of the code execution. */ --export declare enum Outcome { -- /** -- * Unspecified status. This value should not be used. -- */ -- OUTCOME_UNSPECIFIED = "OUTCOME_UNSPECIFIED", -- /** -- * Code execution completed successfully. -- */ -- OUTCOME_OK = "OUTCOME_OK", -- /** -- * Code execution finished but with a failure. `stderr` should contain the reason. -- */ -- OUTCOME_FAILED = "OUTCOME_FAILED", -- /** -- * Code execution ran for too long, and was cancelled. There may or may not be a partial output present. -- */ -- OUTCOME_DEADLINE_EXCEEDED = "OUTCOME_DEADLINE_EXCEEDED" --} -- --/** -- * @license -- * Copyright 2025 Google LLC -- * SPDX-License-Identifier: Apache-2.0 -- */ --/** -- * Pagers for the GenAI List APIs. -- */ --export declare enum PagedItem { -- PAGED_ITEM_BATCH_JOBS = "batchJobs", -- PAGED_ITEM_MODELS = "models", -- PAGED_ITEM_TUNING_JOBS = "tuningJobs", -- PAGED_ITEM_FILES = "files", -- PAGED_ITEM_CACHED_CONTENTS = "cachedContents" --} -- --declare interface PagedItemConfig { -- config?: { -- pageToken?: string; -- pageSize?: number; -- }; --} -- --declare interface PagedItemResponse { -- nextPageToken?: string; -- batchJobs?: T[]; -- models?: T[]; -- tuningJobs?: T[]; -- files?: T[]; -- cachedContents?: T[]; --} -- --/** -- * Pager class for iterating through paginated results. -- */ --export declare class Pager implements AsyncIterable { -- private nameInternal; -- private pageInternal; -- private paramsInternal; -- private pageInternalSize; -- protected requestInternal: (params: PagedItemConfig) => Promise>; -- protected idxInternal: number; -- constructor(name: PagedItem, request: (params: PagedItemConfig) => Promise>, response: PagedItemResponse, params: PagedItemConfig); -- private init; -- private initNextPage; -- /** -- * Returns the current page, which is a list of items. -- * -- * @remarks -- * The first page is retrieved when the pager is created. The returned list of -- * items could be a subset of the entire list. -- */ -- get page(): T[]; -- /** -- * Returns the type of paged item (for example, ``batch_jobs``). -- */ -- get name(): PagedItem; -- /** -- * Returns the length of the page fetched each time by this pager. -- * -- * @remarks -- * The number of items in the page is less than or equal to the page length. -- */ -- get pageSize(): number; -- /** -- * Returns the parameters when making the API request for the next page. -- * -- * @remarks -- * Parameters contain a set of optional configs that can be -- * used to customize the API request. For example, the `pageToken` parameter -- * contains the token to request the next page. -- */ -- get params(): PagedItemConfig; -- /** -- * Returns the total number of items in the current page. -- */ -- get pageLength(): number; -- /** -- * Returns the item at the given index. -- */ -- getItem(index: number): T; -- /** -- * Returns an async iterator that support iterating through all items -- * retrieved from the API. -- * -- * @remarks -- * The iterator will automatically fetch the next page if there are more items -- * to fetch from the API. -- * -- * @example -- * -- * ```ts -- * const pager = await ai.files.list({config: {pageSize: 10}}); -- * for await (const file of pager) { -- * console.log(file.name); -- * } -- * ``` -- */ -- [Symbol.asyncIterator](): AsyncIterator; -- /** -- * Fetches the next page of items. This makes a new API request. -- * -- * @throws {Error} If there are no more pages to fetch. -- * -- * @example -- * -- * ```ts -- * const pager = await ai.files.list({config: {pageSize: 10}}); -- * let page = pager.page; -- * while (true) { -- * for (const file of page) { -- * console.log(file.name); -- * } -- * if (!pager.hasNextPage()) { -- * break; -- * } -- * page = await pager.nextPage(); -- * } -- * ``` -- */ -- nextPage(): Promise; -- /** -- * Returns true if there are more pages to fetch from the API. -- */ -- hasNextPage(): boolean; --} -- --/** A datatype containing media content. -- -- Exactly one field within a Part should be set, representing the specific type -- of content being conveyed. Using multiple fields within the same `Part` -- instance is considered invalid. -- */ --export declare interface Part { -- /** Metadata for a given video. */ -- videoMetadata?: VideoMetadata; -- /** Indicates if the part is thought from the model. */ -- thought?: boolean; -- /** Optional. Inlined bytes data. */ -- inlineData?: Blob_2; -- /** Optional. Result of executing the [ExecutableCode]. */ -- codeExecutionResult?: CodeExecutionResult; -- /** Optional. Code generated by the model that is meant to be executed. */ -- executableCode?: ExecutableCode; -- /** Optional. URI based data. */ -- fileData?: FileData; -- /** Optional. A predicted [FunctionCall] returned from the model that contains a string representing the [FunctionDeclaration.name] with the parameters and their values. */ -- functionCall?: FunctionCall; -- /** Optional. The result output of a [FunctionCall] that contains a string representing the [FunctionDeclaration.name] and a structured JSON object containing any output from the function call. It is used as context to the model. */ -- functionResponse?: FunctionResponse; -- /** Optional. Text part (can be code). */ -- text?: string; --} -- --export declare type PartListUnion = PartUnion[] | PartUnion; -- --/** Tuning spec for Partner models. */ --export declare interface PartnerModelTuningSpec { -- /** Hyperparameters for tuning. The accepted hyper_parameters and their valid range of values will differ depending on the base model. */ -- hyperParameters?: Record; -- /** Required. Cloud Storage path to file containing training dataset for tuning. The dataset must be formatted as a JSONL file. */ -- trainingDatasetUri?: string; -- /** Optional. Cloud Storage path to file containing validation dataset for tuning. The dataset must be formatted as a JSONL file. */ -- validationDatasetUri?: string; --} -- --export declare type PartUnion = Part | string; -- --/** Enum that controls the generation of people. */ --export declare enum PersonGeneration { -- DONT_ALLOW = "DONT_ALLOW", -- ALLOW_ADULT = "ALLOW_ADULT", -- ALLOW_ALL = "ALLOW_ALL" --} -- --/** The configuration for the prebuilt speaker to use. */ --export declare interface PrebuiltVoiceConfig { -- /** The name of the prebuilt voice to use. */ -- voiceName?: string; --} -- --/** Config for proactivity features. */ --export declare interface ProactivityConfig { -- /** If enabled, the model can reject responding to the last prompt. For -- example, this allows the model to ignore out of context speech or to stay -- silent if the user did not make a request, yet. */ -- proactiveAudio?: boolean; --} -- --/** Specifies the context retrieval config. */ --export declare interface RagRetrievalConfig { -- /** Optional. Config for filters. */ -- filter?: RagRetrievalConfigFilter; -- /** Optional. Config for Hybrid Search. */ -- hybridSearch?: RagRetrievalConfigHybridSearch; -- /** Optional. Config for ranking and reranking. */ -- ranking?: RagRetrievalConfigRanking; -- /** Optional. The number of contexts to retrieve. */ -- topK?: number; --} -- --/** Config for filters. */ --export declare interface RagRetrievalConfigFilter { -- /** Optional. String for metadata filtering. */ -- metadataFilter?: string; -- /** Optional. Only returns contexts with vector distance smaller than the threshold. */ -- vectorDistanceThreshold?: number; -- /** Optional. Only returns contexts with vector similarity larger than the threshold. */ -- vectorSimilarityThreshold?: number; --} -- --/** Config for Hybrid Search. */ --export declare interface RagRetrievalConfigHybridSearch { -- /** Optional. Alpha value controls the weight between dense and sparse vector search results. The range is [0, 1], while 0 means sparse vector search only and 1 means dense vector search only. The default value is 0.5 which balances sparse and dense vector search equally. */ -- alpha?: number; --} -- --/** Config for ranking and reranking. */ --export declare interface RagRetrievalConfigRanking { -- /** Optional. Config for LlmRanker. */ -- llmRanker?: RagRetrievalConfigRankingLlmRanker; -- /** Optional. Config for Rank Service. */ -- rankService?: RagRetrievalConfigRankingRankService; --} -- --/** Config for LlmRanker. */ --export declare interface RagRetrievalConfigRankingLlmRanker { -- /** Optional. The model name used for ranking. Format: `gemini-1.5-pro` */ -- modelName?: string; --} -- --/** Config for Rank Service. */ --export declare interface RagRetrievalConfigRankingRankService { -- /** Optional. The model name of the rank service. Format: `semantic-ranker-512@latest` */ -- modelName?: string; --} -- --/** A raw reference image. -- -- A raw reference image represents the base image to edit, provided by the user. -- It can optionally be provided in addition to a mask reference image or -- a style reference image. -- */ --export declare class RawReferenceImage { -- /** The reference image for the editing operation. */ -- referenceImage?: Image_2; -- /** The id of the reference image. */ -- referenceId?: number; -- /** The type of the reference image. Only set by the SDK. */ -- referenceType?: string; -- /** Internal method to convert to ReferenceImageAPIInternal. */ -- toReferenceImageAPI(): any; --} -- --/** Marks the end of user activity. -- -- This can only be sent if automatic (i.e. server-side) activity detection is -- disabled. -- */ --export declare interface RealtimeInputConfig { -- /** If not set, automatic activity detection is enabled by default. If automatic voice detection is disabled, the client must send activity signals. */ -- automaticActivityDetection?: AutomaticActivityDetection; -- /** Defines what effect activity has. */ -- activityHandling?: ActivityHandling; -- /** Defines which input is included in the user's turn. */ -- turnCoverage?: TurnCoverage; --} -- --export declare type ReferenceImage = RawReferenceImage | MaskReferenceImage | ControlReferenceImage | StyleReferenceImage | SubjectReferenceImage; -- --/** Represents a recorded session. */ --export declare interface ReplayFile { -- replayId?: string; -- interactions?: ReplayInteraction[]; --} -- --/** Represents a single interaction, request and response in a replay. */ --export declare interface ReplayInteraction { -- request?: ReplayRequest; -- response?: ReplayResponse; --} -- --/** Represents a single request in a replay. */ --export declare interface ReplayRequest { -- method?: string; -- url?: string; -- headers?: Record; -- bodySegments?: Record[]; --} -- --/** Represents a single response in a replay. */ --export declare class ReplayResponse { -- statusCode?: number; -- headers?: Record; -- bodySegments?: Record[]; -- sdkResponseSegments?: Record[]; --} -- --/** Defines a retrieval tool that model can call to access external knowledge. */ --export declare interface Retrieval { -- /** Optional. Deprecated. This option is no longer supported. */ -- disableAttribution?: boolean; -- /** Set to use data source powered by Vertex AI Search. */ -- vertexAiSearch?: VertexAISearch; -- /** Set to use data source powered by Vertex RAG store. User data is uploaded via the VertexRagDataService. */ -- vertexRagStore?: VertexRagStore; --} -- --/** Retrieval config. -- */ --export declare interface RetrievalConfig { -- /** Optional. The location of the user. */ -- latLng?: LatLng; --} -- --/** Metadata related to retrieval in the grounding flow. */ --export declare interface RetrievalMetadata { -- /** Optional. Score indicating how likely information from Google Search could help answer the prompt. The score is in the range `[0, 1]`, where 0 is the least likely and 1 is the most likely. This score is only populated when Google Search grounding and dynamic retrieval is enabled. It will be compared to the threshold to determine whether to trigger Google Search. */ -- googleSearchDynamicRetrievalScore?: number; --} -- --/** Safety attributes of a GeneratedImage or the user-provided prompt. */ --export declare interface SafetyAttributes { -- /** List of RAI categories. -- */ -- categories?: string[]; -- /** List of scores of each categories. -- */ -- scores?: number[]; -- /** Internal use only. -- */ -- contentType?: string; --} -- --/** Enum that controls the safety filter level for objectionable content. */ --export declare enum SafetyFilterLevel { -- BLOCK_LOW_AND_ABOVE = "BLOCK_LOW_AND_ABOVE", -- BLOCK_MEDIUM_AND_ABOVE = "BLOCK_MEDIUM_AND_ABOVE", -- BLOCK_ONLY_HIGH = "BLOCK_ONLY_HIGH", -- BLOCK_NONE = "BLOCK_NONE" --} -- --/** Safety rating corresponding to the generated content. */ --export declare interface SafetyRating { -- /** Output only. Indicates whether the content was filtered out because of this rating. */ -- blocked?: boolean; -- /** Output only. Harm category. */ -- category?: HarmCategory; -- /** Output only. Harm probability levels in the content. */ -- probability?: HarmProbability; -- /** Output only. Harm probability score. */ -- probabilityScore?: number; -- /** Output only. Harm severity levels in the content. */ -- severity?: HarmSeverity; -- /** Output only. Harm severity score. */ -- severityScore?: number; --} -- --/** Safety settings. */ --export declare interface SafetySetting { -- /** Determines if the harm block method uses probability or probability -- and severity scores. */ -- method?: HarmBlockMethod; -- /** Required. Harm category. */ -- category?: HarmCategory; -- /** Required. The harm block threshold. */ -- threshold?: HarmBlockThreshold; --} -- --/** Scale of the generated music. */ --export declare enum Scale { -- /** -- * Default value. This value is unused. -- */ -- SCALE_UNSPECIFIED = "SCALE_UNSPECIFIED", -- /** -- * C major or A minor. -- */ -- C_MAJOR_A_MINOR = "C_MAJOR_A_MINOR", -- /** -- * Db major or Bb minor. -- */ -- D_FLAT_MAJOR_B_FLAT_MINOR = "D_FLAT_MAJOR_B_FLAT_MINOR", -- /** -- * D major or B minor. -- */ -- D_MAJOR_B_MINOR = "D_MAJOR_B_MINOR", -- /** -- * Eb major or C minor -- */ -- E_FLAT_MAJOR_C_MINOR = "E_FLAT_MAJOR_C_MINOR", -- /** -- * E major or Db minor. -- */ -- E_MAJOR_D_FLAT_MINOR = "E_MAJOR_D_FLAT_MINOR", -- /** -- * F major or D minor. -- */ -- F_MAJOR_D_MINOR = "F_MAJOR_D_MINOR", -- /** -- * Gb major or Eb minor. -- */ -- G_FLAT_MAJOR_E_FLAT_MINOR = "G_FLAT_MAJOR_E_FLAT_MINOR", -- /** -- * G major or E minor. -- */ -- G_MAJOR_E_MINOR = "G_MAJOR_E_MINOR", -- /** -- * Ab major or F minor. -- */ -- A_FLAT_MAJOR_F_MINOR = "A_FLAT_MAJOR_F_MINOR", -- /** -- * A major or Gb minor. -- */ -- A_MAJOR_G_FLAT_MINOR = "A_MAJOR_G_FLAT_MINOR", -- /** -- * Bb major or G minor. -- */ -- B_FLAT_MAJOR_G_MINOR = "B_FLAT_MAJOR_G_MINOR", -- /** -- * B major or Ab minor. -- */ -- B_MAJOR_A_FLAT_MINOR = "B_MAJOR_A_FLAT_MINOR" --} -- --/** Schema is used to define the format of input/output data. Represents a select subset of an [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema-object). More fields may be added in the future as needed. */ --export declare interface Schema { -- /** Optional. The value should be validated against any (one or more) of the subschemas in the list. */ -- anyOf?: Schema[]; -- /** Optional. Default value of the data. */ -- default?: unknown; -- /** Optional. The description of the data. */ -- description?: string; -- /** Optional. Possible values of the element of primitive type with enum format. Examples: 1. We can define direction as : {type:STRING, format:enum, enum:["EAST", NORTH", "SOUTH", "WEST"]} 2. We can define apartment number as : {type:INTEGER, format:enum, enum:["101", "201", "301"]} */ -- enum?: string[]; -- /** Optional. Example of the object. Will only populated when the object is the root. */ -- example?: unknown; -- /** Optional. The format of the data. Supported formats: for NUMBER type: "float", "double" for INTEGER type: "int32", "int64" for STRING type: "email", "byte", etc */ -- format?: string; -- /** Optional. SCHEMA FIELDS FOR TYPE ARRAY Schema of the elements of Type.ARRAY. */ -- items?: Schema; -- /** Optional. Maximum number of the elements for Type.ARRAY. */ -- maxItems?: string; -- /** Optional. Maximum length of the Type.STRING */ -- maxLength?: string; -- /** Optional. Maximum number of the properties for Type.OBJECT. */ -- maxProperties?: string; -- /** Optional. Maximum value of the Type.INTEGER and Type.NUMBER */ -- maximum?: number; -- /** Optional. Minimum number of the elements for Type.ARRAY. */ -- minItems?: string; -- /** Optional. SCHEMA FIELDS FOR TYPE STRING Minimum length of the Type.STRING */ -- minLength?: string; -- /** Optional. Minimum number of the properties for Type.OBJECT. */ -- minProperties?: string; -- /** Optional. SCHEMA FIELDS FOR TYPE INTEGER and NUMBER Minimum value of the Type.INTEGER and Type.NUMBER */ -- minimum?: number; -- /** Optional. Indicates if the value may be null. */ -- nullable?: boolean; -- /** Optional. Pattern of the Type.STRING to restrict a string to a regular expression. */ -- pattern?: string; -- /** Optional. SCHEMA FIELDS FOR TYPE OBJECT Properties of Type.OBJECT. */ -- properties?: Record; -- /** Optional. The order of the properties. Not a standard field in open api spec. Only used to support the order of the properties. */ -- propertyOrdering?: string[]; -- /** Optional. Required properties of Type.OBJECT. */ -- required?: string[]; -- /** Optional. The title of the Schema. */ -- title?: string; -- /** Optional. The type of the data. */ -- type?: Type; --} -- --export declare type SchemaUnion = Schema | unknown; -- --/** Google search entry point. */ --export declare interface SearchEntryPoint { -- /** Optional. Web content snippet that can be embedded in a web page or an app webview. */ -- renderedContent?: string; -- /** Optional. Base64 encoded JSON representing array of tuple. */ -- sdkBlob?: string; --} -- --/** Segment of the content. */ --export declare interface Segment { -- /** Output only. End index in the given Part, measured in bytes. Offset from the start of the Part, exclusive, starting at zero. */ -- endIndex?: number; -- /** Output only. The index of a Part object within its parent Content object. */ -- partIndex?: number; -- /** Output only. Start index in the given Part, measured in bytes. Offset from the start of the Part, inclusive, starting at zero. */ -- startIndex?: number; -- /** Output only. The text corresponding to the segment from the response. */ -- text?: string; --} -- --/** Parameters for sending a message within a chat session. -- -- These parameters are used with the `chat.sendMessage()` method. -- */ --export declare interface SendMessageParameters { -- /** The message to send to the model. -- -- The SDK will combine all parts into a single 'user' content to send to -- the model. -- */ -- message: PartListUnion; -- /** Config for this specific request. -- -- Please note that the per-request config does not change the chat level -- config, nor inherit from it. If you intend to use some values from the -- chat's default config, you must explicitly copy them into this per-request -- config. -- */ -- config?: GenerateContentConfig; --} -- --/** -- Represents a connection to the API. -- -- @experimental -- */ --export declare class Session { -- readonly conn: WebSocket_2; -- private readonly apiClient; -- constructor(conn: WebSocket_2, apiClient: ApiClient); -- private tLiveClientContent; -- private tLiveClienttToolResponse; -- /** -- Send a message over the established connection. -- -- @param params - Contains two **optional** properties, `turns` and -- `turnComplete`. -- -- - `turns` will be converted to a `Content[]` -- - `turnComplete: true` [default] indicates that you are done sending -- content and expect a response. If `turnComplete: false`, the server -- will wait for additional messages before starting generation. -- -- @experimental -- -- @remarks -- There are two ways to send messages to the live API: -- `sendClientContent` and `sendRealtimeInput`. -- -- `sendClientContent` messages are added to the model context **in order**. -- Having a conversation using `sendClientContent` messages is roughly -- equivalent to using the `Chat.sendMessageStream`, except that the state of -- the `chat` history is stored on the API server instead of locally. -- -- Because of `sendClientContent`'s order guarantee, the model cannot respons -- as quickly to `sendClientContent` messages as to `sendRealtimeInput` -- messages. This makes the biggest difference when sending objects that have -- significant preprocessing time (typically images). -- -- The `sendClientContent` message sends a `Content[]` -- which has more options than the `Blob` sent by `sendRealtimeInput`. -- -- So the main use-cases for `sendClientContent` over `sendRealtimeInput` are: -- -- - Sending anything that can't be represented as a `Blob` (text, -- `sendClientContent({turns="Hello?"}`)). -- - Managing turns when not using audio input and voice activity detection. -- (`sendClientContent({turnComplete:true})` or the short form -- `sendClientContent()`) -- - Prefilling a conversation context -- ``` -- sendClientContent({ -- turns: [ -- Content({role:user, parts:...}), -- Content({role:user, parts:...}), -- ... -- ] -- }) -- ``` -- @experimental -- */ -- sendClientContent(params: types.LiveSendClientContentParameters): void; -- /** -- Send a realtime message over the established connection. -- -- @param params - Contains one property, `media`. -- -- - `media` will be converted to a `Blob` -- -- @experimental -- -- @remarks -- Use `sendRealtimeInput` for realtime audio chunks and video frames (images). -- -- With `sendRealtimeInput` the api will respond to audio automatically -- based on voice activity detection (VAD). -- -- `sendRealtimeInput` is optimized for responsivness at the expense of -- deterministic ordering guarantees. Audio and video tokens are to the -- context when they become available. -- -- Note: The Call signature expects a `Blob` object, but only a subset -- of audio and image mimetypes are allowed. -- */ -- sendRealtimeInput(params: types.LiveSendRealtimeInputParameters): void; -- /** -- Send a function response message over the established connection. -- -- @param params - Contains property `functionResponses`. -- -- - `functionResponses` will be converted to a `functionResponses[]` -- -- @remarks -- Use `sendFunctionResponse` to reply to `LiveServerToolCall` from the server. -- -- Use {@link types.LiveConnectConfig#tools} to configure the callable functions. -- -- @experimental -- */ -- sendToolResponse(params: types.LiveSendToolResponseParameters): void; -- /** -- Terminates the WebSocket connection. -- -- @experimental -- -- @example -- ```ts -- let model: string; -- if (GOOGLE_GENAI_USE_VERTEXAI) { -- model = 'gemini-2.0-flash-live-preview-04-09'; -- } else { -- model = 'gemini-2.0-flash-live-001'; -- } -- const session = await ai.live.connect({ -- model: model, -- config: { -- responseModalities: [Modality.AUDIO], -- } -- }); -- -- session.close(); -- ``` -- */ -- close(): void; --} -- --/** Configuration of session resumption mechanism. -- -- Included in `LiveConnectConfig.session_resumption`. If included server -- will send `LiveServerSessionResumptionUpdate` messages. -- */ --export declare interface SessionResumptionConfig { -- /** Session resumption handle of previous session (session to restore). -- -- If not present new session will be started. */ -- handle?: string; -- /** If set the server will send `last_consumed_client_message_index` in the `session_resumption_update` messages to allow for transparent reconnections. */ -- transparent?: boolean; --} -- --/** -- * Overrides the base URLs for the Gemini API and Vertex AI API. -- * -- * @remarks This function should be called before initializing the SDK. If the -- * base URLs are set after initializing the SDK, the base URLs will not be -- * updated. Base URLs provided in the HttpOptions will also take precedence over -- * URLs set here. -- * -- * @example -- * ```ts -- * import {GoogleGenAI, setDefaultBaseUrls} from '@google/genai'; -- * // Override the base URL for the Gemini API. -- * setDefaultBaseUrls({geminiUrl:'https://gemini.google.com'}); -- * -- * // Override the base URL for the Vertex AI API. -- * setDefaultBaseUrls({vertexUrl: 'https://vertexai.googleapis.com'}); -- * -- * const ai = new GoogleGenAI({apiKey: 'GEMINI_API_KEY'}); -- * ``` -- */ --export declare function setDefaultBaseUrls(baseUrlParams: BaseUrlParameters): void; -- --/** Context window will be truncated by keeping only suffix of it. -- -- Context window will always be cut at start of USER role turn. System -- instructions and `BidiGenerateContentSetup.prefix_turns` will not be -- subject to the sliding window mechanism, they will always stay at the -- beginning of context window. -- */ --export declare interface SlidingWindow { -- /** Session reduction target -- how many tokens we should keep. Window shortening operation has some latency costs, so we should avoid running it on every turn. Should be < trigger_tokens. If not set, trigger_tokens/2 is assumed. */ -- targetTokens?: string; --} -- --/** The configuration for the speaker to use. */ --export declare interface SpeakerVoiceConfig { -- /** The name of the speaker to use. Should be the same as in the -- prompt. */ -- speaker?: string; -- /** The configuration for the voice to use. */ -- voiceConfig?: VoiceConfig; --} -- --/** The speech generation configuration. */ --export declare interface SpeechConfig { -- /** The configuration for the speaker to use. -- */ -- voiceConfig?: VoiceConfig; -- /** The configuration for the multi-speaker setup. -- It is mutually exclusive with the voice_config field. -- */ -- multiSpeakerVoiceConfig?: MultiSpeakerVoiceConfig; -- /** Language code (ISO 639. e.g. en-US) for the speech synthesization. -- Only available for Live API. -- */ -- languageCode?: string; --} -- --export declare type SpeechConfigUnion = SpeechConfig | string; -- --/** Start of speech sensitivity. */ --export declare enum StartSensitivity { -- /** -- * The default is START_SENSITIVITY_LOW. -- */ -- START_SENSITIVITY_UNSPECIFIED = "START_SENSITIVITY_UNSPECIFIED", -- /** -- * Automatic detection will detect the start of speech more often. -- */ -- START_SENSITIVITY_HIGH = "START_SENSITIVITY_HIGH", -- /** -- * Automatic detection will detect the start of speech less often. -- */ -- START_SENSITIVITY_LOW = "START_SENSITIVITY_LOW" --} -- --/** Configuration for a Style reference image. */ --export declare interface StyleReferenceConfig { -- /** A text description of the style to use for the generated image. */ -- styleDescription?: string; --} -- --/** A style reference image. -- -- This encapsulates a style reference image provided by the user, and -- additionally optional config parameters for the style reference image. -- -- A raw reference image can also be provided as a destination for the style to -- be applied to. -- */ --export declare class StyleReferenceImage { -- /** The reference image for the editing operation. */ -- referenceImage?: Image_2; -- /** The id of the reference image. */ -- referenceId?: number; -- /** The type of the reference image. Only set by the SDK. */ -- referenceType?: string; -- /** Configuration for the style reference image. */ -- config?: StyleReferenceConfig; -- /** Internal method to convert to ReferenceImageAPIInternal. */ -- toReferenceImageAPI(): any; --} -- --/** Configuration for a Subject reference image. */ --export declare interface SubjectReferenceConfig { -- /** The subject type of a subject reference image. */ -- subjectType?: SubjectReferenceType; -- /** Subject description for the image. */ -- subjectDescription?: string; --} -- --/** A subject reference image. -- -- This encapsulates a subject reference image provided by the user, and -- additionally optional config parameters for the subject reference image. -- -- A raw reference image can also be provided as a destination for the subject to -- be applied to. -- */ --export declare class SubjectReferenceImage { -- /** The reference image for the editing operation. */ -- referenceImage?: Image_2; -- /** The id of the reference image. */ -- referenceId?: number; -- /** The type of the reference image. Only set by the SDK. */ -- referenceType?: string; -- /** Configuration for the subject reference image. */ -- config?: SubjectReferenceConfig; -- toReferenceImageAPI(): any; --} -- --/** Enum representing the subject type of a subject reference image. */ --export declare enum SubjectReferenceType { -- SUBJECT_TYPE_DEFAULT = "SUBJECT_TYPE_DEFAULT", -- SUBJECT_TYPE_PERSON = "SUBJECT_TYPE_PERSON", -- SUBJECT_TYPE_ANIMAL = "SUBJECT_TYPE_ANIMAL", -- SUBJECT_TYPE_PRODUCT = "SUBJECT_TYPE_PRODUCT" --} -- --/** Hyperparameters for SFT. */ --export declare interface SupervisedHyperParameters { -- /** Optional. Adapter size for tuning. */ -- adapterSize?: AdapterSize; -- /** Optional. Number of complete passes the model makes over the entire training dataset during training. */ -- epochCount?: string; -- /** Optional. Multiplier for adjusting the default learning rate. */ -- learningRateMultiplier?: number; --} -- --/** Dataset distribution for Supervised Tuning. */ --export declare interface SupervisedTuningDatasetDistribution { -- /** Output only. Sum of a given population of values that are billable. */ -- billableSum?: string; -- /** Output only. Defines the histogram bucket. */ -- buckets?: SupervisedTuningDatasetDistributionDatasetBucket[]; -- /** Output only. The maximum of the population values. */ -- max?: number; -- /** Output only. The arithmetic mean of the values in the population. */ -- mean?: number; -- /** Output only. The median of the values in the population. */ -- median?: number; -- /** Output only. The minimum of the population values. */ -- min?: number; -- /** Output only. The 5th percentile of the values in the population. */ -- p5?: number; -- /** Output only. The 95th percentile of the values in the population. */ -- p95?: number; -- /** Output only. Sum of a given population of values. */ -- sum?: string; --} -- --/** Dataset bucket used to create a histogram for the distribution given a population of values. */ --export declare interface SupervisedTuningDatasetDistributionDatasetBucket { -- /** Output only. Number of values in the bucket. */ -- count?: number; -- /** Output only. Left bound of the bucket. */ -- left?: number; -- /** Output only. Right bound of the bucket. */ -- right?: number; --} -- --/** Tuning data statistics for Supervised Tuning. */ --export declare interface SupervisedTuningDataStats { -- /** Output only. Number of billable characters in the tuning dataset. */ -- totalBillableCharacterCount?: string; -- /** Output only. Number of billable tokens in the tuning dataset. */ -- totalBillableTokenCount?: string; -- /** The number of examples in the dataset that have been truncated by any amount. */ -- totalTruncatedExampleCount?: string; -- /** Output only. Number of tuning characters in the tuning dataset. */ -- totalTuningCharacterCount?: string; -- /** A partial sample of the indices (starting from 1) of the truncated examples. */ -- truncatedExampleIndices?: string[]; -- /** Output only. Number of examples in the tuning dataset. */ -- tuningDatasetExampleCount?: string; -- /** Output only. Number of tuning steps for this Tuning Job. */ -- tuningStepCount?: string; -- /** Output only. Sample user messages in the training dataset uri. */ -- userDatasetExamples?: Content[]; -- /** Output only. Dataset distributions for the user input tokens. */ -- userInputTokenDistribution?: SupervisedTuningDatasetDistribution; -- /** Output only. Dataset distributions for the messages per example. */ -- userMessagePerExampleDistribution?: SupervisedTuningDatasetDistribution; -- /** Output only. Dataset distributions for the user output tokens. */ -- userOutputTokenDistribution?: SupervisedTuningDatasetDistribution; --} -- --/** Tuning Spec for Supervised Tuning for first party models. */ --export declare interface SupervisedTuningSpec { -- /** Optional. Hyperparameters for SFT. */ -- hyperParameters?: SupervisedHyperParameters; -- /** Required. Cloud Storage path to file containing training dataset for tuning. The dataset must be formatted as a JSONL file. */ -- trainingDatasetUri?: string; -- /** Optional. Cloud Storage path to file containing validation dataset for tuning. The dataset must be formatted as a JSONL file. */ -- validationDatasetUri?: string; -- /** Optional. If set to true, disable intermediate checkpoints for SFT and only the last checkpoint will be exported. */ -- exportLastCheckpointOnly?: boolean; --} -- --export declare interface TestTableFile { -- comment?: string; -- testMethod?: string; -- parameterNames?: string[]; -- testTable?: TestTableItem[]; --} -- --export declare interface TestTableItem { -- /** The name of the test. This is used to derive the replay id. */ -- name?: string; -- /** The parameters to the test. Use pydantic models. */ -- parameters?: Record; -- /** Expects an exception for MLDev matching the string. */ -- exceptionIfMldev?: string; -- /** Expects an exception for Vertex matching the string. */ -- exceptionIfVertex?: string; -- /** Use if you don't want to use the default replay id which is derived from the test name. */ -- overrideReplayId?: string; -- /** True if the parameters contain an unsupported union type. This test will be skipped for languages that do not support the union type. */ -- hasUnion?: boolean; -- /** When set to a reason string, this test will be skipped in the API mode. Use this flag for tests that can not be reproduced with the real API. E.g. a test that deletes a resource. */ -- skipInApiMode?: string; -- /** Keys to ignore when comparing the request and response. This is useful for tests that are not deterministic. */ -- ignoreKeys?: string[]; --} -- --/** The thinking features configuration. */ --export declare interface ThinkingConfig { -- /** Indicates whether to include thoughts in the response. If true, thoughts are returned only if the model supports thought and thoughts are available. -- */ -- includeThoughts?: boolean; -- /** Indicates the thinking budget in tokens. -- */ -- thinkingBudget?: number; --} -- --/** Tokens info with a list of tokens and the corresponding list of token ids. */ --export declare interface TokensInfo { -- /** Optional. Optional fields for the role from the corresponding Content. */ -- role?: string; -- /** A list of token ids from the input. */ -- tokenIds?: string[]; -- /** A list of tokens from the input. */ -- tokens?: string[]; --} -- --/** Tool details of a tool that the model may use to generate a response. */ --export declare interface Tool { -- /** List of function declarations that the tool supports. */ -- functionDeclarations?: FunctionDeclaration[]; -- /** Optional. Retrieval tool type. System will always execute the provided retrieval tool(s) to get external knowledge to answer the prompt. Retrieval results are presented to the model for generation. */ -- retrieval?: Retrieval; -- /** Optional. Google Search tool type. Specialized retrieval tool -- that is powered by Google Search. */ -- googleSearch?: GoogleSearch; -- /** Optional. GoogleSearchRetrieval tool type. Specialized retrieval tool that is powered by Google search. */ -- googleSearchRetrieval?: GoogleSearchRetrieval; -- /** Optional. Enterprise web search tool type. Specialized retrieval -- tool that is powered by Vertex AI Search and Sec4 compliance. */ -- enterpriseWebSearch?: EnterpriseWebSearch; -- /** Optional. Google Maps tool type. Specialized retrieval tool -- that is powered by Google Maps. */ -- googleMaps?: GoogleMaps; -- /** Optional. Tool to support URL context retrieval. */ -- urlContext?: UrlContext; -- /** Optional. CodeExecution tool type. Enables the model to execute code as part of generation. This field is only used by the Gemini Developer API services. */ -- codeExecution?: ToolCodeExecution; --} -- --/** Tool that executes code generated by the model, and automatically returns the result to the model. See also [ExecutableCode]and [CodeExecutionResult] which are input and output to this tool. */ --export declare interface ToolCodeExecution { --} -- --/** Tool config. -- -- This config is shared for all tools provided in the request. -- */ --export declare interface ToolConfig { -- /** Optional. Function calling config. */ -- functionCallingConfig?: FunctionCallingConfig; -- /** Optional. Retrieval config. */ -- retrievalConfig?: RetrievalConfig; --} -- --export declare type ToolListUnion = ToolUnion[]; -- --export declare type ToolUnion = Tool | CallableTool; -- --/** Output only. Traffic type. This shows whether a request consumes Pay-As-You-Go or Provisioned Throughput quota. */ --export declare enum TrafficType { -- /** -- * Unspecified request traffic type. -- */ -- TRAFFIC_TYPE_UNSPECIFIED = "TRAFFIC_TYPE_UNSPECIFIED", -- /** -- * Type for Pay-As-You-Go traffic. -- */ -- ON_DEMAND = "ON_DEMAND", -- /** -- * Type for Provisioned Throughput traffic. -- */ -- PROVISIONED_THROUGHPUT = "PROVISIONED_THROUGHPUT" --} -- --/** Audio transcription in Server Conent. */ --export declare interface Transcription { -- /** Transcription text. -- */ -- text?: string; -- /** The bool indicates the end of the transcription. -- */ -- finished?: boolean; --} -- --export declare interface TunedModel { -- /** Output only. The resource name of the TunedModel. Format: `projects/{project}/locations/{location}/models/{model}`. */ -- model?: string; -- /** Output only. A resource name of an Endpoint. Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`. */ -- endpoint?: string; -- /** The checkpoints associated with this TunedModel. -- This field is only populated for tuning jobs that enable intermediate -- checkpoints. */ -- checkpoints?: TunedModelCheckpoint[]; --} -- --/** TunedModelCheckpoint for the Tuned Model of a Tuning Job. */ --export declare interface TunedModelCheckpoint { -- /** The ID of the checkpoint. -- */ -- checkpointId?: string; -- /** The epoch of the checkpoint. -- */ -- epoch?: string; -- /** The step of the checkpoint. -- */ -- step?: string; -- /** The Endpoint resource name that the checkpoint is deployed to. -- Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`. -- */ -- endpoint?: string; --} -- --/** A tuned machine learning model. */ --export declare interface TunedModelInfo { -- /** ID of the base model that you want to tune. */ -- baseModel?: string; -- /** Date and time when the base model was created. */ -- createTime?: string; -- /** Date and time when the base model was last updated. */ -- updateTime?: string; --} -- --/** Supervised fine-tuning training dataset. */ --export declare interface TuningDataset { -- /** GCS URI of the file containing training dataset in JSONL format. */ -- gcsUri?: string; -- /** Inline examples with simple input/output text. */ -- examples?: TuningExample[]; --} -- --/** The tuning data statistic values for TuningJob. */ --export declare interface TuningDataStats { -- /** Output only. Statistics for distillation. */ -- distillationDataStats?: DistillationDataStats; -- /** The SFT Tuning data stats. */ -- supervisedTuningDataStats?: SupervisedTuningDataStats; --} -- --export declare interface TuningExample { -- /** Text model input. */ -- textInput?: string; -- /** The expected model output. */ -- output?: string; --} -- --/** A tuning job. */ --export declare interface TuningJob { -- /** Output only. Identifier. Resource name of a TuningJob. Format: `projects/{project}/locations/{location}/tuningJobs/{tuning_job}` */ -- name?: string; -- /** Output only. The detailed state of the job. */ -- state?: JobState; -- /** Output only. Time when the TuningJob was created. */ -- createTime?: string; -- /** Output only. Time when the TuningJob for the first time entered the `JOB_STATE_RUNNING` state. */ -- startTime?: string; -- /** Output only. Time when the TuningJob entered any of the following JobStates: `JOB_STATE_SUCCEEDED`, `JOB_STATE_FAILED`, `JOB_STATE_CANCELLED`, `JOB_STATE_EXPIRED`. */ -- endTime?: string; -- /** Output only. Time when the TuningJob was most recently updated. */ -- updateTime?: string; -- /** Output only. Only populated when job's state is `JOB_STATE_FAILED` or `JOB_STATE_CANCELLED`. */ -- error?: GoogleRpcStatus; -- /** Optional. The description of the TuningJob. */ -- description?: string; -- /** The base model that is being tuned, e.g., "gemini-1.0-pro-002". . */ -- baseModel?: string; -- /** Output only. The tuned model resources associated with this TuningJob. */ -- tunedModel?: TunedModel; -- /** Tuning Spec for Supervised Fine Tuning. */ -- supervisedTuningSpec?: SupervisedTuningSpec; -- /** Output only. The tuning data statistics associated with this TuningJob. */ -- tuningDataStats?: TuningDataStats; -- /** Customer-managed encryption key options for a TuningJob. If this is set, then all resources created by the TuningJob will be encrypted with the provided encryption key. */ -- encryptionSpec?: EncryptionSpec; -- /** Tuning Spec for open sourced and third party Partner models. */ -- partnerModelTuningSpec?: PartnerModelTuningSpec; -- /** Tuning Spec for Distillation. */ -- distillationSpec?: DistillationSpec; -- /** Output only. The Experiment associated with this TuningJob. */ -- experiment?: string; -- /** Optional. The labels with user-defined metadata to organize TuningJob and generated resources such as Model and Endpoint. Label keys and values can be no longer than 64 characters (Unicode codepoints), can only contain lowercase letters, numeric characters, underscores and dashes. International characters are allowed. See https://goo.gl/xmQnxf for more information and examples of labels. */ -- labels?: Record; -- /** Output only. The resource name of the PipelineJob associated with the TuningJob. Format: `projects/{project}/locations/{location}/pipelineJobs/{pipeline_job}`. */ -- pipelineJob?: string; -- /** Optional. The display name of the TunedModel. The name can be up to 128 characters long and can consist of any UTF-8 characters. */ -- tunedModelDisplayName?: string; --} -- --declare class Tunings extends BaseModule { -- private readonly apiClient; -- constructor(apiClient: ApiClient); -- /** -- * Gets a TuningJob. -- * -- * @param name - The resource name of the tuning job. -- * @return - A TuningJob object. -- * -- * @experimental - The SDK's tuning implementation is experimental, and may -- * change in future versions. -- */ -- get: (params: types.GetTuningJobParameters) => Promise; -- /** -- * Lists tuning jobs. -- * -- * @param config - The configuration for the list request. -- * @return - A list of tuning jobs. -- * -- * @experimental - The SDK's tuning implementation is experimental, and may -- * change in future versions. -- */ -- list: (params?: types.ListTuningJobsParameters) => Promise>; -- /** -- * Creates a supervised fine-tuning job. -- * -- * @param params - The parameters for the tuning job. -- * @return - A TuningJob operation. -- * -- * @experimental - The SDK's tuning implementation is experimental, and may -- * change in future versions. -- */ -- tune: (params: types.CreateTuningJobParameters) => Promise; -- private getInternal; -- private listInternal; -- private tuneInternal; -- private tuneMldevInternal; --} -- --export declare interface TuningValidationDataset { -- /** GCS URI of the file containing validation dataset in JSONL format. */ -- gcsUri?: string; --} -- --/** Options about which input is included in the user's turn. */ --export declare enum TurnCoverage { -- /** -- * If unspecified, the default behavior is `TURN_INCLUDES_ONLY_ACTIVITY`. -- */ -- TURN_COVERAGE_UNSPECIFIED = "TURN_COVERAGE_UNSPECIFIED", -- /** -- * The users turn only includes activity since the last turn, excluding inactivity (e.g. silence on the audio stream). This is the default behavior. -- */ -- TURN_INCLUDES_ONLY_ACTIVITY = "TURN_INCLUDES_ONLY_ACTIVITY", -- /** -- * The users turn includes all realtime input since the last turn, including inactivity (e.g. silence on the audio stream). -- */ -- TURN_INCLUDES_ALL_INPUT = "TURN_INCLUDES_ALL_INPUT" --} -- --/** Optional. The type of the data. */ --export declare enum Type { -- /** -- * Not specified, should not be used. -- */ -- TYPE_UNSPECIFIED = "TYPE_UNSPECIFIED", -- /** -- * OpenAPI string type -- */ -- STRING = "STRING", -- /** -- * OpenAPI number type -- */ -- NUMBER = "NUMBER", -- /** -- * OpenAPI integer type -- */ -- INTEGER = "INTEGER", -- /** -- * OpenAPI boolean type -- */ -- BOOLEAN = "BOOLEAN", -- /** -- * OpenAPI array type -- */ -- ARRAY = "ARRAY", -- /** -- * OpenAPI object type -- */ -- OBJECT = "OBJECT" --} -- --declare namespace types { -- export { -- createPartFromUri, -- createPartFromText, -- createPartFromFunctionCall, -- createPartFromFunctionResponse, -- createPartFromBase64, -- createPartFromCodeExecutionResult, -- createPartFromExecutableCode, -- createUserContent, -- createModelContent, -- Outcome, -- Language, -- HarmCategory, -- HarmBlockMethod, -- HarmBlockThreshold, -- Type, -- Mode, -- AuthType, -- FinishReason, -- HarmProbability, -- HarmSeverity, -- BlockedReason, -- TrafficType, -- Modality, -- MediaResolution, -- JobState, -- AdapterSize, -- FeatureSelectionPreference, -- Behavior, -- DynamicRetrievalConfigMode, -- FunctionCallingConfigMode, -- UrlRetrievalStatus, -- SafetyFilterLevel, -- PersonGeneration, -- ImagePromptLanguage, -- MaskReferenceMode, -- ControlReferenceType, -- SubjectReferenceType, -- EditMode, -- FileState, -- FileSource, -- MediaModality, -- StartSensitivity, -- EndSensitivity, -- ActivityHandling, -- TurnCoverage, -- FunctionResponseScheduling, -- Scale, -- MusicGenerationMode, -- LiveMusicPlaybackControl, -- VideoMetadata, -- Blob_2 as Blob, -- CodeExecutionResult, -- ExecutableCode, -- FileData, -- FunctionCall, -- FunctionResponse, -- Part, -- Content, -- HttpOptions, -- ModelSelectionConfig, -- SafetySetting, -- Schema, -- FunctionDeclaration, -- Interval, -- GoogleSearch, -- DynamicRetrievalConfig, -- GoogleSearchRetrieval, -- EnterpriseWebSearch, -- ApiKeyConfig, -- AuthConfigGoogleServiceAccountConfig, -- AuthConfigHttpBasicAuthConfig, -- AuthConfigOauthConfig, -- AuthConfigOidcConfig, -- AuthConfig, -- GoogleMaps, -- UrlContext, -- VertexAISearch, -- VertexRagStoreRagResource, -- RagRetrievalConfigFilter, -- RagRetrievalConfigHybridSearch, -- RagRetrievalConfigRankingLlmRanker, -- RagRetrievalConfigRankingRankService, -- RagRetrievalConfigRanking, -- RagRetrievalConfig, -- VertexRagStore, -- Retrieval, -- ToolCodeExecution, -- Tool, -- FunctionCallingConfig, -- LatLng, -- RetrievalConfig, -- ToolConfig, -- PrebuiltVoiceConfig, -- VoiceConfig, -- SpeakerVoiceConfig, -- MultiSpeakerVoiceConfig, -- SpeechConfig, -- AutomaticFunctionCallingConfig, -- ThinkingConfig, -- GenerationConfigRoutingConfigAutoRoutingMode, -- GenerationConfigRoutingConfigManualRoutingMode, -- GenerationConfigRoutingConfig, -- GenerateContentConfig, -- GenerateContentParameters, -- GoogleTypeDate, -- Citation, -- CitationMetadata, -- UrlMetadata, -- UrlContextMetadata, -- GroundingChunkRetrievedContext, -- GroundingChunkWeb, -- GroundingChunk, -- Segment, -- GroundingSupport, -- RetrievalMetadata, -- SearchEntryPoint, -- GroundingMetadata, -- LogprobsResultCandidate, -- LogprobsResultTopCandidates, -- LogprobsResult, -- SafetyRating, -- Candidate, -- GenerateContentResponsePromptFeedback, -- ModalityTokenCount, -- GenerateContentResponseUsageMetadata, -- GenerateContentResponse, -- ReferenceImage, -- EditImageParameters, -- EmbedContentConfig, -- EmbedContentParameters, -- ContentEmbeddingStatistics, -- ContentEmbedding, -- EmbedContentMetadata, -- EmbedContentResponse, -- GenerateImagesConfig, -- GenerateImagesParameters, -- Image_2 as Image, -- SafetyAttributes, -- GeneratedImage, -- GenerateImagesResponse, -- MaskReferenceConfig, -- ControlReferenceConfig, -- StyleReferenceConfig, -- SubjectReferenceConfig, -- EditImageConfig, -- EditImageResponse, -- UpscaleImageResponse, -- GetModelConfig, -- GetModelParameters, -- Endpoint, -- TunedModelInfo, -- Checkpoint, -- Model, -- ListModelsConfig, -- ListModelsParameters, -- ListModelsResponse, -- UpdateModelConfig, -- UpdateModelParameters, -- DeleteModelConfig, -- DeleteModelParameters, -- DeleteModelResponse, -- GenerationConfig, -- CountTokensConfig, -- CountTokensParameters, -- CountTokensResponse, -- ComputeTokensConfig, -- ComputeTokensParameters, -- TokensInfo, -- ComputeTokensResponse, -- GenerateVideosConfig, -- GenerateVideosParameters, -- Video, -- GeneratedVideo, -- GenerateVideosResponse, -- GenerateVideosOperation, -- GetTuningJobConfig, -- GetTuningJobParameters, -- TunedModelCheckpoint, -- TunedModel, -- GoogleRpcStatus, -- SupervisedHyperParameters, -- SupervisedTuningSpec, -- DatasetDistributionDistributionBucket, -- DatasetDistribution, -- DatasetStats, -- DistillationDataStats, -- SupervisedTuningDatasetDistributionDatasetBucket, -- SupervisedTuningDatasetDistribution, -- SupervisedTuningDataStats, -- TuningDataStats, -- EncryptionSpec, -- PartnerModelTuningSpec, -- DistillationHyperParameters, -- DistillationSpec, -- TuningJob, -- ListTuningJobsConfig, -- ListTuningJobsParameters, -- ListTuningJobsResponse, -- TuningExample, -- TuningDataset, -- TuningValidationDataset, -- CreateTuningJobConfig, -- CreateTuningJobParameters, -- Operation, -- CreateCachedContentConfig, -- CreateCachedContentParameters, -- CachedContentUsageMetadata, -- CachedContent, -- GetCachedContentConfig, -- GetCachedContentParameters, -- DeleteCachedContentConfig, -- DeleteCachedContentParameters, -- DeleteCachedContentResponse, -- UpdateCachedContentConfig, -- UpdateCachedContentParameters, -- ListCachedContentsConfig, -- ListCachedContentsParameters, -- ListCachedContentsResponse, -- ListFilesConfig, -- ListFilesParameters, -- FileStatus, -- File_2 as File, -- ListFilesResponse, -- CreateFileConfig, -- CreateFileParameters, -- HttpResponse, -- LiveCallbacks, -- CreateFileResponse, -- GetFileConfig, -- GetFileParameters, -- DeleteFileConfig, -- DeleteFileParameters, -- DeleteFileResponse, -- GetOperationConfig, -- GetOperationParameters, -- FetchPredictOperationConfig, -- FetchPredictOperationParameters, -- TestTableItem, -- TestTableFile, -- ReplayRequest, -- ReplayResponse, -- ReplayInteraction, -- ReplayFile, -- UploadFileConfig, -- DownloadFileConfig, -- DownloadFileParameters, -- UpscaleImageConfig, -- UpscaleImageParameters, -- RawReferenceImage, -- MaskReferenceImage, -- ControlReferenceImage, -- StyleReferenceImage, -- SubjectReferenceImage, -- LiveServerSetupComplete, -- Transcription, -- LiveServerContent, -- LiveServerToolCall, -- LiveServerToolCallCancellation, -- UsageMetadata, -- LiveServerGoAway, -- LiveServerSessionResumptionUpdate, -- LiveServerMessage, -- AutomaticActivityDetection, -- RealtimeInputConfig, -- SessionResumptionConfig, -- SlidingWindow, -- ContextWindowCompressionConfig, -- AudioTranscriptionConfig, -- ProactivityConfig, -- LiveClientSetup, -- LiveClientContent, -- ActivityStart, -- ActivityEnd, -- LiveClientRealtimeInput, -- LiveSendRealtimeInputParameters, -- LiveClientToolResponse, -- LiveClientMessage, -- LiveConnectConfig, -- LiveConnectParameters, -- CreateChatParameters, -- SendMessageParameters, -- LiveSendClientContentParameters, -- LiveSendToolResponseParameters, -- LiveMusicClientSetup, -- WeightedPrompt, -- LiveMusicClientContent, -- LiveMusicGenerationConfig, -- LiveMusicClientMessage, -- LiveMusicServerSetupComplete, -- LiveMusicSourceMetadata, -- AudioChunk, -- LiveMusicServerContent, -- LiveMusicFilteredPrompt, -- LiveMusicServerMessage, -- LiveMusicCallbacks, -- UploadFileParameters, -- CallableTool, -- CallableToolConfig, -- LiveMusicConnectParameters, -- LiveMusicSetConfigParameters, -- LiveMusicSetWeightedPromptsParameters, -- LiveEphemeralParameters, -- CreateAuthTokenConfig, -- OperationGetParameters, -- BlobImageUnion, -- PartUnion, -- PartListUnion, -- ContentUnion, -- ContentListUnion, -- SchemaUnion, -- SpeechConfigUnion, -- ToolUnion, -- ToolListUnion, -- DownloadableFileUnion -- } --} -- --/** Optional parameters for caches.update method. */ --export declare interface UpdateCachedContentConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; -- /** The TTL for this resource. The expiration time is computed: now + TTL. It is a duration string, with up to nine fractional digits, terminated by 's'. Example: "3.5s". */ -- ttl?: string; -- /** Timestamp of when this resource is considered expired. Uses RFC 3339 format, Example: 2014-10-02T15:01:23Z. */ -- expireTime?: string; --} -- --export declare interface UpdateCachedContentParameters { -- /** The server-generated resource name of the cached content. -- */ -- name: string; -- /** Configuration that contains optional parameters. -- */ -- config?: UpdateCachedContentConfig; --} -- --/** Configuration for updating a tuned model. */ --export declare interface UpdateModelConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; -- displayName?: string; -- description?: string; -- defaultCheckpointId?: string; --} -- --/** Configuration for updating a tuned model. */ --export declare interface UpdateModelParameters { -- model: string; -- config?: UpdateModelConfig; --} -- --declare interface Uploader { -- /** -- * Uploads a file to the given upload url. -- * -- * @param file The file to upload. file is in string type or a Blob. -- * @param uploadUrl The upload URL as a string is where the file will be -- * uploaded to. The uploadUrl must be a url that was returned by the -- * https://generativelanguage.googleapis.com/upload/v1beta/files endpoint -- * @param apiClient The ApiClient to use for uploading. -- * @return A Promise that resolves to types.File. -- */ -- upload(file: string | Blob, uploadUrl: string, apiClient: ApiClient): Promise; -- /** -- * Returns the file's mimeType and the size of a given file. If the file is a -- * string path, the file type is determined by the file extension. If the -- * file's type cannot be determined, the type will be set to undefined. -- * -- * @param file The file to get the stat for. Can be a string path or a Blob. -- * @return A Promise that resolves to the file stat of the given file. -- */ -- stat(file: string | Blob): Promise; --} -- --/** Used to override the default configuration. */ --export declare interface UploadFileConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; -- /** The name of the file in the destination (e.g., 'files/sample-image'. If not provided one will be generated. */ -- name?: string; -- /** mime_type: The MIME type of the file. If not provided, it will be inferred from the file extension. */ -- mimeType?: string; -- /** Optional display name of the file. */ -- displayName?: string; --} -- --/** Parameters for the upload file method. */ --export declare interface UploadFileParameters { -- /** The string path to the file to be uploaded or a Blob object. */ -- file: string | globalThis.Blob; -- /** Configuration that contains optional parameters. */ -- config?: UploadFileConfig; --} -- --/** Configuration for upscaling an image. -- -- For more information on this configuration, refer to -- the `Imagen API reference documentation -- `_. -- */ --export declare interface UpscaleImageConfig { -- /** Used to override HTTP request options. */ -- httpOptions?: HttpOptions; -- /** Abort signal which can be used to cancel the request. -- -- NOTE: AbortSignal is a client-only operation. Using it to cancel an -- operation will not cancel the request in the service. You will still -- be charged usage for any applicable operations. -- */ -- abortSignal?: AbortSignal; -- /** Whether to include a reason for filtered-out images in the -- response. */ -- includeRaiReason?: boolean; -- /** The image format that the output should be saved as. */ -- outputMimeType?: string; -- /** The level of compression if the ``output_mime_type`` is -- ``image/jpeg``. */ -- outputCompressionQuality?: number; --} -- --/** User-facing config UpscaleImageParameters. */ --export declare interface UpscaleImageParameters { -- /** The model to use. */ -- model: string; -- /** The input image to upscale. */ -- image: Image_2; -- /** The factor to upscale the image (x2 or x4). */ -- upscaleFactor: string; -- /** Configuration for upscaling. */ -- config?: UpscaleImageConfig; --} -- --export declare class UpscaleImageResponse { -- /** Generated images. */ -- generatedImages?: GeneratedImage[]; --} -- --/** Tool to support URL context retrieval. */ --export declare interface UrlContext { --} -- --/** Metadata related to url context retrieval tool. */ --export declare interface UrlContextMetadata { -- /** List of url context. */ -- urlMetadata?: UrlMetadata[]; --} -- --/** Context for a single url retrieval. */ --export declare interface UrlMetadata { -- /** The URL retrieved by the tool. */ -- retrievedUrl?: string; -- /** Status of the url retrieval. */ -- urlRetrievalStatus?: UrlRetrievalStatus; --} -- --/** Status of the url retrieval. */ --export declare enum UrlRetrievalStatus { -- /** -- * Default value. This value is unused -- */ -- URL_RETRIEVAL_STATUS_UNSPECIFIED = "URL_RETRIEVAL_STATUS_UNSPECIFIED", -- /** -- * Url retrieval is successful. -- */ -- URL_RETRIEVAL_STATUS_SUCCESS = "URL_RETRIEVAL_STATUS_SUCCESS", -- /** -- * Url retrieval is failed due to error. -- */ -- URL_RETRIEVAL_STATUS_ERROR = "URL_RETRIEVAL_STATUS_ERROR" --} -- --/** Usage metadata about response(s). */ --export declare interface UsageMetadata { -- /** Number of tokens in the prompt. When `cached_content` is set, this is still the total effective prompt size meaning this includes the number of tokens in the cached content. */ -- promptTokenCount?: number; -- /** Number of tokens in the cached part of the prompt (the cached content). */ -- cachedContentTokenCount?: number; -- /** Total number of tokens across all the generated response candidates. */ -- responseTokenCount?: number; -- /** Number of tokens present in tool-use prompt(s). */ -- toolUsePromptTokenCount?: number; -- /** Number of tokens of thoughts for thinking models. */ -- thoughtsTokenCount?: number; -- /** Total token count for prompt, response candidates, and tool-use prompts(if present). */ -- totalTokenCount?: number; -- /** List of modalities that were processed in the request input. */ -- promptTokensDetails?: ModalityTokenCount[]; -- /** List of modalities that were processed in the cache input. */ -- cacheTokensDetails?: ModalityTokenCount[]; -- /** List of modalities that were returned in the response. */ -- responseTokensDetails?: ModalityTokenCount[]; -- /** List of modalities that were processed in the tool-use prompt. */ -- toolUsePromptTokensDetails?: ModalityTokenCount[]; -- /** Traffic type. This shows whether a request consumes Pay-As-You-Go -- or Provisioned Throughput quota. */ -- trafficType?: TrafficType; --} -- --/** Retrieve from Vertex AI Search datastore or engine for grounding. datastore and engine are mutually exclusive. See https://cloud.google.com/products/agent-builder */ --export declare interface VertexAISearch { -- /** Optional. Fully-qualified Vertex AI Search data store resource ID. Format: `projects/{project}/locations/{location}/collections/{collection}/dataStores/{dataStore}` */ -- datastore?: string; -- /** Optional. Fully-qualified Vertex AI Search engine resource ID. Format: `projects/{project}/locations/{location}/collections/{collection}/engines/{engine}` */ -- engine?: string; --} -- --/** Retrieve from Vertex RAG Store for grounding. */ --export declare interface VertexRagStore { -- /** Optional. Deprecated. Please use rag_resources instead. */ -- ragCorpora?: string[]; -- /** Optional. The representation of the rag source. It can be used to specify corpus only or ragfiles. Currently only support one corpus or multiple files from one corpus. In the future we may open up multiple corpora support. */ -- ragResources?: VertexRagStoreRagResource[]; -- /** Optional. The retrieval config for the Rag query. */ -- ragRetrievalConfig?: RagRetrievalConfig; -- /** Optional. Number of top k results to return from the selected corpora. */ -- similarityTopK?: number; -- /** Optional. Only return results with vector distance smaller than the threshold. */ -- vectorDistanceThreshold?: number; --} -- --/** The definition of the Rag resource. */ --export declare interface VertexRagStoreRagResource { -- /** Optional. RagCorpora resource name. Format: `projects/{project}/locations/{location}/ragCorpora/{rag_corpus}` */ -- ragCorpus?: string; -- /** Optional. rag_file_id. The files should be in the same rag_corpus set in rag_corpus field. */ -- ragFileIds?: string[]; --} -- --/** A generated video. */ --export declare interface Video { -- /** Path to another storage. */ -- uri?: string; -- /** Video bytes. */ -- videoBytes?: string; -- /** Video encoding, for example "video/mp4". */ -- mimeType?: string; --} -- --/** Describes how the video in the Part should be used by the model. */ --export declare interface VideoMetadata { -- /** The frame rate of the video sent to the model. If not specified, the -- default value will be 1.0. The fps range is (0.0, 24.0]. */ -- fps?: number; -- /** Optional. The end offset of the video. */ -- endOffset?: string; -- /** Optional. The start offset of the video. */ -- startOffset?: string; --} -- --/** The configuration for the voice to use. */ --export declare interface VoiceConfig { -- /** The configuration for the speaker to use. -- */ -- prebuiltVoiceConfig?: PrebuiltVoiceConfig; --} -- --declare interface WebSocket_2 { -- /** -- * Connects the socket to the server. -- */ -- connect(): void; -- /** -- * Sends a message to the server. -- */ -- send(message: string): void; -- /** -- * Closes the socket connection. -- */ -- close(): void; --} -- --/** -- * @license -- * Copyright 2025 Google LLC -- * SPDX-License-Identifier: Apache-2.0 -- */ --declare interface WebSocketCallbacks { -- onopen: () => void; -- onerror: (e: any) => void; -- onmessage: (e: any) => void; -- onclose: (e: any) => void; --} -- --declare interface WebSocketFactory { -- /** -- * Returns a new WebSocket instance. -- */ -- create(url: string, headers: Record, callbacks: WebSocketCallbacks): WebSocket_2; --} -- --/** Maps a prompt to a relative weight to steer music generation. */ --export declare interface WeightedPrompt { -- /** Text prompt. */ -- text?: string; -- /** Weight of the prompt. The weight is used to control the relative -- importance of the prompt. Higher weights are more important than lower -- weights. -- -- Weight must not be 0. Weights of all weighted_prompts in this -- LiveMusicClientContent message will be normalized. */ -- weight?: number; --} -- --export { } +-import { Client } from '@modelcontextprotocol/sdk/client/index.js'; +-import { GoogleAuthOptions } from 'google-auth-library'; +- +-/** Marks the end of user activity. +- +- This can only be sent if automatic (i.e. server-side) activity detection is +- disabled. +- */ +-export declare interface ActivityEnd { +-} +- +-/** The different ways of handling user activity. */ +-export declare enum ActivityHandling { +- /** +- * If unspecified, the default behavior is `START_OF_ACTIVITY_INTERRUPTS`. +- */ +- ACTIVITY_HANDLING_UNSPECIFIED = "ACTIVITY_HANDLING_UNSPECIFIED", +- /** +- * If true, start of activity will interrupt the model's response (also called "barge in"). The model's current response will be cut-off in the moment of the interruption. This is the default behavior. +- */ +- START_OF_ACTIVITY_INTERRUPTS = "START_OF_ACTIVITY_INTERRUPTS", +- /** +- * The model's response will not be interrupted. +- */ +- NO_INTERRUPTION = "NO_INTERRUPTION" +-} +- +-/** Marks the start of user activity. +- +- This can only be sent if automatic (i.e. server-side) activity detection is +- disabled. +- */ +-export declare interface ActivityStart { +-} +- +-/** Optional. Adapter size for tuning. */ +-export declare enum AdapterSize { +- /** +- * Adapter size is unspecified. +- */ +- ADAPTER_SIZE_UNSPECIFIED = "ADAPTER_SIZE_UNSPECIFIED", +- /** +- * Adapter size 1. +- */ +- ADAPTER_SIZE_ONE = "ADAPTER_SIZE_ONE", +- /** +- * Adapter size 2. +- */ +- ADAPTER_SIZE_TWO = "ADAPTER_SIZE_TWO", +- /** +- * Adapter size 4. +- */ +- ADAPTER_SIZE_FOUR = "ADAPTER_SIZE_FOUR", +- /** +- * Adapter size 8. +- */ +- ADAPTER_SIZE_EIGHT = "ADAPTER_SIZE_EIGHT", +- /** +- * Adapter size 16. +- */ +- ADAPTER_SIZE_SIXTEEN = "ADAPTER_SIZE_SIXTEEN", +- /** +- * Adapter size 32. +- */ +- ADAPTER_SIZE_THIRTY_TWO = "ADAPTER_SIZE_THIRTY_TWO" +-} +- +-/** +- * The ApiClient class is used to send requests to the Gemini API or Vertex AI +- * endpoints. +- */ +-declare class ApiClient { +- readonly clientOptions: ApiClientInitOptions; +- constructor(opts: ApiClientInitOptions); +- /** +- * Determines the base URL for Vertex AI based on project and location. +- * Uses the global endpoint if location is 'global' or if project/location +- * are not specified (implying API key usage). +- * @private +- */ +- private baseUrlFromProjectLocation; +- /** +- * Normalizes authentication parameters for Vertex AI. +- * If project and location are provided, API key is cleared. +- * If project and location are not provided (implying API key usage), +- * project and location are cleared. +- * @private +- */ +- private normalizeAuthParameters; +- isVertexAI(): boolean; +- getProject(): string | undefined; +- getLocation(): string | undefined; +- getApiVersion(): string; +- getBaseUrl(): string; +- getRequestUrl(): string; +- getHeaders(): Record; +- private getRequestUrlInternal; +- getBaseResourcePath(): string; +- getApiKey(): string | undefined; +- getWebsocketBaseUrl(): string; +- setBaseUrl(url: string): void; +- private constructUrl; +- private shouldPrependVertexProjectPath; +- request(request: HttpRequest): Promise; +- private patchHttpOptions; +- requestStream(request: HttpRequest): Promise>; +- private includeExtraHttpOptionsToRequestInit; +- private unaryApiCall; +- private streamApiCall; +- processStreamResponse(response: Response): AsyncGenerator; +- private apiCall; +- getDefaultHeaders(): Record; +- private getHeadersInternal; +- /** +- * Uploads a file asynchronously using Gemini API only, this is not supported +- * in Vertex AI. +- * +- * @param file The string path to the file to be uploaded or a Blob object. +- * @param config Optional parameters specified in the `UploadFileConfig` +- * interface. @see {@link UploadFileConfig} +- * @return A promise that resolves to a `File` object. +- * @throws An error if called on a Vertex AI client. +- * @throws An error if the `mimeType` is not provided and can not be inferred, +- */ +- uploadFile(file: string | Blob, config?: UploadFileConfig): Promise; +- /** +- * Downloads a file asynchronously to the specified path. +- * +- * @params params - The parameters for the download request, see {@link +- * DownloadFileParameters} +- */ +- downloadFile(params: DownloadFileParameters): Promise; +- private fetchUploadUrl; +-} +- +-/** +- * Options for initializing the ApiClient. The ApiClient uses the parameters +- * for authentication purposes as well as to infer if SDK should send the +- * request to Vertex AI or Gemini API. +- */ +-declare interface ApiClientInitOptions { +- /** +- * The object used for adding authentication headers to API requests. +- */ +- auth: Auth; +- /** +- * The uploader to use for uploading files. This field is required for +- * creating a client, will be set through the Node_client or Web_client. +- */ +- uploader: Uploader; +- /** +- * Optional. The downloader to use for downloading files. This field is +- * required for creating a client, will be set through the Node_client or +- * Web_client. +- */ +- downloader: Downloader; +- /** +- * Optional. The Google Cloud project ID for Vertex AI users. +- * It is not the numeric project name. +- * If not provided, SDK will try to resolve it from runtime environment. +- */ +- project?: string; +- /** +- * Optional. The Google Cloud project location for Vertex AI users. +- * If not provided, SDK will try to resolve it from runtime environment. +- */ +- location?: string; +- /** +- * The API Key. This is required for Gemini API users. +- */ +- apiKey?: string; +- /** +- * Optional. Set to true if you intend to call Vertex AI endpoints. +- * If unset, default SDK behavior is to call Gemini API. +- */ +- vertexai?: boolean; +- /** +- * Optional. The API version for the endpoint. +- * If unset, SDK will choose a default api version. +- */ +- apiVersion?: string; +- /** +- * Optional. A set of customizable configuration for HTTP requests. +- */ +- httpOptions?: HttpOptions; +- /** +- * Optional. An extra string to append at the end of the User-Agent header. +- * +- * This can be used to e.g specify the runtime and its version. +- */ +- userAgentExtra?: string; +-} +- +-/** Config for authentication with API key. */ +-export declare interface ApiKeyConfig { +- /** The API key to be used in the request directly. */ +- apiKeyString?: string; +-} +- +-/** Representation of an audio chunk. */ +-export declare interface AudioChunk { +- /** Raw byets of audio data. */ +- data?: string; +- /** MIME type of the audio chunk. */ +- mimeType?: string; +- /** Prompts and config used for generating this audio chunk. */ +- sourceMetadata?: LiveMusicSourceMetadata; +-} +- +-/** The audio transcription configuration in Setup. */ +-export declare interface AudioTranscriptionConfig { +-} +- +-/** +- * @license +- * Copyright 2025 Google LLC +- * SPDX-License-Identifier: Apache-2.0 +- */ +-/** +- * The Auth interface is used to authenticate with the API service. +- */ +-declare interface Auth { +- /** +- * Sets the headers needed to authenticate with the API service. +- * +- * @param headers - The Headers object that will be updated with the authentication headers. +- */ +- addAuthHeaders(headers: Headers): Promise; +-} +- +-/** Auth configuration to run the extension. */ +-export declare interface AuthConfig { +- /** Config for API key auth. */ +- apiKeyConfig?: ApiKeyConfig; +- /** Type of auth scheme. */ +- authType?: AuthType; +- /** Config for Google Service Account auth. */ +- googleServiceAccountConfig?: AuthConfigGoogleServiceAccountConfig; +- /** Config for HTTP Basic auth. */ +- httpBasicAuthConfig?: AuthConfigHttpBasicAuthConfig; +- /** Config for user oauth. */ +- oauthConfig?: AuthConfigOauthConfig; +- /** Config for user OIDC auth. */ +- oidcConfig?: AuthConfigOidcConfig; +-} +- +-/** Config for Google Service Account Authentication. */ +-export declare interface AuthConfigGoogleServiceAccountConfig { +- /** Optional. The service account that the extension execution service runs as. - If the service account is specified, the `iam.serviceAccounts.getAccessToken` permission should be granted to Vertex AI Extension Service Agent (https://cloud.google.com/vertex-ai/docs/general/access-control#service-agents) on the specified service account. - If not specified, the Vertex AI Extension Service Agent will be used to execute the Extension. */ +- serviceAccount?: string; +-} +- +-/** Config for HTTP Basic Authentication. */ +-export declare interface AuthConfigHttpBasicAuthConfig { +- /** Required. The name of the SecretManager secret version resource storing the base64 encoded credentials. Format: `projects/{project}/secrets/{secrete}/versions/{version}` - If specified, the `secretmanager.versions.access` permission should be granted to Vertex AI Extension Service Agent (https://cloud.google.com/vertex-ai/docs/general/access-control#service-agents) on the specified resource. */ +- credentialSecret?: string; +-} +- +-/** Config for user oauth. */ +-export declare interface AuthConfigOauthConfig { +- /** Access token for extension endpoint. Only used to propagate token from [[ExecuteExtensionRequest.runtime_auth_config]] at request time. */ +- accessToken?: string; +- /** The service account used to generate access tokens for executing the Extension. - If the service account is specified, the `iam.serviceAccounts.getAccessToken` permission should be granted to Vertex AI Extension Service Agent (https://cloud.google.com/vertex-ai/docs/general/access-control#service-agents) on the provided service account. */ +- serviceAccount?: string; +-} +- +-/** Config for user OIDC auth. */ +-export declare interface AuthConfigOidcConfig { +- /** OpenID Connect formatted ID token for extension endpoint. Only used to propagate token from [[ExecuteExtensionRequest.runtime_auth_config]] at request time. */ +- idToken?: string; +- /** The service account used to generate an OpenID Connect (OIDC)-compatible JWT token signed by the Google OIDC Provider (accounts.google.com) for extension endpoint (https://cloud.google.com/iam/docs/create-short-lived-credentials-direct#sa-credentials-oidc). - The audience for the token will be set to the URL in the server url defined in the OpenApi spec. - If the service account is provided, the service account should grant `iam.serviceAccounts.getOpenIdToken` permission to Vertex AI Extension Service Agent (https://cloud.google.com/vertex-ai/docs/general/access-control#service-agents). */ +- serviceAccount?: string; +-} +- +-/** Type of auth scheme. */ +-export declare enum AuthType { +- AUTH_TYPE_UNSPECIFIED = "AUTH_TYPE_UNSPECIFIED", +- /** +- * No Auth. +- */ +- NO_AUTH = "NO_AUTH", +- /** +- * API Key Auth. +- */ +- API_KEY_AUTH = "API_KEY_AUTH", +- /** +- * HTTP Basic Auth. +- */ +- HTTP_BASIC_AUTH = "HTTP_BASIC_AUTH", +- /** +- * Google Service Account Auth. +- */ +- GOOGLE_SERVICE_ACCOUNT_AUTH = "GOOGLE_SERVICE_ACCOUNT_AUTH", +- /** +- * OAuth auth. +- */ +- OAUTH = "OAUTH", +- /** +- * OpenID Connect (OIDC) Auth. +- */ +- OIDC_AUTH = "OIDC_AUTH" +-} +- +-/** Configures automatic detection of activity. */ +-export declare interface AutomaticActivityDetection { +- /** If enabled, detected voice and text input count as activity. If disabled, the client must send activity signals. */ +- disabled?: boolean; +- /** Determines how likely speech is to be detected. */ +- startOfSpeechSensitivity?: StartSensitivity; +- /** Determines how likely detected speech is ended. */ +- endOfSpeechSensitivity?: EndSensitivity; +- /** The required duration of detected speech before start-of-speech is committed. The lower this value the more sensitive the start-of-speech detection is and the shorter speech can be recognized. However, this also increases the probability of false positives. */ +- prefixPaddingMs?: number; +- /** The required duration of detected non-speech (e.g. silence) before end-of-speech is committed. The larger this value, the longer speech gaps can be without interrupting the user's activity but this will increase the model's latency. */ +- silenceDurationMs?: number; +-} +- +-/** The configuration for automatic function calling. */ +-export declare interface AutomaticFunctionCallingConfig { +- /** Whether to disable automatic function calling. +- If not set or set to False, will enable automatic function calling. +- If set to True, will disable automatic function calling. +- */ +- disable?: boolean; +- /** If automatic function calling is enabled, +- maximum number of remote calls for automatic function calling. +- This number should be a positive integer. +- If not set, SDK will set maximum number of remote calls to 10. +- */ +- maximumRemoteCalls?: number; +- /** If automatic function calling is enabled, +- whether to ignore call history to the response. +- If not set, SDK will set ignore_call_history to false, +- and will append the call history to +- GenerateContentResponse.automatic_function_calling_history. +- */ +- ignoreCallHistory?: boolean; +-} +- +-/** +- * @license +- * Copyright 2025 Google LLC +- * SPDX-License-Identifier: Apache-2.0 +- */ +-declare class BaseModule { +-} +- +-/** +- * Parameters for setting the base URLs for the Gemini API and Vertex AI API. +- */ +-export declare interface BaseUrlParameters { +- geminiUrl?: string; +- vertexUrl?: string; +-} +- +-/** Defines the function behavior. Defaults to `BLOCKING`. */ +-export declare enum Behavior { +- /** +- * This value is unused. +- */ +- UNSPECIFIED = "UNSPECIFIED", +- /** +- * If set, the system will wait to receive the function response before continuing the conversation. +- */ +- BLOCKING = "BLOCKING", +- /** +- * If set, the system will not wait to receive the function response. Instead, it will attempt to handle function responses as they become available while maintaining the conversation between the user and the model. +- */ +- NON_BLOCKING = "NON_BLOCKING" +-} +- +-/** Content blob. */ +-declare interface Blob_2 { +- /** Optional. Display name of the blob. Used to provide a label or filename to distinguish blobs. This field is not currently used in the Gemini GenerateContent calls. */ +- displayName?: string; +- /** Required. Raw bytes. */ +- data?: string; +- /** Required. The IANA standard MIME type of the source data. */ +- mimeType?: string; +-} +-export { Blob_2 as Blob } +- +-export declare type BlobImageUnion = Blob_2; +- +-/** Output only. Blocked reason. */ +-export declare enum BlockedReason { +- /** +- * Unspecified blocked reason. +- */ +- BLOCKED_REASON_UNSPECIFIED = "BLOCKED_REASON_UNSPECIFIED", +- /** +- * Candidates blocked due to safety. +- */ +- SAFETY = "SAFETY", +- /** +- * Candidates blocked due to other reason. +- */ +- OTHER = "OTHER", +- /** +- * Candidates blocked due to the terms which are included from the terminology blocklist. +- */ +- BLOCKLIST = "BLOCKLIST", +- /** +- * Candidates blocked due to prohibited content. +- */ +- PROHIBITED_CONTENT = "PROHIBITED_CONTENT" +-} +- +-/** A resource used in LLM queries for users to explicitly specify what to cache. */ +-export declare interface CachedContent { +- /** The server-generated resource name of the cached content. */ +- name?: string; +- /** The user-generated meaningful display name of the cached content. */ +- displayName?: string; +- /** The name of the publisher model to use for cached content. */ +- model?: string; +- /** Creation time of the cache entry. */ +- createTime?: string; +- /** When the cache entry was last updated in UTC time. */ +- updateTime?: string; +- /** Expiration time of the cached content. */ +- expireTime?: string; +- /** Metadata on the usage of the cached content. */ +- usageMetadata?: CachedContentUsageMetadata; +-} +- +-/** Metadata on the usage of the cached content. */ +-export declare interface CachedContentUsageMetadata { +- /** Duration of audio in seconds. */ +- audioDurationSeconds?: number; +- /** Number of images. */ +- imageCount?: number; +- /** Number of text characters. */ +- textCount?: number; +- /** Total number of tokens that the cached content consumes. */ +- totalTokenCount?: number; +- /** Duration of video in seconds. */ +- videoDurationSeconds?: number; +-} +- +-export declare class Caches extends BaseModule { +- private readonly apiClient; +- constructor(apiClient: ApiClient); +- /** +- * Lists cached content configurations. +- * +- * @param params - The parameters for the list request. +- * @return The paginated results of the list of cached contents. +- * +- * @example +- * ```ts +- * const cachedContents = await ai.caches.list({config: {'pageSize': 2}}); +- * for (const cachedContent of cachedContents) { +- * console.log(cachedContent); +- * } +- * ``` +- */ +- list: (params?: types.ListCachedContentsParameters) => Promise>; +- /** +- * Creates a cached contents resource. +- * +- * @remarks +- * Context caching is only supported for specific models. See [Gemini +- * Developer API reference](https://ai.google.dev/gemini-api/docs/caching?lang=node/context-cac) +- * and [Vertex AI reference](https://cloud.google.com/vertex-ai/generative-ai/docs/context-cache/context-cache-overview#supported_models) +- * for more information. +- * +- * @param params - The parameters for the create request. +- * @return The created cached content. +- * +- * @example +- * ```ts +- * const contents = ...; // Initialize the content to cache. +- * const response = await ai.caches.create({ +- * model: 'gemini-2.0-flash-001', +- * config: { +- * 'contents': contents, +- * 'displayName': 'test cache', +- * 'systemInstruction': 'What is the sum of the two pdfs?', +- * 'ttl': '86400s', +- * } +- * }); +- * ``` +- */ +- create(params: types.CreateCachedContentParameters): Promise; +- /** +- * Gets cached content configurations. +- * +- * @param params - The parameters for the get request. +- * @return The cached content. +- * +- * @example +- * ```ts +- * await ai.caches.get({name: '...'}); // The server-generated resource name. +- * ``` +- */ +- get(params: types.GetCachedContentParameters): Promise; +- /** +- * Deletes cached content. +- * +- * @param params - The parameters for the delete request. +- * @return The empty response returned by the API. +- * +- * @example +- * ```ts +- * await ai.caches.delete({name: '...'}); // The server-generated resource name. +- * ``` +- */ +- delete(params: types.DeleteCachedContentParameters): Promise; +- /** +- * Updates cached content configurations. +- * +- * @param params - The parameters for the update request. +- * @return The updated cached content. +- * +- * @example +- * ```ts +- * const response = await ai.caches.update({ +- * name: '...', // The server-generated resource name. +- * config: {'ttl': '7600s'} +- * }); +- * ``` +- */ +- update(params: types.UpdateCachedContentParameters): Promise; +- private listInternal; +-} +- +-/** +- * CallableTool is an invokable tool that can be executed with external +- * application (e.g., via Model Context Protocol) or local functions with +- * function calling. +- */ +-export declare interface CallableTool { +- /** +- * Returns tool that can be called by Gemini. +- */ +- tool(): Promise; +- /** +- * Executes the callable tool with the given function call arguments and +- * returns the response parts from the tool execution. +- */ +- callTool(functionCalls: FunctionCall[]): Promise; +-} +- +-/** +- * CallableToolConfig is the configuration for a callable tool. +- */ +-export declare interface CallableToolConfig { +- /** +- * Specifies the model's behavior after invoking this tool. +- */ +- behavior?: Behavior; +-} +- +-/** A response candidate generated from the model. */ +-export declare interface Candidate { +- /** Contains the multi-part content of the response. +- */ +- content?: Content; +- /** Source attribution of the generated content. +- */ +- citationMetadata?: CitationMetadata; +- /** Describes the reason the model stopped generating tokens. +- */ +- finishMessage?: string; +- /** Number of tokens for this candidate. +- */ +- tokenCount?: number; +- /** The reason why the model stopped generating tokens. +- If empty, the model has not stopped generating the tokens. +- */ +- finishReason?: FinishReason; +- /** Metadata related to url context retrieval tool. */ +- urlContextMetadata?: UrlContextMetadata; +- /** Output only. Average log probability score of the candidate. */ +- avgLogprobs?: number; +- /** Output only. Metadata specifies sources used to ground generated content. */ +- groundingMetadata?: GroundingMetadata; +- /** Output only. Index of the candidate. */ +- index?: number; +- /** Output only. Log-likelihood scores for the response tokens and top tokens */ +- logprobsResult?: LogprobsResult; +- /** Output only. List of ratings for the safety of a response candidate. There is at most one rating per category. */ +- safetyRatings?: SafetyRating[]; +-} +- +-/** +- * Chat session that enables sending messages to the model with previous +- * conversation context. +- * +- * @remarks +- * The session maintains all the turns between user and model. +- */ +-export declare class Chat { +- private readonly apiClient; +- private readonly modelsModule; +- private readonly model; +- private readonly config; +- private history; +- private sendPromise; +- constructor(apiClient: ApiClient, modelsModule: Models, model: string, config?: types.GenerateContentConfig, history?: types.Content[]); +- /** +- * Sends a message to the model and returns the response. +- * +- * @remarks +- * This method will wait for the previous message to be processed before +- * sending the next message. +- * +- * @see {@link Chat#sendMessageStream} for streaming method. +- * @param params - parameters for sending messages within a chat session. +- * @returns The model's response. +- * +- * @example +- * ```ts +- * const chat = ai.chats.create({model: 'gemini-2.0-flash'}); +- * const response = await chat.sendMessage({ +- * message: 'Why is the sky blue?' +- * }); +- * console.log(response.text); +- * ``` +- */ +- sendMessage(params: types.SendMessageParameters): Promise; +- /** +- * Sends a message to the model and returns the response in chunks. +- * +- * @remarks +- * This method will wait for the previous message to be processed before +- * sending the next message. +- * +- * @see {@link Chat#sendMessage} for non-streaming method. +- * @param params - parameters for sending the message. +- * @return The model's response. +- * +- * @example +- * ```ts +- * const chat = ai.chats.create({model: 'gemini-2.0-flash'}); +- * const response = await chat.sendMessageStream({ +- * message: 'Why is the sky blue?' +- * }); +- * for await (const chunk of response) { +- * console.log(chunk.text); +- * } +- * ``` +- */ +- sendMessageStream(params: types.SendMessageParameters): Promise>; +- /** +- * Returns the chat history. +- * +- * @remarks +- * The history is a list of contents alternating between user and model. +- * +- * There are two types of history: +- * - The `curated history` contains only the valid turns between user and +- * model, which will be included in the subsequent requests sent to the model. +- * - The `comprehensive history` contains all turns, including invalid or +- * empty model outputs, providing a complete record of the history. +- * +- * The history is updated after receiving the response from the model, +- * for streaming response, it means receiving the last chunk of the response. +- * +- * The `comprehensive history` is returned by default. To get the `curated +- * history`, set the `curated` parameter to `true`. +- * +- * @param curated - whether to return the curated history or the comprehensive +- * history. +- * @return History contents alternating between user and model for the entire +- * chat session. +- */ +- getHistory(curated?: boolean): types.Content[]; +- private processStreamResponse; +- private recordHistory; +-} +- +-/** +- * A utility class to create a chat session. +- */ +-export declare class Chats { +- private readonly modelsModule; +- private readonly apiClient; +- constructor(modelsModule: Models, apiClient: ApiClient); +- /** +- * Creates a new chat session. +- * +- * @remarks +- * The config in the params will be used for all requests within the chat +- * session unless overridden by a per-request `config` in +- * @see {@link types.SendMessageParameters#config}. +- * +- * @param params - Parameters for creating a chat session. +- * @returns A new chat session. +- * +- * @example +- * ```ts +- * const chat = ai.chats.create({ +- * model: 'gemini-2.0-flash' +- * config: { +- * temperature: 0.5, +- * maxOutputTokens: 1024, +- * } +- * }); +- * ``` +- */ +- create(params: types.CreateChatParameters): Chat; +-} +- +-/** Describes the machine learning model version checkpoint. */ +-export declare interface Checkpoint { +- /** The ID of the checkpoint. +- */ +- checkpointId?: string; +- /** The epoch of the checkpoint. +- */ +- epoch?: string; +- /** The step of the checkpoint. +- */ +- step?: string; +-} +- +-/** Source attributions for content. */ +-export declare interface Citation { +- /** Output only. End index into the content. */ +- endIndex?: number; +- /** Output only. License of the attribution. */ +- license?: string; +- /** Output only. Publication date of the attribution. */ +- publicationDate?: GoogleTypeDate; +- /** Output only. Start index into the content. */ +- startIndex?: number; +- /** Output only. Title of the attribution. */ +- title?: string; +- /** Output only. Url reference of the attribution. */ +- uri?: string; +-} +- +-/** Citation information when the model quotes another source. */ +-export declare interface CitationMetadata { +- /** Contains citation information when the model directly quotes, at +- length, from another source. Can include traditional websites and code +- repositories. +- */ +- citations?: Citation[]; +-} +- +-/** Result of executing the [ExecutableCode]. Always follows a `part` containing the [ExecutableCode]. */ +-export declare interface CodeExecutionResult { +- /** Required. Outcome of the code execution. */ +- outcome?: Outcome; +- /** Optional. Contains stdout when code execution is successful, stderr or other description otherwise. */ +- output?: string; +-} +- +-/** Optional parameters for computing tokens. */ +-export declare interface ComputeTokensConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +-} +- +-/** Parameters for computing tokens. */ +-export declare interface ComputeTokensParameters { +- /** ID of the model to use. For a list of models, see `Google models +- `_. */ +- model: string; +- /** Input content. */ +- contents: ContentListUnion; +- /** Optional parameters for the request. +- */ +- config?: ComputeTokensConfig; +-} +- +-/** Response for computing tokens. */ +-export declare class ComputeTokensResponse { +- /** Lists of tokens info from the input. A ComputeTokensRequest could have multiple instances with a prompt in each instance. We also need to return lists of tokens info for the request with multiple instances. */ +- tokensInfo?: TokensInfo[]; +-} +- +-/** Contains the multi-part content of a message. */ +-export declare interface Content { +- /** List of parts that constitute a single message. Each part may have +- a different IANA MIME type. */ +- parts?: Part[]; +- /** Optional. The producer of the content. Must be either 'user' or +- 'model'. Useful to set for multi-turn conversations, otherwise can be +- empty. If role is not specified, SDK will determine the role. */ +- role?: string; +-} +- +-/** The embedding generated from an input content. */ +-export declare interface ContentEmbedding { +- /** A list of floats representing an embedding. +- */ +- values?: number[]; +- /** Vertex API only. Statistics of the input text associated with this +- embedding. +- */ +- statistics?: ContentEmbeddingStatistics; +-} +- +-/** Statistics of the input text associated with the result of content embedding. */ +-export declare interface ContentEmbeddingStatistics { +- /** Vertex API only. If the input text was truncated due to having +- a length longer than the allowed maximum input. +- */ +- truncated?: boolean; +- /** Vertex API only. Number of tokens of the input text. +- */ +- tokenCount?: number; +-} +- +-export declare type ContentListUnion = Content | Content[] | PartUnion | PartUnion[]; +- +-export declare type ContentUnion = Content | PartUnion[] | PartUnion; +- +-/** Enables context window compression -- mechanism managing model context window so it does not exceed given length. */ +-export declare interface ContextWindowCompressionConfig { +- /** Number of tokens (before running turn) that triggers context window compression mechanism. */ +- triggerTokens?: string; +- /** Sliding window compression mechanism. */ +- slidingWindow?: SlidingWindow; +-} +- +-/** Configuration for a Control reference image. */ +-export declare interface ControlReferenceConfig { +- /** The type of control reference image to use. */ +- controlType?: ControlReferenceType; +- /** Defaults to False. When set to True, the control image will be +- computed by the model based on the control type. When set to False, +- the control image must be provided by the user. */ +- enableControlImageComputation?: boolean; +-} +- +-/** A control reference image. +- +- The image of the control reference image is either a control image provided +- by the user, or a regular image which the backend will use to generate a +- control image of. In the case of the latter, the +- enable_control_image_computation field in the config should be set to True. +- +- A control image is an image that represents a sketch image of areas for the +- model to fill in based on the prompt. +- */ +-export declare class ControlReferenceImage { +- /** The reference image for the editing operation. */ +- referenceImage?: Image_2; +- /** The id of the reference image. */ +- referenceId?: number; +- /** The type of the reference image. Only set by the SDK. */ +- referenceType?: string; +- /** Configuration for the control reference image. */ +- config?: ControlReferenceConfig; +- /** Internal method to convert to ReferenceImageAPIInternal. */ +- toReferenceImageAPI(): any; +-} +- +-/** Enum representing the control type of a control reference image. */ +-export declare enum ControlReferenceType { +- CONTROL_TYPE_DEFAULT = "CONTROL_TYPE_DEFAULT", +- CONTROL_TYPE_CANNY = "CONTROL_TYPE_CANNY", +- CONTROL_TYPE_SCRIBBLE = "CONTROL_TYPE_SCRIBBLE", +- CONTROL_TYPE_FACE_MESH = "CONTROL_TYPE_FACE_MESH" +-} +- +-/** Config for the count_tokens method. */ +-export declare interface CountTokensConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +- /** Instructions for the model to steer it toward better performance. +- */ +- systemInstruction?: ContentUnion; +- /** Code that enables the system to interact with external systems to +- perform an action outside of the knowledge and scope of the model. +- */ +- tools?: Tool[]; +- /** Configuration that the model uses to generate the response. Not +- supported by the Gemini Developer API. +- */ +- generationConfig?: GenerationConfig; +-} +- +-/** Parameters for counting tokens. */ +-export declare interface CountTokensParameters { +- /** ID of the model to use. For a list of models, see `Google models +- `_. */ +- model: string; +- /** Input content. */ +- contents: ContentListUnion; +- /** Configuration for counting tokens. */ +- config?: CountTokensConfig; +-} +- +-/** Response for counting tokens. */ +-export declare class CountTokensResponse { +- /** Total number of tokens. */ +- totalTokens?: number; +- /** Number of tokens in the cached part of the prompt (the cached content). */ +- cachedContentTokenCount?: number; +-} +- +-/** Optional parameters. */ +-export declare interface CreateAuthTokenConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +- /** An optional time after which, when using the resulting token, +- messages in Live API sessions will be rejected. (Gemini may +- preemptively close the session after this time.) +- +- If not set then this defaults to 30 minutes in the future. If set, this +- value must be less than 20 hours in the future. */ +- expireTime?: string; +- /** The time after which new Live API sessions using the token +- resulting from this request will be rejected. +- +- If not set this defaults to 60 seconds in the future. If set, this value +- must be less than 20 hours in the future. */ +- newSessionExpireTime?: string; +- /** The number of times the token can be used. If this value is zero +- then no limit is applied. Default is 1. Resuming a Live API session does +- not count as a use. */ +- uses?: number; +- /** Configuration specific to Live API connections created using this token. */ +- liveEphemeralParameters?: LiveEphemeralParameters; +- /** Additional fields to lock in the effective LiveConnectParameters. */ +- lockAdditionalFields?: string[]; +-} +- +-/** Optional configuration for cached content creation. */ +-export declare interface CreateCachedContentConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +- /** The TTL for this resource. The expiration time is computed: now + TTL. It is a duration string, with up to nine fractional digits, terminated by 's'. Example: "3.5s". */ +- ttl?: string; +- /** Timestamp of when this resource is considered expired. Uses RFC 3339 format, Example: 2014-10-02T15:01:23Z. */ +- expireTime?: string; +- /** The user-generated meaningful display name of the cached content. +- */ +- displayName?: string; +- /** The content to cache. +- */ +- contents?: ContentListUnion; +- /** Developer set system instruction. +- */ +- systemInstruction?: ContentUnion; +- /** A list of `Tools` the model may use to generate the next response. +- */ +- tools?: Tool[]; +- /** Configuration for the tools to use. This config is shared for all tools. +- */ +- toolConfig?: ToolConfig; +- /** The Cloud KMS resource identifier of the customer managed +- encryption key used to protect a resource. +- The key needs to be in the same region as where the compute resource is +- created. See +- https://cloud.google.com/vertex-ai/docs/general/cmek for more +- details. If this is set, then all created CachedContent objects +- will be encrypted with the provided encryption key. +- Allowed formats: projects/{project}/locations/{location}/keyRings/{key_ring}/cryptoKeys/{crypto_key} +- */ +- kmsKeyName?: string; +-} +- +-/** Parameters for caches.create method. */ +-export declare interface CreateCachedContentParameters { +- /** ID of the model to use. Example: gemini-2.0-flash */ +- model: string; +- /** Configuration that contains optional parameters. +- */ +- config?: CreateCachedContentConfig; +-} +- +-/** Parameters for initializing a new chat session. +- +- These parameters are used when creating a chat session with the +- `chats.create()` method. +- */ +-export declare interface CreateChatParameters { +- /** The name of the model to use for the chat session. +- +- For example: 'gemini-2.0-flash', 'gemini-2.0-flash-lite', etc. See Gemini API +- docs to find the available models. +- */ +- model: string; +- /** Config for the entire chat session. +- +- This config applies to all requests within the session +- unless overridden by a per-request `config` in `SendMessageParameters`. +- */ +- config?: GenerateContentConfig; +- /** The initial conversation history for the chat session. +- +- This allows you to start the chat with a pre-existing history. The history +- must be a list of `Content` alternating between 'user' and 'model' roles. +- It should start with a 'user' message. +- */ +- history?: Content[]; +-} +- +-/** Used to override the default configuration. */ +-export declare interface CreateFileConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +-} +- +-/** Generates the parameters for the private _create method. */ +-export declare interface CreateFileParameters { +- /** The file to be uploaded. +- mime_type: (Required) The MIME type of the file. Must be provided. +- name: (Optional) The name of the file in the destination (e.g. +- 'files/sample-image'). +- display_name: (Optional) The display name of the file. +- */ +- file: File_2; +- /** Used to override the default configuration. */ +- config?: CreateFileConfig; +-} +- +-/** Response for the create file method. */ +-export declare class CreateFileResponse { +- /** Used to retain the full HTTP response. */ +- sdkHttpResponse?: HttpResponse; +-} +- +-/** +- * Creates a `Content` object with a model role from a `PartListUnion` object or `string`. +- */ +-export declare function createModelContent(partOrString: PartListUnion | string): Content; +- +-/** +- * Creates a `Part` object from a `base64` encoded `string`. +- */ +-export declare function createPartFromBase64(data: string, mimeType: string): Part; +- +-/** +- * Creates a `Part` object from the `outcome` and `output` of a `CodeExecutionResult` object. +- */ +-export declare function createPartFromCodeExecutionResult(outcome: Outcome, output: string): Part; +- +-/** +- * Creates a `Part` object from the `code` and `language` of an `ExecutableCode` object. +- */ +-export declare function createPartFromExecutableCode(code: string, language: Language): Part; +- +-/** +- * Creates a `Part` object from a `FunctionCall` object. +- */ +-export declare function createPartFromFunctionCall(name: string, args: Record): Part; +- +-/** +- * Creates a `Part` object from a `FunctionResponse` object. +- */ +-export declare function createPartFromFunctionResponse(id: string, name: string, response: Record): Part; +- +-/** +- * Creates a `Part` object from a `text` string. +- */ +-export declare function createPartFromText(text: string): Part; +- +-/** +- * Creates a `Part` object from a `URI` string. +- */ +-export declare function createPartFromUri(uri: string, mimeType: string): Part; +- +-/** Supervised fine-tuning job creation request - optional fields. */ +-export declare interface CreateTuningJobConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +- /** Cloud Storage path to file containing training dataset for tuning. The dataset must be formatted as a JSONL file. */ +- validationDataset?: TuningValidationDataset; +- /** The display name of the tuned Model. The name can be up to 128 characters long and can consist of any UTF-8 characters. */ +- tunedModelDisplayName?: string; +- /** The description of the TuningJob */ +- description?: string; +- /** Number of complete passes the model makes over the entire training dataset during training. */ +- epochCount?: number; +- /** Multiplier for adjusting the default learning rate. */ +- learningRateMultiplier?: number; +- /** If set to true, disable intermediate checkpoints for SFT and only the last checkpoint will be exported. Otherwise, enable intermediate checkpoints for SFT. */ +- exportLastCheckpointOnly?: boolean; +- /** Adapter size for tuning. */ +- adapterSize?: AdapterSize; +- /** The batch size hyperparameter for tuning. If not set, a default of 4 or 16 will be used based on the number of training examples. */ +- batchSize?: number; +- /** The learning rate hyperparameter for tuning. If not set, a default of 0.001 or 0.0002 will be calculated based on the number of training examples. */ +- learningRate?: number; +-} +- +-/** Supervised fine-tuning job creation parameters - optional fields. */ +-export declare interface CreateTuningJobParameters { +- /** The base model that is being tuned, e.g., "gemini-1.0-pro-002". */ +- baseModel: string; +- /** Cloud Storage path to file containing training dataset for tuning. The dataset must be formatted as a JSONL file. */ +- trainingDataset: TuningDataset; +- /** Configuration for the tuning job. */ +- config?: CreateTuningJobConfig; +-} +- +-/** +- * Creates a `Content` object with a user role from a `PartListUnion` object or `string`. +- */ +-export declare function createUserContent(partOrString: PartListUnion | string): Content; +- +-/** Distribution computed over a tuning dataset. */ +-export declare interface DatasetDistribution { +- /** Output only. Defines the histogram bucket. */ +- buckets?: DatasetDistributionDistributionBucket[]; +- /** Output only. The maximum of the population values. */ +- max?: number; +- /** Output only. The arithmetic mean of the values in the population. */ +- mean?: number; +- /** Output only. The median of the values in the population. */ +- median?: number; +- /** Output only. The minimum of the population values. */ +- min?: number; +- /** Output only. The 5th percentile of the values in the population. */ +- p5?: number; +- /** Output only. The 95th percentile of the values in the population. */ +- p95?: number; +- /** Output only. Sum of a given population of values. */ +- sum?: number; +-} +- +-/** Dataset bucket used to create a histogram for the distribution given a population of values. */ +-export declare interface DatasetDistributionDistributionBucket { +- /** Output only. Number of values in the bucket. */ +- count?: string; +- /** Output only. Left bound of the bucket. */ +- left?: number; +- /** Output only. Right bound of the bucket. */ +- right?: number; +-} +- +-/** Statistics computed over a tuning dataset. */ +-export declare interface DatasetStats { +- /** Output only. Number of billable characters in the tuning dataset. */ +- totalBillableCharacterCount?: string; +- /** Output only. Number of tuning characters in the tuning dataset. */ +- totalTuningCharacterCount?: string; +- /** Output only. Number of examples in the tuning dataset. */ +- tuningDatasetExampleCount?: string; +- /** Output only. Number of tuning steps for this Tuning Job. */ +- tuningStepCount?: string; +- /** Output only. Sample user messages in the training dataset uri. */ +- userDatasetExamples?: Content[]; +- /** Output only. Dataset distributions for the user input tokens. */ +- userInputTokenDistribution?: DatasetDistribution; +- /** Output only. Dataset distributions for the messages per example. */ +- userMessagePerExampleDistribution?: DatasetDistribution; +- /** Output only. Dataset distributions for the user output tokens. */ +- userOutputTokenDistribution?: DatasetDistribution; +-} +- +-/** Optional parameters for caches.delete method. */ +-export declare interface DeleteCachedContentConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +-} +- +-/** Parameters for caches.delete method. */ +-export declare interface DeleteCachedContentParameters { +- /** The server-generated resource name of the cached content. +- */ +- name: string; +- /** Optional parameters for the request. +- */ +- config?: DeleteCachedContentConfig; +-} +- +-/** Empty response for caches.delete method. */ +-export declare class DeleteCachedContentResponse { +-} +- +-/** Used to override the default configuration. */ +-export declare interface DeleteFileConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +-} +- +-/** Generates the parameters for the get method. */ +-export declare interface DeleteFileParameters { +- /** The name identifier for the file to be deleted. */ +- name: string; +- /** Used to override the default configuration. */ +- config?: DeleteFileConfig; +-} +- +-/** Response for the delete file method. */ +-export declare class DeleteFileResponse { +-} +- +-/** Configuration for deleting a tuned model. */ +-export declare interface DeleteModelConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +-} +- +-/** Parameters for deleting a tuned model. */ +-export declare interface DeleteModelParameters { +- model: string; +- /** Optional parameters for the request. */ +- config?: DeleteModelConfig; +-} +- +-export declare class DeleteModelResponse { +-} +- +-/** Statistics computed for datasets used for distillation. */ +-export declare interface DistillationDataStats { +- /** Output only. Statistics computed for the training dataset. */ +- trainingDatasetStats?: DatasetStats; +-} +- +-/** Hyperparameters for Distillation. */ +-export declare interface DistillationHyperParameters { +- /** Optional. Adapter size for distillation. */ +- adapterSize?: AdapterSize; +- /** Optional. Number of complete passes the model makes over the entire training dataset during training. */ +- epochCount?: string; +- /** Optional. Multiplier for adjusting the default learning rate. */ +- learningRateMultiplier?: number; +-} +- +-/** Tuning Spec for Distillation. */ +-export declare interface DistillationSpec { +- /** The base teacher model that is being distilled, e.g., "gemini-1.0-pro-002". */ +- baseTeacherModel?: string; +- /** Optional. Hyperparameters for Distillation. */ +- hyperParameters?: DistillationHyperParameters; +- /** Required. A path in a Cloud Storage bucket, which will be treated as the root output directory of the distillation pipeline. It is used by the system to generate the paths of output artifacts. */ +- pipelineRootDirectory?: string; +- /** The student model that is being tuned, e.g., "google/gemma-2b-1.1-it". */ +- studentModel?: string; +- /** Required. Cloud Storage path to file containing training dataset for tuning. The dataset must be formatted as a JSONL file. */ +- trainingDatasetUri?: string; +- /** The resource name of the Tuned teacher model. Format: `projects/{project}/locations/{location}/models/{model}`. */ +- tunedTeacherModelSource?: string; +- /** Optional. Cloud Storage path to file containing validation dataset for tuning. The dataset must be formatted as a JSONL file. */ +- validationDatasetUri?: string; +-} +- +-export declare type DownloadableFileUnion = string | File_2 | GeneratedVideo | Video; +- +-declare interface Downloader { +- /** +- * Downloads a file to the given location. +- * +- * @param params The parameters for downloading the file. +- * @param apiClient The ApiClient to use for uploading. +- * @return A Promises that resolves when the download is complete. +- */ +- download(params: DownloadFileParameters, apiClient: ApiClient): Promise; +-} +- +-/** Used to override the default configuration. */ +-export declare interface DownloadFileConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +-} +- +-/** Parameters used to download a file. */ +-export declare interface DownloadFileParameters { +- /** The file to download. It can be a file name, a file object or a generated video. */ +- file: DownloadableFileUnion; +- /** Location where the file should be downloaded to. */ +- downloadPath: string; +- /** Configuration to for the download operation. */ +- config?: DownloadFileConfig; +-} +- +-/** Describes the options to customize dynamic retrieval. */ +-export declare interface DynamicRetrievalConfig { +- /** The mode of the predictor to be used in dynamic retrieval. */ +- mode?: DynamicRetrievalConfigMode; +- /** Optional. The threshold to be used in dynamic retrieval. If not set, a system default value is used. */ +- dynamicThreshold?: number; +-} +- +-/** Config for the dynamic retrieval config mode. */ +-export declare enum DynamicRetrievalConfigMode { +- /** +- * Always trigger retrieval. +- */ +- MODE_UNSPECIFIED = "MODE_UNSPECIFIED", +- /** +- * Run retrieval only when system decides it is necessary. +- */ +- MODE_DYNAMIC = "MODE_DYNAMIC" +-} +- +-/** Configuration for editing an image. */ +-export declare interface EditImageConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +- /** Cloud Storage URI used to store the generated images. +- */ +- outputGcsUri?: string; +- /** Description of what to discourage in the generated images. +- */ +- negativePrompt?: string; +- /** Number of images to generate. +- */ +- numberOfImages?: number; +- /** Aspect ratio of the generated images. +- */ +- aspectRatio?: string; +- /** Controls how much the model adheres to the text prompt. Large +- values increase output and prompt alignment, but may compromise image +- quality. +- */ +- guidanceScale?: number; +- /** Random seed for image generation. This is not available when +- ``add_watermark`` is set to true. +- */ +- seed?: number; +- /** Filter level for safety filtering. +- */ +- safetyFilterLevel?: SafetyFilterLevel; +- /** Allows generation of people by the model. +- */ +- personGeneration?: PersonGeneration; +- /** Whether to report the safety scores of each generated image and +- the positive prompt in the response. +- */ +- includeSafetyAttributes?: boolean; +- /** Whether to include the Responsible AI filter reason if the image +- is filtered out of the response. +- */ +- includeRaiReason?: boolean; +- /** Language of the text in the prompt. +- */ +- language?: ImagePromptLanguage; +- /** MIME type of the generated image. +- */ +- outputMimeType?: string; +- /** Compression quality of the generated image (for ``image/jpeg`` +- only). +- */ +- outputCompressionQuality?: number; +- /** Describes the editing mode for the request. */ +- editMode?: EditMode; +- /** The number of sampling steps. A higher value has better image +- quality, while a lower value has better latency. */ +- baseSteps?: number; +-} +- +-/** Parameters for the request to edit an image. */ +-export declare interface EditImageParameters { +- /** The model to use. */ +- model: string; +- /** A text description of the edit to apply to the image. */ +- prompt: string; +- /** The reference images for Imagen 3 editing. */ +- referenceImages: ReferenceImage[]; +- /** Configuration for editing. */ +- config?: EditImageConfig; +-} +- +-/** Response for the request to edit an image. */ +-export declare class EditImageResponse { +- /** Generated images. */ +- generatedImages?: GeneratedImage[]; +-} +- +-/** Enum representing the Imagen 3 Edit mode. */ +-export declare enum EditMode { +- EDIT_MODE_DEFAULT = "EDIT_MODE_DEFAULT", +- EDIT_MODE_INPAINT_REMOVAL = "EDIT_MODE_INPAINT_REMOVAL", +- EDIT_MODE_INPAINT_INSERTION = "EDIT_MODE_INPAINT_INSERTION", +- EDIT_MODE_OUTPAINT = "EDIT_MODE_OUTPAINT", +- EDIT_MODE_CONTROLLED_EDITING = "EDIT_MODE_CONTROLLED_EDITING", +- EDIT_MODE_STYLE = "EDIT_MODE_STYLE", +- EDIT_MODE_BGSWAP = "EDIT_MODE_BGSWAP", +- EDIT_MODE_PRODUCT_IMAGE = "EDIT_MODE_PRODUCT_IMAGE" +-} +- +-/** Optional parameters for the embed_content method. */ +-export declare interface EmbedContentConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +- /** Type of task for which the embedding will be used. +- */ +- taskType?: string; +- /** Title for the text. Only applicable when TaskType is +- `RETRIEVAL_DOCUMENT`. +- */ +- title?: string; +- /** Reduced dimension for the output embedding. If set, +- excessive values in the output embedding are truncated from the end. +- Supported by newer models since 2024 only. You cannot set this value if +- using the earlier model (`models/embedding-001`). +- */ +- outputDimensionality?: number; +- /** Vertex API only. The MIME type of the input. +- */ +- mimeType?: string; +- /** Vertex API only. Whether to silently truncate inputs longer than +- the max sequence length. If this option is set to false, oversized inputs +- will lead to an INVALID_ARGUMENT error, similar to other text APIs. +- */ +- autoTruncate?: boolean; +-} +- +-/** Request-level metadata for the Vertex Embed Content API. */ +-export declare interface EmbedContentMetadata { +- /** Vertex API only. The total number of billable characters included +- in the request. +- */ +- billableCharacterCount?: number; +-} +- +-/** Parameters for the embed_content method. */ +-export declare interface EmbedContentParameters { +- /** ID of the model to use. For a list of models, see `Google models +- `_. */ +- model: string; +- /** The content to embed. Only the `parts.text` fields will be counted. +- */ +- contents: ContentListUnion; +- /** Configuration that contains optional parameters. +- */ +- config?: EmbedContentConfig; +-} +- +-/** Response for the embed_content method. */ +-export declare class EmbedContentResponse { +- /** The embeddings for each request, in the same order as provided in +- the batch request. +- */ +- embeddings?: ContentEmbedding[]; +- /** Vertex API only. Metadata about the request. +- */ +- metadata?: EmbedContentMetadata; +-} +- +-/** Represents a customer-managed encryption key spec that can be applied to a top-level resource. */ +-export declare interface EncryptionSpec { +- /** Required. The Cloud KMS resource identifier of the customer managed encryption key used to protect a resource. Has the form: `projects/my-project/locations/my-region/keyRings/my-kr/cryptoKeys/my-key`. The key needs to be in the same region as where the compute resource is created. */ +- kmsKeyName?: string; +-} +- +-/** An endpoint where you deploy models. */ +-export declare interface Endpoint { +- /** Resource name of the endpoint. */ +- name?: string; +- /** ID of the model that's deployed to the endpoint. */ +- deployedModelId?: string; +-} +- +-/** End of speech sensitivity. */ +-export declare enum EndSensitivity { +- /** +- * The default is END_SENSITIVITY_LOW. +- */ +- END_SENSITIVITY_UNSPECIFIED = "END_SENSITIVITY_UNSPECIFIED", +- /** +- * Automatic detection ends speech more often. +- */ +- END_SENSITIVITY_HIGH = "END_SENSITIVITY_HIGH", +- /** +- * Automatic detection ends speech less often. +- */ +- END_SENSITIVITY_LOW = "END_SENSITIVITY_LOW" +-} +- +-/** Tool to search public web data, powered by Vertex AI Search and Sec4 compliance. */ +-export declare interface EnterpriseWebSearch { +-} +- +-/** Code generated by the model that is meant to be executed, and the result returned to the model. Generated when using the [FunctionDeclaration] tool and [FunctionCallingConfig] mode is set to [Mode.CODE]. */ +-export declare interface ExecutableCode { +- /** Required. The code to be executed. */ +- code?: string; +- /** Required. Programming language of the `code`. */ +- language?: Language; +-} +- +-/** Options for feature selection preference. */ +-export declare enum FeatureSelectionPreference { +- FEATURE_SELECTION_PREFERENCE_UNSPECIFIED = "FEATURE_SELECTION_PREFERENCE_UNSPECIFIED", +- PRIORITIZE_QUALITY = "PRIORITIZE_QUALITY", +- BALANCED = "BALANCED", +- PRIORITIZE_COST = "PRIORITIZE_COST" +-} +- +-export declare interface FetchPredictOperationConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +-} +- +-/** Parameters for the fetchPredictOperation method. */ +-export declare interface FetchPredictOperationParameters { +- /** The server-assigned name for the operation. */ +- operationName: string; +- resourceName: string; +- /** Used to override the default configuration. */ +- config?: FetchPredictOperationConfig; +-} +- +-/** A file uploaded to the API. */ +-declare interface File_2 { +- /** The `File` resource name. The ID (name excluding the "files/" prefix) can contain up to 40 characters that are lowercase alphanumeric or dashes (-). The ID cannot start or end with a dash. If the name is empty on create, a unique name will be generated. Example: `files/123-456` */ +- name?: string; +- /** Optional. The human-readable display name for the `File`. The display name must be no more than 512 characters in length, including spaces. Example: 'Welcome Image' */ +- displayName?: string; +- /** Output only. MIME type of the file. */ +- mimeType?: string; +- /** Output only. Size of the file in bytes. */ +- sizeBytes?: string; +- /** Output only. The timestamp of when the `File` was created. */ +- createTime?: string; +- /** Output only. The timestamp of when the `File` will be deleted. Only set if the `File` is scheduled to expire. */ +- expirationTime?: string; +- /** Output only. The timestamp of when the `File` was last updated. */ +- updateTime?: string; +- /** Output only. SHA-256 hash of the uploaded bytes. The hash value is encoded in base64 format. */ +- sha256Hash?: string; +- /** Output only. The URI of the `File`. */ +- uri?: string; +- /** Output only. The URI of the `File`, only set for downloadable (generated) files. */ +- downloadUri?: string; +- /** Output only. Processing state of the File. */ +- state?: FileState; +- /** Output only. The source of the `File`. */ +- source?: FileSource; +- /** Output only. Metadata for a video. */ +- videoMetadata?: Record; +- /** Output only. Error status if File processing failed. */ +- error?: FileStatus; +-} +-export { File_2 as File } +- +-/** URI based data. */ +-export declare interface FileData { +- /** Required. URI. */ +- fileUri?: string; +- /** Required. The IANA standard MIME type of the source data. */ +- mimeType?: string; +-} +- +-export declare class Files extends BaseModule { +- private readonly apiClient; +- constructor(apiClient: ApiClient); +- /** +- * Lists all current project files from the service. +- * +- * @param params - The parameters for the list request +- * @return The paginated results of the list of files +- * +- * @example +- * The following code prints the names of all files from the service, the +- * size of each page is 10. +- * +- * ```ts +- * const listResponse = await ai.files.list({config: {'pageSize': 10}}); +- * for await (const file of listResponse) { +- * console.log(file.name); +- * } +- * ``` +- */ +- list: (params?: types.ListFilesParameters) => Promise>; +- /** +- * Uploads a file asynchronously to the Gemini API. +- * This method is not available in Vertex AI. +- * Supported upload sources: +- * - Node.js: File path (string) or Blob object. +- * - Browser: Blob object (e.g., File). +- * +- * @remarks +- * The `mimeType` can be specified in the `config` parameter. If omitted: +- * - For file path (string) inputs, the `mimeType` will be inferred from the +- * file extension. +- * - For Blob object inputs, the `mimeType` will be set to the Blob's `type` +- * property. +- * Somex eamples for file extension to mimeType mapping: +- * .txt -> text/plain +- * .json -> application/json +- * .jpg -> image/jpeg +- * .png -> image/png +- * .mp3 -> audio/mpeg +- * .mp4 -> video/mp4 +- * +- * This section can contain multiple paragraphs and code examples. +- * +- * @param params - Optional parameters specified in the +- * `types.UploadFileParameters` interface. +- * @see {@link types.UploadFileParameters#config} for the optional +- * config in the parameters. +- * @return A promise that resolves to a `types.File` object. +- * @throws An error if called on a Vertex AI client. +- * @throws An error if the `mimeType` is not provided and can not be inferred, +- * the `mimeType` can be provided in the `params.config` parameter. +- * @throws An error occurs if a suitable upload location cannot be established. +- * +- * @example +- * The following code uploads a file to Gemini API. +- * +- * ```ts +- * const file = await ai.files.upload({file: 'file.txt', config: { +- * mimeType: 'text/plain', +- * }}); +- * console.log(file.name); +- * ``` +- */ +- upload(params: types.UploadFileParameters): Promise; +- /** +- * Downloads a remotely stored file asynchronously to a location specified in +- * the `params` object. This method only works on Node environment, to +- * download files in the browser, use a browser compliant method like an +- * tag. +- * +- * @param params - The parameters for the download request. +- * +- * @example +- * The following code downloads an example file named "files/mehozpxf877d" as +- * "file.txt". +- * +- * ```ts +- * await ai.files.download({file: file.name, downloadPath: 'file.txt'}); +- * ``` +- */ +- download(params: types.DownloadFileParameters): Promise; +- private listInternal; +- private createInternal; +- /** +- * Retrieves the file information from the service. +- * +- * @param params - The parameters for the get request +- * @return The Promise that resolves to the types.File object requested. +- * +- * @example +- * ```ts +- * const config: GetFileParameters = { +- * name: fileName, +- * }; +- * file = await ai.files.get(config); +- * console.log(file.name); +- * ``` +- */ +- get(params: types.GetFileParameters): Promise; +- /** +- * Deletes a remotely stored file. +- * +- * @param params - The parameters for the delete request. +- * @return The DeleteFileResponse, the response for the delete method. +- * +- * @example +- * The following code deletes an example file named "files/mehozpxf877d". +- * +- * ```ts +- * await ai.files.delete({name: file.name}); +- * ``` +- */ +- delete(params: types.DeleteFileParameters): Promise; +-} +- +-/** Source of the File. */ +-export declare enum FileSource { +- SOURCE_UNSPECIFIED = "SOURCE_UNSPECIFIED", +- UPLOADED = "UPLOADED", +- GENERATED = "GENERATED" +-} +- +-/** +- * Represents the size and mimeType of a file. The information is used to +- * request the upload URL from the https://generativelanguage.googleapis.com/upload/v1beta/files endpoint. +- * This interface defines the structure for constructing and executing HTTP +- * requests. +- */ +-declare interface FileStat { +- /** +- * The size of the file in bytes. +- */ +- size: number; +- /** +- * The MIME type of the file. +- */ +- type: string | undefined; +-} +- +-/** State for the lifecycle of a File. */ +-export declare enum FileState { +- STATE_UNSPECIFIED = "STATE_UNSPECIFIED", +- PROCESSING = "PROCESSING", +- ACTIVE = "ACTIVE", +- FAILED = "FAILED" +-} +- +-/** Status of a File that uses a common error model. */ +-export declare interface FileStatus { +- /** A list of messages that carry the error details. There is a common set of message types for APIs to use. */ +- details?: Record[]; +- /** A list of messages that carry the error details. There is a common set of message types for APIs to use. */ +- message?: string; +- /** The status code. 0 for OK, 1 for CANCELLED */ +- code?: number; +-} +- +-/** Output only. The reason why the model stopped generating tokens. +- +- If empty, the model has not stopped generating the tokens. +- */ +-export declare enum FinishReason { +- /** +- * The finish reason is unspecified. +- */ +- FINISH_REASON_UNSPECIFIED = "FINISH_REASON_UNSPECIFIED", +- /** +- * Token generation reached a natural stopping point or a configured stop sequence. +- */ +- STOP = "STOP", +- /** +- * Token generation reached the configured maximum output tokens. +- */ +- MAX_TOKENS = "MAX_TOKENS", +- /** +- * Token generation stopped because the content potentially contains safety violations. NOTE: When streaming, [content][] is empty if content filters blocks the output. +- */ +- SAFETY = "SAFETY", +- /** +- * The token generation stopped because of potential recitation. +- */ +- RECITATION = "RECITATION", +- /** +- * The token generation stopped because of using an unsupported language. +- */ +- LANGUAGE = "LANGUAGE", +- /** +- * All other reasons that stopped the token generation. +- */ +- OTHER = "OTHER", +- /** +- * Token generation stopped because the content contains forbidden terms. +- */ +- BLOCKLIST = "BLOCKLIST", +- /** +- * Token generation stopped for potentially containing prohibited content. +- */ +- PROHIBITED_CONTENT = "PROHIBITED_CONTENT", +- /** +- * Token generation stopped because the content potentially contains Sensitive Personally Identifiable Information (SPII). +- */ +- SPII = "SPII", +- /** +- * The function call generated by the model is invalid. +- */ +- MALFORMED_FUNCTION_CALL = "MALFORMED_FUNCTION_CALL", +- /** +- * Token generation stopped because generated images have safety violations. +- */ +- IMAGE_SAFETY = "IMAGE_SAFETY" +-} +- +-/** A function call. */ +-export declare interface FunctionCall { +- /** The unique id of the function call. If populated, the client to execute the +- `function_call` and return the response with the matching `id`. */ +- id?: string; +- /** Optional. Required. The function parameters and values in JSON object format. See [FunctionDeclaration.parameters] for parameter details. */ +- args?: Record; +- /** Required. The name of the function to call. Matches [FunctionDeclaration.name]. */ +- name?: string; +-} +- +-/** Function calling config. */ +-export declare interface FunctionCallingConfig { +- /** Optional. Function calling mode. */ +- mode?: FunctionCallingConfigMode; +- /** Optional. Function names to call. Only set when the Mode is ANY. Function names should match [FunctionDeclaration.name]. With mode set to ANY, model will predict a function call from the set of function names provided. */ +- allowedFunctionNames?: string[]; +-} +- +-/** Config for the function calling config mode. */ +-export declare enum FunctionCallingConfigMode { +- /** +- * The function calling config mode is unspecified. Should not be used. +- */ +- MODE_UNSPECIFIED = "MODE_UNSPECIFIED", +- /** +- * Default model behavior, model decides to predict either function calls or natural language response. +- */ +- AUTO = "AUTO", +- /** +- * Model is constrained to always predicting function calls only. If "allowed_function_names" are set, the predicted function calls will be limited to any one of "allowed_function_names", else the predicted function calls will be any one of the provided "function_declarations". +- */ +- ANY = "ANY", +- /** +- * Model will not predict any function calls. Model behavior is same as when not passing any function declarations. +- */ +- NONE = "NONE" +-} +- +-/** Defines a function that the model can generate JSON inputs for. +- +- The inputs are based on `OpenAPI 3.0 specifications +- `_. +- */ +-export declare interface FunctionDeclaration { +- /** Defines the function behavior. */ +- behavior?: Behavior; +- /** Optional. Description and purpose of the function. Model uses it to decide how and whether to call the function. */ +- description?: string; +- /** Required. The name of the function to call. Must start with a letter or an underscore. Must be a-z, A-Z, 0-9, or contain underscores, dots and dashes, with a maximum length of 64. */ +- name?: string; +- /** Optional. Describes the parameters to this function in JSON Schema Object format. Reflects the Open API 3.03 Parameter Object. string Key: the name of the parameter. Parameter names are case sensitive. Schema Value: the Schema defining the type used for the parameter. For function with no parameters, this can be left unset. Parameter names must start with a letter or an underscore and must only contain chars a-z, A-Z, 0-9, or underscores with a maximum length of 64. Example with 1 required and 1 optional parameter: type: OBJECT properties: param1: type: STRING param2: type: INTEGER required: - param1 */ +- parameters?: Schema; +- /** Optional. Describes the output from this function in JSON Schema format. Reflects the Open API 3.03 Response Object. The Schema defines the type used for the response value of the function. */ +- response?: Schema; +-} +- +-/** A function response. */ +-export declare class FunctionResponse { +- /** Signals that function call continues, and more responses will be returned, turning the function call into a generator. Is only applicable to NON_BLOCKING function calls (see FunctionDeclaration.behavior for details), ignored otherwise. If false, the default, future responses will not be considered. Is only applicable to NON_BLOCKING function calls, is ignored otherwise. If set to false, future responses will not be considered. It is allowed to return empty `response` with `will_continue=False` to signal that the function call is finished. */ +- willContinue?: boolean; +- /** Specifies how the response should be scheduled in the conversation. Only applicable to NON_BLOCKING function calls, is ignored otherwise. Defaults to WHEN_IDLE. */ +- scheduling?: FunctionResponseScheduling; +- /** Optional. The id of the function call this response is for. Populated by the client to match the corresponding function call `id`. */ +- id?: string; +- /** Required. The name of the function to call. Matches [FunctionDeclaration.name] and [FunctionCall.name]. */ +- name?: string; +- /** Required. The function response in JSON object format. Use "output" key to specify function output and "error" key to specify error details (if any). If "output" and "error" keys are not specified, then whole "response" is treated as function output. */ +- response?: Record; +-} +- +-/** Specifies how the response should be scheduled in the conversation. */ +-export declare enum FunctionResponseScheduling { +- /** +- * This value is unused. +- */ +- SCHEDULING_UNSPECIFIED = "SCHEDULING_UNSPECIFIED", +- /** +- * Only add the result to the conversation context, do not interrupt or trigger generation. +- */ +- SILENT = "SILENT", +- /** +- * Add the result to the conversation context, and prompt to generate output without interrupting ongoing generation. +- */ +- WHEN_IDLE = "WHEN_IDLE", +- /** +- * Add the result to the conversation context, interrupt ongoing generation and prompt to generate output. +- */ +- INTERRUPT = "INTERRUPT" +-} +- +-/** Optional model configuration parameters. +- +- For more information, see `Content generation parameters +- `_. +- */ +-export declare interface GenerateContentConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +- /** Instructions for the model to steer it toward better performance. +- For example, "Answer as concisely as possible" or "Don't use technical +- terms in your response". +- */ +- systemInstruction?: ContentUnion; +- /** Value that controls the degree of randomness in token selection. +- Lower temperatures are good for prompts that require a less open-ended or +- creative response, while higher temperatures can lead to more diverse or +- creative results. +- */ +- temperature?: number; +- /** Tokens are selected from the most to least probable until the sum +- of their probabilities equals this value. Use a lower value for less +- random responses and a higher value for more random responses. +- */ +- topP?: number; +- /** For each token selection step, the ``top_k`` tokens with the +- highest probabilities are sampled. Then tokens are further filtered based +- on ``top_p`` with the final token selected using temperature sampling. Use +- a lower number for less random responses and a higher number for more +- random responses. +- */ +- topK?: number; +- /** Number of response variations to return. +- */ +- candidateCount?: number; +- /** Maximum number of tokens that can be generated in the response. +- */ +- maxOutputTokens?: number; +- /** List of strings that tells the model to stop generating text if one +- of the strings is encountered in the response. +- */ +- stopSequences?: string[]; +- /** Whether to return the log probabilities of the tokens that were +- chosen by the model at each step. +- */ +- responseLogprobs?: boolean; +- /** Number of top candidate tokens to return the log probabilities for +- at each generation step. +- */ +- logprobs?: number; +- /** Positive values penalize tokens that already appear in the +- generated text, increasing the probability of generating more diverse +- content. +- */ +- presencePenalty?: number; +- /** Positive values penalize tokens that repeatedly appear in the +- generated text, increasing the probability of generating more diverse +- content. +- */ +- frequencyPenalty?: number; +- /** When ``seed`` is fixed to a specific number, the model makes a best +- effort to provide the same response for repeated requests. By default, a +- random number is used. +- */ +- seed?: number; +- /** Output response mimetype of the generated candidate text. +- Supported mimetype: +- - `text/plain`: (default) Text output. +- - `application/json`: JSON response in the candidates. +- The model needs to be prompted to output the appropriate response type, +- otherwise the behavior is undefined. +- This is a preview feature. +- */ +- responseMimeType?: string; +- /** The `Schema` object allows the definition of input and output data types. +- These types can be objects, but also primitives and arrays. +- Represents a select subset of an [OpenAPI 3.0 schema +- object](https://spec.openapis.org/oas/v3.0.3#schema). +- If set, a compatible response_mime_type must also be set. +- Compatible mimetypes: `application/json`: Schema for JSON response. +- */ +- responseSchema?: SchemaUnion; +- /** Configuration for model router requests. +- */ +- routingConfig?: GenerationConfigRoutingConfig; +- /** Configuration for model selection. +- */ +- modelSelectionConfig?: ModelSelectionConfig; +- /** Safety settings in the request to block unsafe content in the +- response. +- */ +- safetySettings?: SafetySetting[]; +- /** Code that enables the system to interact with external systems to +- perform an action outside of the knowledge and scope of the model. +- */ +- tools?: ToolListUnion; +- /** Associates model output to a specific function call. +- */ +- toolConfig?: ToolConfig; +- /** Labels with user-defined metadata to break down billed charges. */ +- labels?: Record; +- /** Resource name of a context cache that can be used in subsequent +- requests. +- */ +- cachedContent?: string; +- /** The requested modalities of the response. Represents the set of +- modalities that the model can return. +- */ +- responseModalities?: string[]; +- /** If specified, the media resolution specified will be used. +- */ +- mediaResolution?: MediaResolution; +- /** The speech generation configuration. +- */ +- speechConfig?: SpeechConfigUnion; +- /** If enabled, audio timestamp will be included in the request to the +- model. +- */ +- audioTimestamp?: boolean; +- /** The configuration for automatic function calling. +- */ +- automaticFunctionCalling?: AutomaticFunctionCallingConfig; +- /** The thinking features configuration. +- */ +- thinkingConfig?: ThinkingConfig; +-} +- +-/** Config for models.generate_content parameters. */ +-export declare interface GenerateContentParameters { +- /** ID of the model to use. For a list of models, see `Google models +- `_. */ +- model: string; +- /** Content of the request. +- */ +- contents: ContentListUnion; +- /** Configuration that contains optional model parameters. +- */ +- config?: GenerateContentConfig; +-} +- +-/** Response message for PredictionService.GenerateContent. */ +-export declare class GenerateContentResponse { +- /** Response variations returned by the model. +- */ +- candidates?: Candidate[]; +- /** Timestamp when the request is made to the server. +- */ +- createTime?: string; +- /** Identifier for each response. +- */ +- responseId?: string; +- /** The history of automatic function calling. +- */ +- automaticFunctionCallingHistory?: Content[]; +- /** Output only. The model version used to generate the response. */ +- modelVersion?: string; +- /** Output only. Content filter results for a prompt sent in the request. Note: Sent only in the first stream chunk. Only happens when no candidates were generated due to content violations. */ +- promptFeedback?: GenerateContentResponsePromptFeedback; +- /** Usage metadata about the response(s). */ +- usageMetadata?: GenerateContentResponseUsageMetadata; +- /** +- * Returns the concatenation of all text parts from the first candidate in the response. +- * +- * @remarks +- * If there are multiple candidates in the response, the text from the first +- * one will be returned. +- * If there are non-text parts in the response, the concatenation of all text +- * parts will be returned, and a warning will be logged. +- * If there are thought parts in the response, the concatenation of all text +- * parts excluding the thought parts will be returned. +- * +- * @example +- * ```ts +- * const response = await ai.models.generateContent({ +- * model: 'gemini-2.0-flash', +- * contents: +- * 'Why is the sky blue?', +- * }); +- * +- * console.debug(response.text); +- * ``` +- */ +- get text(): string | undefined; +- /** +- * Returns the concatenation of all inline data parts from the first candidate +- * in the response. +- * +- * @remarks +- * If there are multiple candidates in the response, the inline data from the +- * first one will be returned. If there are non-inline data parts in the +- * response, the concatenation of all inline data parts will be returned, and +- * a warning will be logged. +- */ +- get data(): string | undefined; +- /** +- * Returns the function calls from the first candidate in the response. +- * +- * @remarks +- * If there are multiple candidates in the response, the function calls from +- * the first one will be returned. +- * If there are no function calls in the response, undefined will be returned. +- * +- * @example +- * ```ts +- * const controlLightFunctionDeclaration: FunctionDeclaration = { +- * name: 'controlLight', +- * parameters: { +- * type: Type.OBJECT, +- * description: 'Set the brightness and color temperature of a room light.', +- * properties: { +- * brightness: { +- * type: Type.NUMBER, +- * description: +- * 'Light level from 0 to 100. Zero is off and 100 is full brightness.', +- * }, +- * colorTemperature: { +- * type: Type.STRING, +- * description: +- * 'Color temperature of the light fixture which can be `daylight`, `cool` or `warm`.', +- * }, +- * }, +- * required: ['brightness', 'colorTemperature'], +- * }; +- * const response = await ai.models.generateContent({ +- * model: 'gemini-2.0-flash', +- * contents: 'Dim the lights so the room feels cozy and warm.', +- * config: { +- * tools: [{functionDeclarations: [controlLightFunctionDeclaration]}], +- * toolConfig: { +- * functionCallingConfig: { +- * mode: FunctionCallingConfigMode.ANY, +- * allowedFunctionNames: ['controlLight'], +- * }, +- * }, +- * }, +- * }); +- * console.debug(JSON.stringify(response.functionCalls)); +- * ``` +- */ +- get functionCalls(): FunctionCall[] | undefined; +- /** +- * Returns the first executable code from the first candidate in the response. +- * +- * @remarks +- * If there are multiple candidates in the response, the executable code from +- * the first one will be returned. +- * If there are no executable code in the response, undefined will be +- * returned. +- * +- * @example +- * ```ts +- * const response = await ai.models.generateContent({ +- * model: 'gemini-2.0-flash', +- * contents: +- * 'What is the sum of the first 50 prime numbers? Generate and run code for the calculation, and make sure you get all 50.' +- * config: { +- * tools: [{codeExecution: {}}], +- * }, +- * }); +- * +- * console.debug(response.executableCode); +- * ``` +- */ +- get executableCode(): string | undefined; +- /** +- * Returns the first code execution result from the first candidate in the response. +- * +- * @remarks +- * If there are multiple candidates in the response, the code execution result from +- * the first one will be returned. +- * If there are no code execution result in the response, undefined will be returned. +- * +- * @example +- * ```ts +- * const response = await ai.models.generateContent({ +- * model: 'gemini-2.0-flash', +- * contents: +- * 'What is the sum of the first 50 prime numbers? Generate and run code for the calculation, and make sure you get all 50.' +- * config: { +- * tools: [{codeExecution: {}}], +- * }, +- * }); +- * +- * console.debug(response.codeExecutionResult); +- * ``` +- */ +- get codeExecutionResult(): string | undefined; +-} +- +-/** Content filter results for a prompt sent in the request. */ +-export declare class GenerateContentResponsePromptFeedback { +- /** Output only. Blocked reason. */ +- blockReason?: BlockedReason; +- /** Output only. A readable block reason message. */ +- blockReasonMessage?: string; +- /** Output only. Safety ratings. */ +- safetyRatings?: SafetyRating[]; +-} +- +-/** Usage metadata about response(s). */ +-export declare class GenerateContentResponseUsageMetadata { +- /** Output only. List of modalities of the cached content in the request input. */ +- cacheTokensDetails?: ModalityTokenCount[]; +- /** Output only. Number of tokens in the cached part in the input (the cached content). */ +- cachedContentTokenCount?: number; +- /** Number of tokens in the response(s). */ +- candidatesTokenCount?: number; +- /** Output only. List of modalities that were returned in the response. */ +- candidatesTokensDetails?: ModalityTokenCount[]; +- /** Number of tokens in the request. When `cached_content` is set, this is still the total effective prompt size meaning this includes the number of tokens in the cached content. */ +- promptTokenCount?: number; +- /** Output only. List of modalities that were processed in the request input. */ +- promptTokensDetails?: ModalityTokenCount[]; +- /** Output only. Number of tokens present in thoughts output. */ +- thoughtsTokenCount?: number; +- /** Output only. Number of tokens present in tool-use prompt(s). */ +- toolUsePromptTokenCount?: number; +- /** Output only. List of modalities that were processed for tool-use request inputs. */ +- toolUsePromptTokensDetails?: ModalityTokenCount[]; +- /** Total token count for prompt, response candidates, and tool-use prompts (if present). */ +- totalTokenCount?: number; +- /** Output only. Traffic type. This shows whether a request consumes Pay-As-You-Go or Provisioned Throughput quota. */ +- trafficType?: TrafficType; +-} +- +-/** An output image. */ +-export declare interface GeneratedImage { +- /** The output image data. +- */ +- image?: Image_2; +- /** Responsible AI filter reason if the image is filtered out of the +- response. +- */ +- raiFilteredReason?: string; +- /** Safety attributes of the image. Lists of RAI categories and their +- scores of each content. +- */ +- safetyAttributes?: SafetyAttributes; +- /** The rewritten prompt used for the image generation if the prompt +- enhancer is enabled. +- */ +- enhancedPrompt?: string; +-} +- +-/** A generated video. */ +-export declare interface GeneratedVideo { +- /** The output video */ +- video?: Video; +-} +- +-/** The config for generating an images. */ +-export declare interface GenerateImagesConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +- /** Cloud Storage URI used to store the generated images. +- */ +- outputGcsUri?: string; +- /** Description of what to discourage in the generated images. +- */ +- negativePrompt?: string; +- /** Number of images to generate. +- */ +- numberOfImages?: number; +- /** Aspect ratio of the generated images. +- */ +- aspectRatio?: string; +- /** Controls how much the model adheres to the text prompt. Large +- values increase output and prompt alignment, but may compromise image +- quality. +- */ +- guidanceScale?: number; +- /** Random seed for image generation. This is not available when +- ``add_watermark`` is set to true. +- */ +- seed?: number; +- /** Filter level for safety filtering. +- */ +- safetyFilterLevel?: SafetyFilterLevel; +- /** Allows generation of people by the model. +- */ +- personGeneration?: PersonGeneration; +- /** Whether to report the safety scores of each generated image and +- the positive prompt in the response. +- */ +- includeSafetyAttributes?: boolean; +- /** Whether to include the Responsible AI filter reason if the image +- is filtered out of the response. +- */ +- includeRaiReason?: boolean; +- /** Language of the text in the prompt. +- */ +- language?: ImagePromptLanguage; +- /** MIME type of the generated image. +- */ +- outputMimeType?: string; +- /** Compression quality of the generated image (for ``image/jpeg`` +- only). +- */ +- outputCompressionQuality?: number; +- /** Whether to add a watermark to the generated images. +- */ +- addWatermark?: boolean; +- /** Whether to use the prompt rewriting logic. +- */ +- enhancePrompt?: boolean; +-} +- +-/** The parameters for generating images. */ +-export declare interface GenerateImagesParameters { +- /** ID of the model to use. For a list of models, see `Google models +- `_. */ +- model: string; +- /** Text prompt that typically describes the images to output. +- */ +- prompt: string; +- /** Configuration for generating images. +- */ +- config?: GenerateImagesConfig; +-} +- +-/** The output images response. */ +-export declare class GenerateImagesResponse { +- /** List of generated images. +- */ +- generatedImages?: GeneratedImage[]; +- /** Safety attributes of the positive prompt. Only populated if +- ``include_safety_attributes`` is set to True. +- */ +- positivePromptSafetyAttributes?: SafetyAttributes; +-} +- +-/** Configuration for generating videos. */ +-export declare interface GenerateVideosConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +- /** Number of output videos. */ +- numberOfVideos?: number; +- /** The gcs bucket where to save the generated videos. */ +- outputGcsUri?: string; +- /** Frames per second for video generation. */ +- fps?: number; +- /** Duration of the clip for video generation in seconds. */ +- durationSeconds?: number; +- /** The RNG seed. If RNG seed is exactly same for each request with unchanged inputs, the prediction results will be consistent. Otherwise, a random RNG seed will be used each time to produce a different result. */ +- seed?: number; +- /** The aspect ratio for the generated video. 16:9 (landscape) and 9:16 (portrait) are supported. */ +- aspectRatio?: string; +- /** The resolution for the generated video. 1280x720, 1920x1080 are supported. */ +- resolution?: string; +- /** Whether allow to generate person videos, and restrict to specific ages. Supported values are: dont_allow, allow_adult. */ +- personGeneration?: string; +- /** The pubsub topic where to publish the video generation progress. */ +- pubsubTopic?: string; +- /** Optional field in addition to the text content. Negative prompts can be explicitly stated here to help generate the video. */ +- negativePrompt?: string; +- /** Whether to use the prompt rewriting logic. */ +- enhancePrompt?: boolean; +-} +- +-/** A video generation operation. */ +-export declare interface GenerateVideosOperation { +- /** The server-assigned name, which is only unique within the same service that originally returns it. If you use the default HTTP mapping, the `name` should be a resource name ending with `operations/{unique_id}`. */ +- name?: string; +- /** Service-specific metadata associated with the operation. It typically contains progress information and common metadata such as create time. Some services might not provide such metadata. Any method that returns a long-running operation should document the metadata type, if any. */ +- metadata?: Record; +- /** If the value is `false`, it means the operation is still in progress. If `true`, the operation is completed, and either `error` or `response` is available. */ +- done?: boolean; +- /** The error result of the operation in case of failure or cancellation. */ +- error?: Record; +- /** The generated videos. */ +- response?: GenerateVideosResponse; +-} +- +-/** Class that represents the parameters for generating an image. */ +-export declare interface GenerateVideosParameters { +- /** ID of the model to use. For a list of models, see `Google models +- `_. */ +- model: string; +- /** The text prompt for generating the videos. Optional for image to video use cases. */ +- prompt?: string; +- /** The input image for generating the videos. +- Optional if prompt is provided. */ +- image?: Image_2; +- /** Configuration for generating videos. */ +- config?: GenerateVideosConfig; +-} +- +-/** Response with generated videos. */ +-export declare class GenerateVideosResponse { +- /** List of the generated videos */ +- generatedVideos?: GeneratedVideo[]; +- /** Returns if any videos were filtered due to RAI policies. */ +- raiMediaFilteredCount?: number; +- /** Returns rai failure reasons if any. */ +- raiMediaFilteredReasons?: string[]; +-} +- +-/** Generation config. */ +-export declare interface GenerationConfig { +- /** Optional. If enabled, audio timestamp will be included in the request to the model. */ +- audioTimestamp?: boolean; +- /** Optional. Number of candidates to generate. */ +- candidateCount?: number; +- /** Optional. Frequency penalties. */ +- frequencyPenalty?: number; +- /** Optional. Logit probabilities. */ +- logprobs?: number; +- /** Optional. The maximum number of output tokens to generate per message. */ +- maxOutputTokens?: number; +- /** Optional. If specified, the media resolution specified will be used. */ +- mediaResolution?: MediaResolution; +- /** Optional. Positive penalties. */ +- presencePenalty?: number; +- /** Optional. If true, export the logprobs results in response. */ +- responseLogprobs?: boolean; +- /** Optional. Output response mimetype of the generated candidate text. Supported mimetype: - `text/plain`: (default) Text output. - `application/json`: JSON response in the candidates. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. This is a preview feature. */ +- responseMimeType?: string; +- /** Optional. The `Schema` object allows the definition of input and output data types. These types can be objects, but also primitives and arrays. Represents a select subset of an [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema). If set, a compatible response_mime_type must also be set. Compatible mimetypes: `application/json`: Schema for JSON response. */ +- responseSchema?: Schema; +- /** Optional. Routing configuration. */ +- routingConfig?: GenerationConfigRoutingConfig; +- /** Optional. Seed. */ +- seed?: number; +- /** Optional. Stop sequences. */ +- stopSequences?: string[]; +- /** Optional. Controls the randomness of predictions. */ +- temperature?: number; +- /** Optional. If specified, top-k sampling will be used. */ +- topK?: number; +- /** Optional. If specified, nucleus sampling will be used. */ +- topP?: number; +-} +- +-/** The configuration for routing the request to a specific model. */ +-export declare interface GenerationConfigRoutingConfig { +- /** Automated routing. */ +- autoMode?: GenerationConfigRoutingConfigAutoRoutingMode; +- /** Manual routing. */ +- manualMode?: GenerationConfigRoutingConfigManualRoutingMode; +-} +- +-/** When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. */ +-export declare interface GenerationConfigRoutingConfigAutoRoutingMode { +- /** The model routing preference. */ +- modelRoutingPreference?: 'UNKNOWN' | 'PRIORITIZE_QUALITY' | 'BALANCED' | 'PRIORITIZE_COST'; +-} +- +-/** When manual routing is set, the specified model will be used directly. */ +-export declare interface GenerationConfigRoutingConfigManualRoutingMode { +- /** The model name to use. Only the public LLM models are accepted. e.g. 'gemini-1.5-pro-001'. */ +- modelName?: string; +-} +- +-/** Optional parameters for caches.get method. */ +-export declare interface GetCachedContentConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +-} +- +-/** Parameters for caches.get method. */ +-export declare interface GetCachedContentParameters { +- /** The server-generated resource name of the cached content. +- */ +- name: string; +- /** Optional parameters for the request. +- */ +- config?: GetCachedContentConfig; +-} +- +-/** Used to override the default configuration. */ +-export declare interface GetFileConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +-} +- +-/** Generates the parameters for the get method. */ +-export declare interface GetFileParameters { +- /** The name identifier for the file to retrieve. */ +- name: string; +- /** Used to override the default configuration. */ +- config?: GetFileConfig; +-} +- +-/** Optional parameters for models.get method. */ +-export declare interface GetModelConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +-} +- +-export declare interface GetModelParameters { +- model: string; +- /** Optional parameters for the request. */ +- config?: GetModelConfig; +-} +- +-export declare interface GetOperationConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +-} +- +-/** Parameters for the GET method. */ +-export declare interface GetOperationParameters { +- /** The server-assigned name for the operation. */ +- operationName: string; +- /** Used to override the default configuration. */ +- config?: GetOperationConfig; +-} +- +-/** Optional parameters for tunings.get method. */ +-export declare interface GetTuningJobConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +-} +- +-/** Parameters for the get method. */ +-export declare interface GetTuningJobParameters { +- name: string; +- /** Optional parameters for the request. */ +- config?: GetTuningJobConfig; +-} +- +-/** +- * The Google GenAI SDK. +- * +- * @remarks +- * Provides access to the GenAI features through either the {@link +- * https://cloud.google.com/vertex-ai/docs/reference/rest | Gemini API} or +- * the {@link https://cloud.google.com/vertex-ai/docs/reference/rest | Vertex AI +- * API}. +- * +- * The {@link GoogleGenAIOptions.vertexai} value determines which of the API +- * services to use. +- * +- * When using the Gemini API, a {@link GoogleGenAIOptions.apiKey} must also be +- * set. When using Vertex AI, currently only {@link GoogleGenAIOptions.apiKey} +- * is supported via Express mode. {@link GoogleGenAIOptions.project} and {@link +- * GoogleGenAIOptions.location} should not be set. +- * +- * @example +- * Initializing the SDK for using the Gemini API: +- * ```ts +- * import {GoogleGenAI} from '@google/genai'; +- * const ai = new GoogleGenAI({apiKey: 'GEMINI_API_KEY'}); +- * ``` +- * +- * @example +- * Initializing the SDK for using the Vertex AI API: +- * ```ts +- * import {GoogleGenAI} from '@google/genai'; +- * const ai = new GoogleGenAI({ +- * vertexai: true, +- * project: 'PROJECT_ID', +- * location: 'PROJECT_LOCATION' +- * }); +- * ``` +- * +- */ +-export declare class GoogleGenAI { +- protected readonly apiClient: ApiClient; +- private readonly apiKey?; +- readonly vertexai: boolean; +- private readonly apiVersion?; +- readonly models: Models; +- readonly live: Live; +- readonly chats: Chats; +- readonly caches: Caches; +- readonly files: Files; +- readonly operations: Operations; +- readonly tunings: Tunings; +- constructor(options: GoogleGenAIOptions); +-} +- +-/** +- * Google Gen AI SDK's configuration options. +- * +- * See {@link GoogleGenAI} for usage samples. +- */ +-export declare interface GoogleGenAIOptions { +- /** +- * Optional. Determines whether to use the Vertex AI or the Gemini API. +- * +- * @remarks +- * When true, the {@link https://cloud.google.com/vertex-ai/docs/reference/rest | Vertex AI API} will used. +- * When false, the {@link https://ai.google.dev/api | Gemini API} will be used. +- * +- * If unset, default SDK behavior is to use the Gemini API service. +- */ +- vertexai?: boolean; +- /** +- * Optional. The Google Cloud project ID for Vertex AI clients. +- * +- * Find your project ID: https://cloud.google.com/resource-manager/docs/creating-managing-projects#identifying_projects +- * +- * @remarks +- * Only supported on Node runtimes, ignored on browser runtimes. +- */ +- project?: string; +- /** +- * Optional. The Google Cloud project {@link https://cloud.google.com/vertex-ai/generative-ai/docs/learn/locations | location} for Vertex AI clients. +- * +- * @remarks +- * Only supported on Node runtimes, ignored on browser runtimes. +- * +- */ +- location?: string; +- /** +- * The API Key, required for Gemini API clients. +- * +- * @remarks +- * Required on browser runtimes. +- */ +- apiKey?: string; +- /** +- * Optional. The API version to use. +- * +- * @remarks +- * If unset, the default API version will be used. +- */ +- apiVersion?: string; +- /** +- * Optional. Authentication options defined by the by google-auth-library for Vertex AI clients. +- * +- * @remarks +- * @see {@link https://github.com/googleapis/google-auth-library-nodejs/blob/v9.15.0/src/auth/googleauth.ts | GoogleAuthOptions interface in google-auth-library-nodejs}. +- * +- * Only supported on Node runtimes, ignored on browser runtimes. +- * +- */ +- googleAuthOptions?: GoogleAuthOptions; +- /** +- * Optional. A set of customizable configuration for HTTP requests. +- */ +- httpOptions?: HttpOptions; +-} +- +-/** Tool to support Google Maps in Model. */ +-export declare interface GoogleMaps { +- /** Optional. Auth config for the Google Maps tool. */ +- authConfig?: AuthConfig; +-} +- +-/** The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). */ +-export declare interface GoogleRpcStatus { +- /** The status code, which should be an enum value of google.rpc.Code. */ +- code?: number; +- /** A list of messages that carry the error details. There is a common set of message types for APIs to use. */ +- details?: Record[]; +- /** A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. */ +- message?: string; +-} +- +-/** Tool to support Google Search in Model. Powered by Google. */ +-export declare interface GoogleSearch { +- /** Optional. Filter search results to a specific time range. +- If customers set a start time, they must set an end time (and vice versa). +- */ +- timeRangeFilter?: Interval; +-} +- +-/** Tool to retrieve public web data for grounding, powered by Google. */ +-export declare interface GoogleSearchRetrieval { +- /** Specifies the dynamic retrieval configuration for the given source. */ +- dynamicRetrievalConfig?: DynamicRetrievalConfig; +-} +- +-/** Represents a whole or partial calendar date, such as a birthday. The time of day and time zone are either specified elsewhere or are insignificant. The date is relative to the Gregorian Calendar. This can represent one of the following: * A full date, with non-zero year, month, and day values. * A month and day, with a zero year (for example, an anniversary). * A year on its own, with a zero month and a zero day. * A year and month, with a zero day (for example, a credit card expiration date). Related types: * google.type.TimeOfDay * google.type.DateTime * google.protobuf.Timestamp */ +-export declare interface GoogleTypeDate { +- /** Day of a month. Must be from 1 to 31 and valid for the year and month, or 0 to specify a year by itself or a year and month where the day isn't significant. */ +- day?: number; +- /** Month of a year. Must be from 1 to 12, or 0 to specify a year without a month and day. */ +- month?: number; +- /** Year of the date. Must be from 1 to 9999, or 0 to specify a date without a year. */ +- year?: number; +-} +- +-/** Grounding chunk. */ +-export declare interface GroundingChunk { +- /** Grounding chunk from context retrieved by the retrieval tools. */ +- retrievedContext?: GroundingChunkRetrievedContext; +- /** Grounding chunk from the web. */ +- web?: GroundingChunkWeb; +-} +- +-/** Chunk from context retrieved by the retrieval tools. */ +-export declare interface GroundingChunkRetrievedContext { +- /** Text of the attribution. */ +- text?: string; +- /** Title of the attribution. */ +- title?: string; +- /** URI reference of the attribution. */ +- uri?: string; +-} +- +-/** Chunk from the web. */ +-export declare interface GroundingChunkWeb { +- /** Domain of the (original) URI. */ +- domain?: string; +- /** Title of the chunk. */ +- title?: string; +- /** URI reference of the chunk. */ +- uri?: string; +-} +- +-/** Metadata returned to client when grounding is enabled. */ +-export declare interface GroundingMetadata { +- /** List of supporting references retrieved from specified grounding source. */ +- groundingChunks?: GroundingChunk[]; +- /** Optional. List of grounding support. */ +- groundingSupports?: GroundingSupport[]; +- /** Optional. Output only. Retrieval metadata. */ +- retrievalMetadata?: RetrievalMetadata; +- /** Optional. Queries executed by the retrieval tools. */ +- retrievalQueries?: string[]; +- /** Optional. Google search entry for the following-up web searches. */ +- searchEntryPoint?: SearchEntryPoint; +- /** Optional. Web search queries for the following-up web search. */ +- webSearchQueries?: string[]; +-} +- +-/** Grounding support. */ +-export declare interface GroundingSupport { +- /** Confidence score of the support references. Ranges from 0 to 1. 1 is the most confident. This list must have the same size as the grounding_chunk_indices. */ +- confidenceScores?: number[]; +- /** A list of indices (into 'grounding_chunk') specifying the citations associated with the claim. For instance [1,3,4] means that grounding_chunk[1], grounding_chunk[3], grounding_chunk[4] are the retrieved content attributed to the claim. */ +- groundingChunkIndices?: number[]; +- /** Segment of the content this support belongs to. */ +- segment?: Segment; +-} +- +-/** Optional. Specify if the threshold is used for probability or severity score. If not specified, the threshold is used for probability score. */ +-export declare enum HarmBlockMethod { +- /** +- * The harm block method is unspecified. +- */ +- HARM_BLOCK_METHOD_UNSPECIFIED = "HARM_BLOCK_METHOD_UNSPECIFIED", +- /** +- * The harm block method uses both probability and severity scores. +- */ +- SEVERITY = "SEVERITY", +- /** +- * The harm block method uses the probability score. +- */ +- PROBABILITY = "PROBABILITY" +-} +- +-/** Required. The harm block threshold. */ +-export declare enum HarmBlockThreshold { +- /** +- * Unspecified harm block threshold. +- */ +- HARM_BLOCK_THRESHOLD_UNSPECIFIED = "HARM_BLOCK_THRESHOLD_UNSPECIFIED", +- /** +- * Block low threshold and above (i.e. block more). +- */ +- BLOCK_LOW_AND_ABOVE = "BLOCK_LOW_AND_ABOVE", +- /** +- * Block medium threshold and above. +- */ +- BLOCK_MEDIUM_AND_ABOVE = "BLOCK_MEDIUM_AND_ABOVE", +- /** +- * Block only high threshold (i.e. block less). +- */ +- BLOCK_ONLY_HIGH = "BLOCK_ONLY_HIGH", +- /** +- * Block none. +- */ +- BLOCK_NONE = "BLOCK_NONE", +- /** +- * Turn off the safety filter. +- */ +- OFF = "OFF" +-} +- +-/** Required. Harm category. */ +-export declare enum HarmCategory { +- /** +- * The harm category is unspecified. +- */ +- HARM_CATEGORY_UNSPECIFIED = "HARM_CATEGORY_UNSPECIFIED", +- /** +- * The harm category is hate speech. +- */ +- HARM_CATEGORY_HATE_SPEECH = "HARM_CATEGORY_HATE_SPEECH", +- /** +- * The harm category is dangerous content. +- */ +- HARM_CATEGORY_DANGEROUS_CONTENT = "HARM_CATEGORY_DANGEROUS_CONTENT", +- /** +- * The harm category is harassment. +- */ +- HARM_CATEGORY_HARASSMENT = "HARM_CATEGORY_HARASSMENT", +- /** +- * The harm category is sexually explicit content. +- */ +- HARM_CATEGORY_SEXUALLY_EXPLICIT = "HARM_CATEGORY_SEXUALLY_EXPLICIT", +- /** +- * The harm category is civic integrity. +- */ +- HARM_CATEGORY_CIVIC_INTEGRITY = "HARM_CATEGORY_CIVIC_INTEGRITY" +-} +- +-/** Output only. Harm probability levels in the content. */ +-export declare enum HarmProbability { +- /** +- * Harm probability unspecified. +- */ +- HARM_PROBABILITY_UNSPECIFIED = "HARM_PROBABILITY_UNSPECIFIED", +- /** +- * Negligible level of harm. +- */ +- NEGLIGIBLE = "NEGLIGIBLE", +- /** +- * Low level of harm. +- */ +- LOW = "LOW", +- /** +- * Medium level of harm. +- */ +- MEDIUM = "MEDIUM", +- /** +- * High level of harm. +- */ +- HIGH = "HIGH" +-} +- +-/** Output only. Harm severity levels in the content. */ +-export declare enum HarmSeverity { +- /** +- * Harm severity unspecified. +- */ +- HARM_SEVERITY_UNSPECIFIED = "HARM_SEVERITY_UNSPECIFIED", +- /** +- * Negligible level of harm severity. +- */ +- HARM_SEVERITY_NEGLIGIBLE = "HARM_SEVERITY_NEGLIGIBLE", +- /** +- * Low level of harm severity. +- */ +- HARM_SEVERITY_LOW = "HARM_SEVERITY_LOW", +- /** +- * Medium level of harm severity. +- */ +- HARM_SEVERITY_MEDIUM = "HARM_SEVERITY_MEDIUM", +- /** +- * High level of harm severity. +- */ +- HARM_SEVERITY_HIGH = "HARM_SEVERITY_HIGH" +-} +- +-/** HTTP options to be used in each of the requests. */ +-export declare interface HttpOptions { +- /** The base URL for the AI platform service endpoint. */ +- baseUrl?: string; +- /** Specifies the version of the API to use. */ +- apiVersion?: string; +- /** Additional HTTP headers to be sent with the request. */ +- headers?: Record; +- /** Timeout for the request in milliseconds. */ +- timeout?: number; +-} +- +-/** +- * Represents the necessary information to send a request to an API endpoint. +- * This interface defines the structure for constructing and executing HTTP +- * requests. +- */ +-declare interface HttpRequest { +- /** +- * URL path from the modules, this path is appended to the base API URL to +- * form the complete request URL. +- * +- * If you wish to set full URL, use httpOptions.baseUrl instead. Example to +- * set full URL in the request: +- * +- * const request: HttpRequest = { +- * path: '', +- * httpOptions: { +- * baseUrl: 'https://', +- * apiVersion: '', +- * }, +- * httpMethod: 'GET', +- * }; +- * +- * The result URL will be: https:// +- * +- */ +- path: string; +- /** +- * Optional query parameters to be appended to the request URL. +- */ +- queryParams?: Record; +- /** +- * Optional request body in json string or Blob format, GET request doesn't +- * need a request body. +- */ +- body?: string | Blob; +- /** +- * The HTTP method to be used for the request. +- */ +- httpMethod: 'GET' | 'POST' | 'PATCH' | 'DELETE'; +- /** +- * Optional set of customizable configuration for HTTP requests. +- */ +- httpOptions?: HttpOptions; +- /** +- * Optional abort signal which can be used to cancel the request. +- */ +- abortSignal?: AbortSignal; +-} +- +-/** A wrapper class for the http response. */ +-export declare class HttpResponse { +- /** Used to retain the processed HTTP headers in the response. */ +- headers?: Record; +- /** +- * The original http response. +- */ +- responseInternal: Response; +- constructor(response: Response); +- json(): Promise; +-} +- +-/** An image. */ +-declare interface Image_2 { +- /** The Cloud Storage URI of the image. ``Image`` can contain a value +- for this field or the ``image_bytes`` field but not both. +- */ +- gcsUri?: string; +- /** The image bytes data. ``Image`` can contain a value for this field +- or the ``gcs_uri`` field but not both. +- */ +- imageBytes?: string; +- /** The MIME type of the image. */ +- mimeType?: string; +-} +-export { Image_2 as Image } +- +-/** Enum that specifies the language of the text in the prompt. */ +-export declare enum ImagePromptLanguage { +- auto = "auto", +- en = "en", +- ja = "ja", +- ko = "ko", +- hi = "hi" +-} +- +-/** Represents a time interval, encoded as a start time (inclusive) and an end time (exclusive). +- +- The start time must be less than or equal to the end time. +- When the start equals the end time, the interval is an empty interval. +- (matches no time) +- When both start and end are unspecified, the interval matches any time. +- */ +-export declare interface Interval { +- /** The start time of the interval. */ +- startTime?: string; +- /** The end time of the interval. */ +- endTime?: string; +-} +- +-/** Output only. The detailed state of the job. */ +-export declare enum JobState { +- /** +- * The job state is unspecified. +- */ +- JOB_STATE_UNSPECIFIED = "JOB_STATE_UNSPECIFIED", +- /** +- * The job has been just created or resumed and processing has not yet begun. +- */ +- JOB_STATE_QUEUED = "JOB_STATE_QUEUED", +- /** +- * The service is preparing to run the job. +- */ +- JOB_STATE_PENDING = "JOB_STATE_PENDING", +- /** +- * The job is in progress. +- */ +- JOB_STATE_RUNNING = "JOB_STATE_RUNNING", +- /** +- * The job completed successfully. +- */ +- JOB_STATE_SUCCEEDED = "JOB_STATE_SUCCEEDED", +- /** +- * The job failed. +- */ +- JOB_STATE_FAILED = "JOB_STATE_FAILED", +- /** +- * The job is being cancelled. From this state the job may only go to either `JOB_STATE_SUCCEEDED`, `JOB_STATE_FAILED` or `JOB_STATE_CANCELLED`. +- */ +- JOB_STATE_CANCELLING = "JOB_STATE_CANCELLING", +- /** +- * The job has been cancelled. +- */ +- JOB_STATE_CANCELLED = "JOB_STATE_CANCELLED", +- /** +- * The job has been stopped, and can be resumed. +- */ +- JOB_STATE_PAUSED = "JOB_STATE_PAUSED", +- /** +- * The job has expired. +- */ +- JOB_STATE_EXPIRED = "JOB_STATE_EXPIRED", +- /** +- * The job is being updated. Only jobs in the `RUNNING` state can be updated. After updating, the job goes back to the `RUNNING` state. +- */ +- JOB_STATE_UPDATING = "JOB_STATE_UPDATING", +- /** +- * The job is partially succeeded, some results may be missing due to errors. +- */ +- JOB_STATE_PARTIALLY_SUCCEEDED = "JOB_STATE_PARTIALLY_SUCCEEDED" +-} +- +-/** Required. Programming language of the `code`. */ +-export declare enum Language { +- /** +- * Unspecified language. This value should not be used. +- */ +- LANGUAGE_UNSPECIFIED = "LANGUAGE_UNSPECIFIED", +- /** +- * Python >= 3.10, with numpy and simpy available. +- */ +- PYTHON = "PYTHON" +-} +- +-/** An object that represents a latitude/longitude pair. +- +- This is expressed as a pair of doubles to represent degrees latitude and +- degrees longitude. Unless specified otherwise, this object must conform to the +- +- WGS84 standard. Values must be within normalized ranges. +- */ +-export declare interface LatLng { +- /** The latitude in degrees. It must be in the range [-90.0, +90.0]. */ +- latitude?: number; +- /** The longitude in degrees. It must be in the range [-180.0, +180.0] */ +- longitude?: number; +-} +- +-/** Config for caches.list method. */ +-export declare interface ListCachedContentsConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +- pageSize?: number; +- pageToken?: string; +-} +- +-/** Parameters for caches.list method. */ +-export declare interface ListCachedContentsParameters { +- /** Configuration that contains optional parameters. +- */ +- config?: ListCachedContentsConfig; +-} +- +-export declare class ListCachedContentsResponse { +- nextPageToken?: string; +- /** List of cached contents. +- */ +- cachedContents?: CachedContent[]; +-} +- +-/** Used to override the default configuration. */ +-export declare interface ListFilesConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +- pageSize?: number; +- pageToken?: string; +-} +- +-/** Generates the parameters for the list method. */ +-export declare interface ListFilesParameters { +- /** Used to override the default configuration. */ +- config?: ListFilesConfig; +-} +- +-/** Response for the list files method. */ +-export declare class ListFilesResponse { +- /** A token to retrieve next page of results. */ +- nextPageToken?: string; +- /** The list of files. */ +- files?: File_2[]; +-} +- +-export declare interface ListModelsConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +- pageSize?: number; +- pageToken?: string; +- filter?: string; +- /** Set true to list base models, false to list tuned models. */ +- queryBase?: boolean; +-} +- +-export declare interface ListModelsParameters { +- config?: ListModelsConfig; +-} +- +-export declare class ListModelsResponse { +- nextPageToken?: string; +- models?: Model[]; +-} +- +-/** Configuration for the list tuning jobs method. */ +-export declare interface ListTuningJobsConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +- pageSize?: number; +- pageToken?: string; +- filter?: string; +-} +- +-/** Parameters for the list tuning jobs method. */ +-export declare interface ListTuningJobsParameters { +- config?: ListTuningJobsConfig; +-} +- +-/** Response for the list tuning jobs method. */ +-export declare class ListTuningJobsResponse { +- /** A token to retrieve the next page of results. Pass to ListTuningJobsRequest.page_token to obtain that page. */ +- nextPageToken?: string; +- /** List of TuningJobs in the requested page. */ +- tuningJobs?: TuningJob[]; +-} +- +-/** +- Live class encapsulates the configuration for live interaction with the +- Generative Language API. It embeds ApiClient for general API settings. +- +- @experimental +- */ +-export declare class Live { +- private readonly apiClient; +- private readonly auth; +- private readonly webSocketFactory; +- readonly music: LiveMusic; +- constructor(apiClient: ApiClient, auth: Auth, webSocketFactory: WebSocketFactory); +- /** +- Establishes a connection to the specified model with the given +- configuration and returns a Session object representing that connection. +- +- @experimental Built-in MCP support is an experimental feature, may change in +- future versions. +- +- @remarks +- +- @param params - The parameters for establishing a connection to the model. +- @return A live session. +- +- @example +- ```ts +- let model: string; +- if (GOOGLE_GENAI_USE_VERTEXAI) { +- model = 'gemini-2.0-flash-live-preview-04-09'; +- } else { +- model = 'gemini-2.0-flash-live-001'; +- } +- const session = await ai.live.connect({ +- model: model, +- config: { +- responseModalities: [Modality.AUDIO], +- }, +- callbacks: { +- onopen: () => { +- console.log('Connected to the socket.'); +- }, +- onmessage: (e: MessageEvent) => { +- console.log('Received message from the server: %s\n', debug(e.data)); +- }, +- onerror: (e: ErrorEvent) => { +- console.log('Error occurred: %s\n', debug(e.error)); +- }, +- onclose: (e: CloseEvent) => { +- console.log('Connection closed.'); +- }, +- }, +- }); +- ``` +- */ +- connect(params: types.LiveConnectParameters): Promise; +- private isCallableTool; +-} +- +-/** Callbacks for the live API. */ +-export declare interface LiveCallbacks { +- /** +- * Called when the websocket connection is established. +- */ +- onopen?: (() => void) | null; +- /** +- * Called when a message is received from the server. +- */ +- onmessage: (e: LiveServerMessage) => void; +- /** +- * Called when an error occurs. +- */ +- onerror?: ((e: ErrorEvent) => void) | null; +- /** +- * Called when the websocket connection is closed. +- */ +- onclose?: ((e: CloseEvent) => void) | null; +-} +- +-/** Incremental update of the current conversation delivered from the client. +- +- All the content here will unconditionally be appended to the conversation +- history and used as part of the prompt to the model to generate content. +- +- A message here will interrupt any current model generation. +- */ +-export declare interface LiveClientContent { +- /** The content appended to the current conversation with the model. +- +- For single-turn queries, this is a single instance. For multi-turn +- queries, this is a repeated field that contains conversation history and +- latest request. +- */ +- turns?: Content[]; +- /** If true, indicates that the server content generation should start with +- the currently accumulated prompt. Otherwise, the server will await +- additional messages before starting generation. */ +- turnComplete?: boolean; +-} +- +-/** Messages sent by the client in the API call. */ +-export declare interface LiveClientMessage { +- /** Message to be sent by the system when connecting to the API. SDK users should not send this message. */ +- setup?: LiveClientSetup; +- /** Incremental update of the current conversation delivered from the client. */ +- clientContent?: LiveClientContent; +- /** User input that is sent in real time. */ +- realtimeInput?: LiveClientRealtimeInput; +- /** Response to a `ToolCallMessage` received from the server. */ +- toolResponse?: LiveClientToolResponse; +-} +- +-/** User input that is sent in real time. +- +- This is different from `LiveClientContent` in a few ways: +- +- - Can be sent continuously without interruption to model generation. +- - If there is a need to mix data interleaved across the +- `LiveClientContent` and the `LiveClientRealtimeInput`, server attempts to +- optimize for best response, but there are no guarantees. +- - End of turn is not explicitly specified, but is rather derived from user +- activity (for example, end of speech). +- - Even before the end of turn, the data is processed incrementally +- to optimize for a fast start of the response from the model. +- - Is always assumed to be the user's input (cannot be used to populate +- conversation history). +- */ +-export declare interface LiveClientRealtimeInput { +- /** Inlined bytes data for media input. */ +- mediaChunks?: Blob_2[]; +- /** The realtime audio input stream. */ +- audio?: Blob_2; +- /** +- Indicates that the audio stream has ended, e.g. because the microphone was +- turned off. +- +- This should only be sent when automatic activity detection is enabled +- (which is the default). +- +- The client can reopen the stream by sending an audio message. +- */ +- audioStreamEnd?: boolean; +- /** The realtime video input stream. */ +- video?: Blob_2; +- /** The realtime text input stream. */ +- text?: string; +- /** Marks the start of user activity. */ +- activityStart?: ActivityStart; +- /** Marks the end of user activity. */ +- activityEnd?: ActivityEnd; +-} +- +-/** Message contains configuration that will apply for the duration of the streaming session. */ +-export declare interface LiveClientSetup { +- /** +- The fully qualified name of the publisher model or tuned model endpoint to +- use. +- */ +- model?: string; +- /** The generation configuration for the session. +- Note: only a subset of fields are supported. +- */ +- generationConfig?: GenerationConfig; +- /** The user provided system instructions for the model. +- Note: only text should be used in parts and content in each part will be +- in a separate paragraph. */ +- systemInstruction?: ContentUnion; +- /** A list of `Tools` the model may use to generate the next response. +- +- A `Tool` is a piece of code that enables the system to interact with +- external systems to perform an action, or set of actions, outside of +- knowledge and scope of the model. */ +- tools?: ToolListUnion; +- /** Configures the realtime input behavior in BidiGenerateContent. */ +- realtimeInputConfig?: RealtimeInputConfig; +- /** Configures session resumption mechanism. +- +- If included server will send SessionResumptionUpdate messages. */ +- sessionResumption?: SessionResumptionConfig; +- /** Configures context window compression mechanism. +- +- If included, server will compress context window to fit into given length. */ +- contextWindowCompression?: ContextWindowCompressionConfig; +- /** The transcription of the input aligns with the input audio language. +- */ +- inputAudioTranscription?: AudioTranscriptionConfig; +- /** The transcription of the output aligns with the language code +- specified for the output audio. +- */ +- outputAudioTranscription?: AudioTranscriptionConfig; +- /** Configures the proactivity of the model. This allows the model to respond proactively to +- the input and to ignore irrelevant input. */ +- proactivity?: ProactivityConfig; +-} +- +-/** Client generated response to a `ToolCall` received from the server. +- +- Individual `FunctionResponse` objects are matched to the respective +- `FunctionCall` objects by the `id` field. +- +- Note that in the unary and server-streaming GenerateContent APIs function +- calling happens by exchanging the `Content` parts, while in the bidi +- GenerateContent APIs function calling happens over this dedicated set of +- messages. +- */ +-export declare class LiveClientToolResponse { +- /** The response to the function calls. */ +- functionResponses?: FunctionResponse[]; +-} +- +-/** Session config for the API connection. */ +-export declare interface LiveConnectConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +- /** The generation configuration for the session. */ +- generationConfig?: GenerationConfig; +- /** The requested modalities of the response. Represents the set of +- modalities that the model can return. Defaults to AUDIO if not specified. +- */ +- responseModalities?: Modality[]; +- /** Value that controls the degree of randomness in token selection. +- Lower temperatures are good for prompts that require a less open-ended or +- creative response, while higher temperatures can lead to more diverse or +- creative results. +- */ +- temperature?: number; +- /** Tokens are selected from the most to least probable until the sum +- of their probabilities equals this value. Use a lower value for less +- random responses and a higher value for more random responses. +- */ +- topP?: number; +- /** For each token selection step, the ``top_k`` tokens with the +- highest probabilities are sampled. Then tokens are further filtered based +- on ``top_p`` with the final token selected using temperature sampling. Use +- a lower number for less random responses and a higher number for more +- random responses. +- */ +- topK?: number; +- /** Maximum number of tokens that can be generated in the response. +- */ +- maxOutputTokens?: number; +- /** If specified, the media resolution specified will be used. +- */ +- mediaResolution?: MediaResolution; +- /** When ``seed`` is fixed to a specific number, the model makes a best +- effort to provide the same response for repeated requests. By default, a +- random number is used. +- */ +- seed?: number; +- /** The speech generation configuration. +- */ +- speechConfig?: SpeechConfig; +- /** If enabled, the model will detect emotions and adapt its responses accordingly. */ +- enableAffectiveDialog?: boolean; +- /** The user provided system instructions for the model. +- Note: only text should be used in parts and content in each part will be +- in a separate paragraph. */ +- systemInstruction?: ContentUnion; +- /** A list of `Tools` the model may use to generate the next response. +- +- A `Tool` is a piece of code that enables the system to interact with +- external systems to perform an action, or set of actions, outside of +- knowledge and scope of the model. */ +- tools?: ToolListUnion; +- /** Configures session resumption mechanism. +- +- If included the server will send SessionResumptionUpdate messages. */ +- sessionResumption?: SessionResumptionConfig; +- /** The transcription of the input aligns with the input audio language. +- */ +- inputAudioTranscription?: AudioTranscriptionConfig; +- /** The transcription of the output aligns with the language code +- specified for the output audio. +- */ +- outputAudioTranscription?: AudioTranscriptionConfig; +- /** Configures the realtime input behavior in BidiGenerateContent. */ +- realtimeInputConfig?: RealtimeInputConfig; +- /** Configures context window compression mechanism. +- +- If included, server will compress context window to fit into given length. */ +- contextWindowCompression?: ContextWindowCompressionConfig; +- /** Configures the proactivity of the model. This allows the model to respond proactively to +- the input and to ignore irrelevant input. */ +- proactivity?: ProactivityConfig; +-} +- +-/** Parameters for connecting to the live API. */ +-export declare interface LiveConnectParameters { +- /** ID of the model to use. For a list of models, see `Google models +- `_. */ +- model: string; +- /** callbacks */ +- callbacks: LiveCallbacks; +- /** Optional configuration parameters for the request. +- */ +- config?: LiveConnectConfig; +-} +- +-/** Config for LiveEphemeralParameters for Auth Token creation. */ +-export declare interface LiveEphemeralParameters { +- /** ID of the model to configure in the ephemeral token for Live API. +- For a list of models, see `Gemini models +- `. */ +- model?: string; +- /** Configuration specific to Live API connections created using this token. */ +- config?: LiveConnectConfig; +-} +- +-/** +- LiveMusic class encapsulates the configuration for live music +- generation via Lyria Live models. +- +- @experimental +- */ +-declare class LiveMusic { +- private readonly apiClient; +- private readonly auth; +- private readonly webSocketFactory; +- constructor(apiClient: ApiClient, auth: Auth, webSocketFactory: WebSocketFactory); +- /** +- Establishes a connection to the specified model and returns a +- LiveMusicSession object representing that connection. +- +- @experimental +- +- @remarks +- +- @param params - The parameters for establishing a connection to the model. +- @return A live session. +- +- @example +- ```ts +- let model = 'models/lyria-realtime-exp'; +- const session = await ai.live.music.connect({ +- model: model, +- callbacks: { +- onmessage: (e: MessageEvent) => { +- console.log('Received message from the server: %s\n', debug(e.data)); +- }, +- onerror: (e: ErrorEvent) => { +- console.log('Error occurred: %s\n', debug(e.error)); +- }, +- onclose: (e: CloseEvent) => { +- console.log('Connection closed.'); +- }, +- }, +- }); +- ``` +- */ +- connect(params: types.LiveMusicConnectParameters): Promise; +-} +- +-/** Callbacks for the realtime music API. */ +-export declare interface LiveMusicCallbacks { +- /** +- * Called when a message is received from the server. +- */ +- onmessage: (e: LiveMusicServerMessage) => void; +- /** +- * Called when an error occurs. +- */ +- onerror?: ((e: ErrorEvent) => void) | null; +- /** +- * Called when the websocket connection is closed. +- */ +- onclose?: ((e: CloseEvent) => void) | null; +-} +- +-/** User input to start or steer the music. */ +-export declare interface LiveMusicClientContent { +- /** Weighted prompts as the model input. */ +- weightedPrompts?: WeightedPrompt[]; +-} +- +-/** Messages sent by the client in the LiveMusicClientMessage call. */ +-export declare interface LiveMusicClientMessage { +- /** Message to be sent in the first (and only in the first) `LiveMusicClientMessage`. +- Clients should wait for a `LiveMusicSetupComplete` message before +- sending any additional messages. */ +- setup?: LiveMusicClientSetup; +- /** User input to influence music generation. */ +- clientContent?: LiveMusicClientContent; +- /** Configuration for music generation. */ +- musicGenerationConfig?: LiveMusicGenerationConfig; +- /** Playback control signal for the music generation. */ +- playbackControl?: LiveMusicPlaybackControl; +-} +- +-/** Message to be sent by the system when connecting to the API. */ +-export declare interface LiveMusicClientSetup { +- /** The model's resource name. Format: `models/{model}`. */ +- model?: string; +-} +- +-/** Parameters for connecting to the live API. */ +-export declare interface LiveMusicConnectParameters { +- /** The model's resource name. */ +- model: string; +- /** Callbacks invoked on server events. */ +- callbacks: LiveMusicCallbacks; +-} +- +-/** A prompt that was filtered with the reason. */ +-export declare interface LiveMusicFilteredPrompt { +- /** The text prompt that was filtered. */ +- text?: string; +- /** The reason the prompt was filtered. */ +- filteredReason?: string; +-} +- +-/** Configuration for music generation. */ +-export declare interface LiveMusicGenerationConfig { +- /** Controls the variance in audio generation. Higher values produce +- higher variance. Range is [0.0, 3.0]. */ +- temperature?: number; +- /** Controls how the model selects tokens for output. Samples the topK +- tokens with the highest probabilities. Range is [1, 1000]. */ +- topK?: number; +- /** Seeds audio generation. If not set, the request uses a randomly +- generated seed. */ +- seed?: number; +- /** Controls how closely the model follows prompts. +- Higher guidance follows more closely, but will make transitions more +- abrupt. Range is [0.0, 6.0]. */ +- guidance?: number; +- /** Beats per minute. Range is [60, 200]. */ +- bpm?: number; +- /** Density of sounds. Range is [0.0, 1.0]. */ +- density?: number; +- /** Brightness of the music. Range is [0.0, 1.0]. */ +- brightness?: number; +- /** Scale of the generated music. */ +- scale?: Scale; +- /** Whether the audio output should contain bass. */ +- muteBass?: boolean; +- /** Whether the audio output should contain drums. */ +- muteDrums?: boolean; +- /** Whether the audio output should contain only bass and drums. */ +- onlyBassAndDrums?: boolean; +- /** The mode of music generation. Default mode is QUALITY. */ +- musicGenerationMode?: MusicGenerationMode; +-} +- +-/** The playback control signal to apply to the music generation. */ +-export declare enum LiveMusicPlaybackControl { +- /** +- * This value is unused. +- */ +- PLAYBACK_CONTROL_UNSPECIFIED = "PLAYBACK_CONTROL_UNSPECIFIED", +- /** +- * Start generating the music. +- */ +- PLAY = "PLAY", +- /** +- * Hold the music generation. Use PLAY to resume from the current position. +- */ +- PAUSE = "PAUSE", +- /** +- * Stop the music generation and reset the context (prompts retained). +- Use PLAY to restart the music generation. +- */ +- STOP = "STOP", +- /** +- * Reset the context of the music generation without stopping it. +- Retains the current prompts and config. +- */ +- RESET_CONTEXT = "RESET_CONTEXT" +-} +- +-/** Server update generated by the model in response to client messages. +- +- Content is generated as quickly as possible, and not in real time. +- Clients may choose to buffer and play it out in real time. +- */ +-export declare interface LiveMusicServerContent { +- /** The audio chunks that the model has generated. */ +- audioChunks?: AudioChunk[]; +-} +- +-/** Response message for the LiveMusicClientMessage call. */ +-export declare class LiveMusicServerMessage { +- /** Message sent in response to a `LiveMusicClientSetup` message from the client. +- Clients should wait for this message before sending any additional messages. */ +- setupComplete?: LiveMusicServerSetupComplete; +- /** Content generated by the model in response to client messages. */ +- serverContent?: LiveMusicServerContent; +- /** A prompt that was filtered with the reason. */ +- filteredPrompt?: LiveMusicFilteredPrompt; +- /** +- * Returns the first audio chunk from the server content, if present. +- * +- * @remarks +- * If there are no audio chunks in the response, undefined will be returned. +- */ +- get audioChunk(): AudioChunk | undefined; +-} +- +-/** Sent in response to a `LiveMusicClientSetup` message from the client. */ +-export declare interface LiveMusicServerSetupComplete { +-} +- +-/** +- Represents a connection to the API. +- +- @experimental +- */ +-export declare class LiveMusicSession { +- readonly conn: WebSocket_2; +- private readonly apiClient; +- constructor(conn: WebSocket_2, apiClient: ApiClient); +- /** +- Sets inputs to steer music generation. Updates the session's current +- weighted prompts. +- +- @param params - Contains one property, `weightedPrompts`. +- +- - `weightedPrompts` to send to the model; weights are normalized to +- sum to 1.0. +- +- @experimental +- */ +- setWeightedPrompts(params: types.LiveMusicSetWeightedPromptsParameters): Promise; +- /** +- Sets a configuration to the model. Updates the session's current +- music generation config. +- +- @param params - Contains one property, `musicGenerationConfig`. +- +- - `musicGenerationConfig` to set in the model. Passing an empty or +- undefined config to the model will reset the config to defaults. +- +- @experimental +- */ +- setMusicGenerationConfig(params: types.LiveMusicSetConfigParameters): Promise; +- private sendPlaybackControl; +- /** +- * Start the music stream. +- * +- * @experimental +- */ +- play(): void; +- /** +- * Temporarily halt the music stream. Use `play` to resume from the current +- * position. +- * +- * @experimental +- */ +- pause(): void; +- /** +- * Stop the music stream and reset the state. Retains the current prompts +- * and config. +- * +- * @experimental +- */ +- stop(): void; +- /** +- * Resets the context of the music generation without stopping it. +- * Retains the current prompts and config. +- * +- * @experimental +- */ +- resetContext(): void; +- /** +- Terminates the WebSocket connection. +- +- @experimental +- */ +- close(): void; +-} +- +-/** Parameters for setting config for the live music API. */ +-export declare interface LiveMusicSetConfigParameters { +- /** Configuration for music generation. */ +- musicGenerationConfig: LiveMusicGenerationConfig; +-} +- +-/** Parameters for setting weighted prompts for the live music API. */ +-export declare interface LiveMusicSetWeightedPromptsParameters { +- /** A map of text prompts to weights to use for the generation request. */ +- weightedPrompts: WeightedPrompt[]; +-} +- +-/** Prompts and config used for generating this audio chunk. */ +-export declare interface LiveMusicSourceMetadata { +- /** Weighted prompts for generating this audio chunk. */ +- clientContent?: LiveMusicClientContent; +- /** Music generation config for generating this audio chunk. */ +- musicGenerationConfig?: LiveMusicGenerationConfig; +-} +- +-/** Parameters for sending client content to the live API. */ +-export declare interface LiveSendClientContentParameters { +- /** Client content to send to the session. */ +- turns?: ContentListUnion; +- /** If true, indicates that the server content generation should start with +- the currently accumulated prompt. Otherwise, the server will await +- additional messages before starting generation. */ +- turnComplete?: boolean; +-} +- +-/** Parameters for sending realtime input to the live API. */ +-export declare interface LiveSendRealtimeInputParameters { +- /** Realtime input to send to the session. */ +- media?: BlobImageUnion; +- /** The realtime audio input stream. */ +- audio?: Blob_2; +- /** +- Indicates that the audio stream has ended, e.g. because the microphone was +- turned off. +- +- This should only be sent when automatic activity detection is enabled +- (which is the default). +- +- The client can reopen the stream by sending an audio message. +- */ +- audioStreamEnd?: boolean; +- /** The realtime video input stream. */ +- video?: BlobImageUnion; +- /** The realtime text input stream. */ +- text?: string; +- /** Marks the start of user activity. */ +- activityStart?: ActivityStart; +- /** Marks the end of user activity. */ +- activityEnd?: ActivityEnd; +-} +- +-/** Parameters for sending tool responses to the live API. */ +-export declare class LiveSendToolResponseParameters { +- /** Tool responses to send to the session. */ +- functionResponses: FunctionResponse[] | FunctionResponse; +-} +- +-/** Incremental server update generated by the model in response to client messages. +- +- Content is generated as quickly as possible, and not in real time. Clients +- may choose to buffer and play it out in real time. +- */ +-export declare interface LiveServerContent { +- /** The content that the model has generated as part of the current conversation with the user. */ +- modelTurn?: Content; +- /** If true, indicates that the model is done generating. Generation will only start in response to additional client messages. Can be set alongside `content`, indicating that the `content` is the last in the turn. */ +- turnComplete?: boolean; +- /** If true, indicates that a client message has interrupted current model generation. If the client is playing out the content in realtime, this is a good signal to stop and empty the current queue. */ +- interrupted?: boolean; +- /** Metadata returned to client when grounding is enabled. */ +- groundingMetadata?: GroundingMetadata; +- /** If true, indicates that the model is done generating. When model is +- interrupted while generating there will be no generation_complete message +- in interrupted turn, it will go through interrupted > turn_complete. +- When model assumes realtime playback there will be delay between +- generation_complete and turn_complete that is caused by model +- waiting for playback to finish. If true, indicates that the model +- has finished generating all content. This is a signal to the client +- that it can stop sending messages. */ +- generationComplete?: boolean; +- /** Input transcription. The transcription is independent to the model +- turn which means it doesn’t imply any ordering between transcription and +- model turn. */ +- inputTranscription?: Transcription; +- /** Output transcription. The transcription is independent to the model +- turn which means it doesn’t imply any ordering between transcription and +- model turn. +- */ +- outputTranscription?: Transcription; +- /** Metadata related to url context retrieval tool. */ +- urlContextMetadata?: UrlContextMetadata; +-} +- +-/** Server will not be able to service client soon. */ +-export declare interface LiveServerGoAway { +- /** The remaining time before the connection will be terminated as ABORTED. The minimal time returned here is specified differently together with the rate limits for a given model. */ +- timeLeft?: string; +-} +- +-/** Response message for API call. */ +-export declare class LiveServerMessage { +- /** Sent in response to a `LiveClientSetup` message from the client. */ +- setupComplete?: LiveServerSetupComplete; +- /** Content generated by the model in response to client messages. */ +- serverContent?: LiveServerContent; +- /** Request for the client to execute the `function_calls` and return the responses with the matching `id`s. */ +- toolCall?: LiveServerToolCall; +- /** Notification for the client that a previously issued `ToolCallMessage` with the specified `id`s should have been not executed and should be cancelled. */ +- toolCallCancellation?: LiveServerToolCallCancellation; +- /** Usage metadata about model response(s). */ +- usageMetadata?: UsageMetadata; +- /** Server will disconnect soon. */ +- goAway?: LiveServerGoAway; +- /** Update of the session resumption state. */ +- sessionResumptionUpdate?: LiveServerSessionResumptionUpdate; +- /** +- * Returns the concatenation of all text parts from the server content if present. +- * +- * @remarks +- * If there are non-text parts in the response, the concatenation of all text +- * parts will be returned, and a warning will be logged. +- */ +- get text(): string | undefined; +- /** +- * Returns the concatenation of all inline data parts from the server content if present. +- * +- * @remarks +- * If there are non-inline data parts in the +- * response, the concatenation of all inline data parts will be returned, and +- * a warning will be logged. +- */ +- get data(): string | undefined; +-} +- +-/** Update of the session resumption state. +- +- Only sent if `session_resumption` was set in the connection config. +- */ +-export declare interface LiveServerSessionResumptionUpdate { +- /** New handle that represents state that can be resumed. Empty if `resumable`=false. */ +- newHandle?: string; +- /** True if session can be resumed at this point. It might be not possible to resume session at some points. In that case we send update empty new_handle and resumable=false. Example of such case could be model executing function calls or just generating. Resuming session (using previous session token) in such state will result in some data loss. */ +- resumable?: boolean; +- /** Index of last message sent by client that is included in state represented by this SessionResumptionToken. Only sent when `SessionResumptionConfig.transparent` is set. +- +- Presence of this index allows users to transparently reconnect and avoid issue of losing some part of realtime audio input/video. If client wishes to temporarily disconnect (for example as result of receiving GoAway) they can do it without losing state by buffering messages sent since last `SessionResmumptionTokenUpdate`. This field will enable them to limit buffering (avoid keeping all requests in RAM). +- +- Note: This should not be used for when resuming a session at some time later -- in those cases partial audio and video frames arelikely not needed. */ +- lastConsumedClientMessageIndex?: string; +-} +- +-export declare interface LiveServerSetupComplete { +-} +- +-/** Request for the client to execute the `function_calls` and return the responses with the matching `id`s. */ +-export declare interface LiveServerToolCall { +- /** The function call to be executed. */ +- functionCalls?: FunctionCall[]; +-} +- +-/** Notification for the client that a previously issued `ToolCallMessage` with the specified `id`s should have been not executed and should be cancelled. +- +- If there were side-effects to those tool calls, clients may attempt to undo +- the tool calls. This message occurs only in cases where the clients interrupt +- server turns. +- */ +-export declare interface LiveServerToolCallCancellation { +- /** The ids of the tool calls to be cancelled. */ +- ids?: string[]; +-} +- +-/** Logprobs Result */ +-export declare interface LogprobsResult { +- /** Length = total number of decoding steps. The chosen candidates may or may not be in top_candidates. */ +- chosenCandidates?: LogprobsResultCandidate[]; +- /** Length = total number of decoding steps. */ +- topCandidates?: LogprobsResultTopCandidates[]; +-} +- +-/** Candidate for the logprobs token and score. */ +-export declare interface LogprobsResultCandidate { +- /** The candidate's log probability. */ +- logProbability?: number; +- /** The candidate's token string value. */ +- token?: string; +- /** The candidate's token id value. */ +- tokenId?: number; +-} +- +-/** Candidates with top log probabilities at each decoding step. */ +-export declare interface LogprobsResultTopCandidates { +- /** Sorted by log probability in descending order. */ +- candidates?: LogprobsResultCandidate[]; +-} +- +-/** Configuration for a Mask reference image. */ +-export declare interface MaskReferenceConfig { +- /** Prompts the model to generate a mask instead of you needing to +- provide one (unless MASK_MODE_USER_PROVIDED is used). */ +- maskMode?: MaskReferenceMode; +- /** A list of up to 5 class ids to use for semantic segmentation. +- Automatically creates an image mask based on specific objects. */ +- segmentationClasses?: number[]; +- /** Dilation percentage of the mask provided. +- Float between 0 and 1. */ +- maskDilation?: number; +-} +- +-/** A mask reference image. +- +- This encapsulates either a mask image provided by the user and configs for +- the user provided mask, or only config parameters for the model to generate +- a mask. +- +- A mask image is an image whose non-zero values indicate where to edit the base +- image. If the user provides a mask image, the mask must be in the same +- dimensions as the raw image. +- */ +-export declare class MaskReferenceImage { +- /** The reference image for the editing operation. */ +- referenceImage?: Image_2; +- /** The id of the reference image. */ +- referenceId?: number; +- /** The type of the reference image. Only set by the SDK. */ +- referenceType?: string; +- /** Configuration for the mask reference image. */ +- config?: MaskReferenceConfig; +- /** Internal method to convert to ReferenceImageAPIInternal. */ +- toReferenceImageAPI(): any; +-} +- +-/** Enum representing the mask mode of a mask reference image. */ +-export declare enum MaskReferenceMode { +- MASK_MODE_DEFAULT = "MASK_MODE_DEFAULT", +- MASK_MODE_USER_PROVIDED = "MASK_MODE_USER_PROVIDED", +- MASK_MODE_BACKGROUND = "MASK_MODE_BACKGROUND", +- MASK_MODE_FOREGROUND = "MASK_MODE_FOREGROUND", +- MASK_MODE_SEMANTIC = "MASK_MODE_SEMANTIC" +-} +- +-/** +- * Creates a McpCallableTool from MCP clients and an optional config. +- * +- * The callable tool can invoke the MCP clients with given function call +- * arguments. (often for automatic function calling). +- * Use the config to modify tool parameters such as behavior. +- * +- * @experimental Built-in MCP support is an experimental feature, may change in future +- * versions. +- */ +-export declare function mcpToTool(...args: [...Client[], CallableToolConfig | Client]): CallableTool; +- +-/** Server content modalities. */ +-export declare enum MediaModality { +- /** +- * The modality is unspecified. +- */ +- MODALITY_UNSPECIFIED = "MODALITY_UNSPECIFIED", +- /** +- * Plain text. +- */ +- TEXT = "TEXT", +- /** +- * Images. +- */ +- IMAGE = "IMAGE", +- /** +- * Video. +- */ +- VIDEO = "VIDEO", +- /** +- * Audio. +- */ +- AUDIO = "AUDIO", +- /** +- * Document, e.g. PDF. +- */ +- DOCUMENT = "DOCUMENT" +-} +- +-/** The media resolution to use. */ +-export declare enum MediaResolution { +- /** +- * Media resolution has not been set +- */ +- MEDIA_RESOLUTION_UNSPECIFIED = "MEDIA_RESOLUTION_UNSPECIFIED", +- /** +- * Media resolution set to low (64 tokens). +- */ +- MEDIA_RESOLUTION_LOW = "MEDIA_RESOLUTION_LOW", +- /** +- * Media resolution set to medium (256 tokens). +- */ +- MEDIA_RESOLUTION_MEDIUM = "MEDIA_RESOLUTION_MEDIUM", +- /** +- * Media resolution set to high (zoomed reframing with 256 tokens). +- */ +- MEDIA_RESOLUTION_HIGH = "MEDIA_RESOLUTION_HIGH" +-} +- +-/** Server content modalities. */ +-export declare enum Modality { +- /** +- * The modality is unspecified. +- */ +- MODALITY_UNSPECIFIED = "MODALITY_UNSPECIFIED", +- /** +- * Indicates the model should return text +- */ +- TEXT = "TEXT", +- /** +- * Indicates the model should return images. +- */ +- IMAGE = "IMAGE", +- /** +- * Indicates the model should return images. +- */ +- AUDIO = "AUDIO" +-} +- +-/** Represents token counting info for a single modality. */ +-export declare interface ModalityTokenCount { +- /** The modality associated with this token count. */ +- modality?: MediaModality; +- /** Number of tokens. */ +- tokenCount?: number; +-} +- +-/** The mode of the predictor to be used in dynamic retrieval. */ +-export declare enum Mode { +- /** +- * Always trigger retrieval. +- */ +- MODE_UNSPECIFIED = "MODE_UNSPECIFIED", +- /** +- * Run retrieval only when system decides it is necessary. +- */ +- MODE_DYNAMIC = "MODE_DYNAMIC" +-} +- +-/** A trained machine learning model. */ +-export declare interface Model { +- /** Resource name of the model. */ +- name?: string; +- /** Display name of the model. */ +- displayName?: string; +- /** Description of the model. */ +- description?: string; +- /** Version ID of the model. A new version is committed when a new +- model version is uploaded or trained under an existing model ID. The +- version ID is an auto-incrementing decimal number in string +- representation. */ +- version?: string; +- /** List of deployed models created from this base model. Note that a +- model could have been deployed to endpoints in different locations. */ +- endpoints?: Endpoint[]; +- /** Labels with user-defined metadata to organize your models. */ +- labels?: Record; +- /** Information about the tuned model from the base model. */ +- tunedModelInfo?: TunedModelInfo; +- /** The maximum number of input tokens that the model can handle. */ +- inputTokenLimit?: number; +- /** The maximum number of output tokens that the model can generate. */ +- outputTokenLimit?: number; +- /** List of actions that are supported by the model. */ +- supportedActions?: string[]; +- /** The default checkpoint id of a model version. +- */ +- defaultCheckpointId?: string; +- /** The checkpoints of the model. */ +- checkpoints?: Checkpoint[]; +-} +- +-export declare class Models extends BaseModule { +- private readonly apiClient; +- constructor(apiClient: ApiClient); +- /** +- * Makes an API request to generate content with a given model. +- * +- * For the `model` parameter, supported formats for Vertex AI API include: +- * - The Gemini model ID, for example: 'gemini-2.0-flash' +- * - The full resource name starts with 'projects/', for example: +- * 'projects/my-project-id/locations/us-central1/publishers/google/models/gemini-2.0-flash' +- * - The partial resource name with 'publishers/', for example: +- * 'publishers/google/models/gemini-2.0-flash' or +- * 'publishers/meta/models/llama-3.1-405b-instruct-maas' +- * - `/` separated publisher and model name, for example: +- * 'google/gemini-2.0-flash' or 'meta/llama-3.1-405b-instruct-maas' +- * +- * For the `model` parameter, supported formats for Gemini API include: +- * - The Gemini model ID, for example: 'gemini-2.0-flash' +- * - The model name starts with 'models/', for example: +- * 'models/gemini-2.0-flash' +- * - For tuned models, the model name starts with 'tunedModels/', +- * for example: +- * 'tunedModels/1234567890123456789' +- * +- * Some models support multimodal input and output. +- * +- * @param params - The parameters for generating content. +- * @return The response from generating content. +- * +- * @example +- * ```ts +- * const response = await ai.models.generateContent({ +- * model: 'gemini-2.0-flash', +- * contents: 'why is the sky blue?', +- * config: { +- * candidateCount: 2, +- * } +- * }); +- * console.log(response); +- * ``` +- */ +- generateContent: (params: types.GenerateContentParameters) => Promise; +- /** +- * Makes an API request to generate content with a given model and yields the +- * response in chunks. +- * +- * For the `model` parameter, supported formats for Vertex AI API include: +- * - The Gemini model ID, for example: 'gemini-2.0-flash' +- * - The full resource name starts with 'projects/', for example: +- * 'projects/my-project-id/locations/us-central1/publishers/google/models/gemini-2.0-flash' +- * - The partial resource name with 'publishers/', for example: +- * 'publishers/google/models/gemini-2.0-flash' or +- * 'publishers/meta/models/llama-3.1-405b-instruct-maas' +- * - `/` separated publisher and model name, for example: +- * 'google/gemini-2.0-flash' or 'meta/llama-3.1-405b-instruct-maas' +- * +- * For the `model` parameter, supported formats for Gemini API include: +- * - The Gemini model ID, for example: 'gemini-2.0-flash' +- * - The model name starts with 'models/', for example: +- * 'models/gemini-2.0-flash' +- * - For tuned models, the model name starts with 'tunedModels/', +- * for example: +- * 'tunedModels/1234567890123456789' +- * +- * Some models support multimodal input and output. +- * +- * @param params - The parameters for generating content with streaming response. +- * @return The response from generating content. +- * +- * @example +- * ```ts +- * const response = await ai.models.generateContentStream({ +- * model: 'gemini-2.0-flash', +- * contents: 'why is the sky blue?', +- * config: { +- * maxOutputTokens: 200, +- * } +- * }); +- * for await (const chunk of response) { +- * console.log(chunk); +- * } +- * ``` +- */ +- generateContentStream: (params: types.GenerateContentParameters) => Promise>; +- /** +- * Transforms the CallableTools in the parameters to be simply Tools, it +- * copies the params into a new object and replaces the tools, it does not +- * modify the original params. Also sets the MCP usage header if there are +- * MCP tools in the parameters. +- */ +- private processParamsForMcpUsage; +- private initAfcToolsMap; +- private processAfcStream; +- /** +- * Generates an image based on a text description and configuration. +- * +- * @param params - The parameters for generating images. +- * @return The response from the API. +- * +- * @example +- * ```ts +- * const response = await client.models.generateImages({ +- * model: 'imagen-3.0-generate-002', +- * prompt: 'Robot holding a red skateboard', +- * config: { +- * numberOfImages: 1, +- * includeRaiReason: true, +- * }, +- * }); +- * console.log(response?.generatedImages?.[0]?.image?.imageBytes); +- * ``` +- */ +- generateImages: (params: types.GenerateImagesParameters) => Promise; +- list: (params?: types.ListModelsParameters) => Promise>; +- /** +- * Edits an image based on a prompt, list of reference images, and configuration. +- * +- * @param params - The parameters for editing an image. +- * @return The response from the API. +- * +- * @example +- * ```ts +- * const response = await client.models.editImage({ +- * model: 'imagen-3.0-capability-001', +- * prompt: 'Generate an image containing a mug with the product logo [1] visible on the side of the mug.', +- * referenceImages: [subjectReferenceImage] +- * config: { +- * numberOfImages: 1, +- * includeRaiReason: true, +- * }, +- * }); +- * console.log(response?.generatedImages?.[0]?.image?.imageBytes); +- * ``` +- */ +- editImage: (params: types.EditImageParameters) => Promise; +- /** +- * Upscales an image based on an image, upscale factor, and configuration. +- * Only supported in Vertex AI currently. +- * +- * @param params - The parameters for upscaling an image. +- * @return The response from the API. +- * +- * @example +- * ```ts +- * const response = await client.models.upscaleImage({ +- * model: 'imagen-3.0-generate-002', +- * image: image, +- * upscaleFactor: 'x2', +- * config: { +- * includeRaiReason: true, +- * }, +- * }); +- * console.log(response?.generatedImages?.[0]?.image?.imageBytes); +- * ``` +- */ +- upscaleImage: (params: types.UpscaleImageParameters) => Promise; +- private generateContentInternal; +- private generateContentStreamInternal; +- /** +- * Calculates embeddings for the given contents. Only text is supported. +- * +- * @param params - The parameters for embedding contents. +- * @return The response from the API. +- * +- * @example +- * ```ts +- * const response = await ai.models.embedContent({ +- * model: 'text-embedding-004', +- * contents: [ +- * 'What is your name?', +- * 'What is your favorite color?', +- * ], +- * config: { +- * outputDimensionality: 64, +- * }, +- * }); +- * console.log(response); +- * ``` +- */ +- embedContent(params: types.EmbedContentParameters): Promise; +- /** +- * Generates an image based on a text description and configuration. +- * +- * @param params - The parameters for generating images. +- * @return The response from the API. +- * +- * @example +- * ```ts +- * const response = await ai.models.generateImages({ +- * model: 'imagen-3.0-generate-002', +- * prompt: 'Robot holding a red skateboard', +- * config: { +- * numberOfImages: 1, +- * includeRaiReason: true, +- * }, +- * }); +- * console.log(response?.generatedImages?.[0]?.image?.imageBytes); +- * ``` +- */ +- private generateImagesInternal; +- private editImageInternal; +- private upscaleImageInternal; +- /** +- * Fetches information about a model by name. +- * +- * @example +- * ```ts +- * const modelInfo = await ai.models.get({model: 'gemini-2.0-flash'}); +- * ``` +- */ +- get(params: types.GetModelParameters): Promise; +- private listInternal; +- /** +- * Updates a tuned model by its name. +- * +- * @param params - The parameters for updating the model. +- * @return The response from the API. +- * +- * @example +- * ```ts +- * const response = await ai.models.update({ +- * model: 'tuned-model-name', +- * config: { +- * displayName: 'New display name', +- * description: 'New description', +- * }, +- * }); +- * ``` +- */ +- update(params: types.UpdateModelParameters): Promise; +- /** +- * Deletes a tuned model by its name. +- * +- * @param params - The parameters for deleting the model. +- * @return The response from the API. +- * +- * @example +- * ```ts +- * const response = await ai.models.delete({model: 'tuned-model-name'}); +- * ``` +- */ +- delete(params: types.DeleteModelParameters): Promise; +- /** +- * Counts the number of tokens in the given contents. Multimodal input is +- * supported for Gemini models. +- * +- * @param params - The parameters for counting tokens. +- * @return The response from the API. +- * +- * @example +- * ```ts +- * const response = await ai.models.countTokens({ +- * model: 'gemini-2.0-flash', +- * contents: 'The quick brown fox jumps over the lazy dog.' +- * }); +- * console.log(response); +- * ``` +- */ +- countTokens(params: types.CountTokensParameters): Promise; +- /** +- * Given a list of contents, returns a corresponding TokensInfo containing +- * the list of tokens and list of token ids. +- * +- * This method is not supported by the Gemini Developer API. +- * +- * @param params - The parameters for computing tokens. +- * @return The response from the API. +- * +- * @example +- * ```ts +- * const response = await ai.models.computeTokens({ +- * model: 'gemini-2.0-flash', +- * contents: 'What is your name?' +- * }); +- * console.log(response); +- * ``` +- */ +- computeTokens(params: types.ComputeTokensParameters): Promise; +- /** +- * Generates videos based on a text description and configuration. +- * +- * @param params - The parameters for generating videos. +- * @return A Promise which allows you to track the progress and eventually retrieve the generated videos using the operations.get method. +- * +- * @example +- * ```ts +- * const operation = await ai.models.generateVideos({ +- * model: 'veo-2.0-generate-001', +- * prompt: 'A neon hologram of a cat driving at top speed', +- * config: { +- * numberOfVideos: 1 +- * }); +- * +- * while (!operation.done) { +- * await new Promise(resolve => setTimeout(resolve, 10000)); +- * operation = await ai.operations.getVideosOperation({operation: operation}); +- * } +- * +- * console.log(operation.response?.generatedVideos?.[0]?.video?.uri); +- * ``` +- */ +- generateVideos(params: types.GenerateVideosParameters): Promise; +-} +- +-/** Config for model selection. */ +-export declare interface ModelSelectionConfig { +- /** Options for feature selection preference. */ +- featureSelectionPreference?: FeatureSelectionPreference; +-} +- +-/** The configuration for the multi-speaker setup. */ +-export declare interface MultiSpeakerVoiceConfig { +- /** The configuration for the speaker to use. */ +- speakerVoiceConfigs?: SpeakerVoiceConfig[]; +-} +- +-/** The mode of music generation. */ +-export declare enum MusicGenerationMode { +- /** +- * This value is unused. +- */ +- MUSIC_GENERATION_MODE_UNSPECIFIED = "MUSIC_GENERATION_MODE_UNSPECIFIED", +- /** +- * Steer text prompts to regions of latent space with higher quality +- music. +- */ +- QUALITY = "QUALITY", +- /** +- * Steer text prompts to regions of latent space with a larger diversity +- of music. +- */ +- DIVERSITY = "DIVERSITY" +-} +- +-/** A long-running operation. */ +-export declare interface Operation { +- /** The server-assigned name, which is only unique within the same service that originally returns it. If you use the default HTTP mapping, the `name` should be a resource name ending with `operations/{unique_id}`. */ +- name?: string; +- /** Service-specific metadata associated with the operation. It typically contains progress information and common metadata such as create time. Some services might not provide such metadata. Any method that returns a long-running operation should document the metadata type, if any. */ +- metadata?: Record; +- /** If the value is `false`, it means the operation is still in progress. If `true`, the operation is completed, and either `error` or `response` is available. */ +- done?: boolean; +- /** The error result of the operation in case of failure or cancellation. */ +- error?: Record; +-} +- +-/** Parameters for the get method of the operations module. */ +-export declare interface OperationGetParameters { +- /** The operation to be retrieved. */ +- operation: GenerateVideosOperation; +- /** Used to override the default configuration. */ +- config?: GetOperationConfig; +-} +- +-export declare class Operations extends BaseModule { +- private readonly apiClient; +- constructor(apiClient: ApiClient); +- /** +- * Gets the status of a long-running operation. +- * +- * @param parameters The parameters for the get operation request. +- * @return The updated Operation object, with the latest status or result. +- */ +- getVideosOperation(parameters: types.OperationGetParameters): Promise; +- private getVideosOperationInternal; +- private fetchPredictVideosOperationInternal; +-} +- +-/** +- * @license +- * Copyright 2025 Google LLC +- * SPDX-License-Identifier: Apache-2.0 +- */ +-/** Required. Outcome of the code execution. */ +-export declare enum Outcome { +- /** +- * Unspecified status. This value should not be used. +- */ +- OUTCOME_UNSPECIFIED = "OUTCOME_UNSPECIFIED", +- /** +- * Code execution completed successfully. +- */ +- OUTCOME_OK = "OUTCOME_OK", +- /** +- * Code execution finished but with a failure. `stderr` should contain the reason. +- */ +- OUTCOME_FAILED = "OUTCOME_FAILED", +- /** +- * Code execution ran for too long, and was cancelled. There may or may not be a partial output present. +- */ +- OUTCOME_DEADLINE_EXCEEDED = "OUTCOME_DEADLINE_EXCEEDED" +-} +- +-/** +- * @license +- * Copyright 2025 Google LLC +- * SPDX-License-Identifier: Apache-2.0 +- */ +-/** +- * Pagers for the GenAI List APIs. +- */ +-export declare enum PagedItem { +- PAGED_ITEM_BATCH_JOBS = "batchJobs", +- PAGED_ITEM_MODELS = "models", +- PAGED_ITEM_TUNING_JOBS = "tuningJobs", +- PAGED_ITEM_FILES = "files", +- PAGED_ITEM_CACHED_CONTENTS = "cachedContents" +-} +- +-declare interface PagedItemConfig { +- config?: { +- pageToken?: string; +- pageSize?: number; +- }; +-} +- +-declare interface PagedItemResponse { +- nextPageToken?: string; +- batchJobs?: T[]; +- models?: T[]; +- tuningJobs?: T[]; +- files?: T[]; +- cachedContents?: T[]; +-} +- +-/** +- * Pager class for iterating through paginated results. +- */ +-export declare class Pager implements AsyncIterable { +- private nameInternal; +- private pageInternal; +- private paramsInternal; +- private pageInternalSize; +- protected requestInternal: (params: PagedItemConfig) => Promise>; +- protected idxInternal: number; +- constructor(name: PagedItem, request: (params: PagedItemConfig) => Promise>, response: PagedItemResponse, params: PagedItemConfig); +- private init; +- private initNextPage; +- /** +- * Returns the current page, which is a list of items. +- * +- * @remarks +- * The first page is retrieved when the pager is created. The returned list of +- * items could be a subset of the entire list. +- */ +- get page(): T[]; +- /** +- * Returns the type of paged item (for example, ``batch_jobs``). +- */ +- get name(): PagedItem; +- /** +- * Returns the length of the page fetched each time by this pager. +- * +- * @remarks +- * The number of items in the page is less than or equal to the page length. +- */ +- get pageSize(): number; +- /** +- * Returns the parameters when making the API request for the next page. +- * +- * @remarks +- * Parameters contain a set of optional configs that can be +- * used to customize the API request. For example, the `pageToken` parameter +- * contains the token to request the next page. +- */ +- get params(): PagedItemConfig; +- /** +- * Returns the total number of items in the current page. +- */ +- get pageLength(): number; +- /** +- * Returns the item at the given index. +- */ +- getItem(index: number): T; +- /** +- * Returns an async iterator that support iterating through all items +- * retrieved from the API. +- * +- * @remarks +- * The iterator will automatically fetch the next page if there are more items +- * to fetch from the API. +- * +- * @example +- * +- * ```ts +- * const pager = await ai.files.list({config: {pageSize: 10}}); +- * for await (const file of pager) { +- * console.log(file.name); +- * } +- * ``` +- */ +- [Symbol.asyncIterator](): AsyncIterator; +- /** +- * Fetches the next page of items. This makes a new API request. +- * +- * @throws {Error} If there are no more pages to fetch. +- * +- * @example +- * +- * ```ts +- * const pager = await ai.files.list({config: {pageSize: 10}}); +- * let page = pager.page; +- * while (true) { +- * for (const file of page) { +- * console.log(file.name); +- * } +- * if (!pager.hasNextPage()) { +- * break; +- * } +- * page = await pager.nextPage(); +- * } +- * ``` +- */ +- nextPage(): Promise; +- /** +- * Returns true if there are more pages to fetch from the API. +- */ +- hasNextPage(): boolean; +-} +- +-/** A datatype containing media content. +- +- Exactly one field within a Part should be set, representing the specific type +- of content being conveyed. Using multiple fields within the same `Part` +- instance is considered invalid. +- */ +-export declare interface Part { +- /** Metadata for a given video. */ +- videoMetadata?: VideoMetadata; +- /** Indicates if the part is thought from the model. */ +- thought?: boolean; +- /** Optional. Inlined bytes data. */ +- inlineData?: Blob_2; +- /** Optional. Result of executing the [ExecutableCode]. */ +- codeExecutionResult?: CodeExecutionResult; +- /** Optional. Code generated by the model that is meant to be executed. */ +- executableCode?: ExecutableCode; +- /** Optional. URI based data. */ +- fileData?: FileData; +- /** Optional. A predicted [FunctionCall] returned from the model that contains a string representing the [FunctionDeclaration.name] with the parameters and their values. */ +- functionCall?: FunctionCall; +- /** Optional. The result output of a [FunctionCall] that contains a string representing the [FunctionDeclaration.name] and a structured JSON object containing any output from the function call. It is used as context to the model. */ +- functionResponse?: FunctionResponse; +- /** Optional. Text part (can be code). */ +- text?: string; +-} +- +-export declare type PartListUnion = PartUnion[] | PartUnion; +- +-/** Tuning spec for Partner models. */ +-export declare interface PartnerModelTuningSpec { +- /** Hyperparameters for tuning. The accepted hyper_parameters and their valid range of values will differ depending on the base model. */ +- hyperParameters?: Record; +- /** Required. Cloud Storage path to file containing training dataset for tuning. The dataset must be formatted as a JSONL file. */ +- trainingDatasetUri?: string; +- /** Optional. Cloud Storage path to file containing validation dataset for tuning. The dataset must be formatted as a JSONL file. */ +- validationDatasetUri?: string; +-} +- +-export declare type PartUnion = Part | string; +- +-/** Enum that controls the generation of people. */ +-export declare enum PersonGeneration { +- DONT_ALLOW = "DONT_ALLOW", +- ALLOW_ADULT = "ALLOW_ADULT", +- ALLOW_ALL = "ALLOW_ALL" +-} +- +-/** The configuration for the prebuilt speaker to use. */ +-export declare interface PrebuiltVoiceConfig { +- /** The name of the prebuilt voice to use. */ +- voiceName?: string; +-} +- +-/** Config for proactivity features. */ +-export declare interface ProactivityConfig { +- /** If enabled, the model can reject responding to the last prompt. For +- example, this allows the model to ignore out of context speech or to stay +- silent if the user did not make a request, yet. */ +- proactiveAudio?: boolean; +-} +- +-/** Specifies the context retrieval config. */ +-export declare interface RagRetrievalConfig { +- /** Optional. Config for filters. */ +- filter?: RagRetrievalConfigFilter; +- /** Optional. Config for Hybrid Search. */ +- hybridSearch?: RagRetrievalConfigHybridSearch; +- /** Optional. Config for ranking and reranking. */ +- ranking?: RagRetrievalConfigRanking; +- /** Optional. The number of contexts to retrieve. */ +- topK?: number; +-} +- +-/** Config for filters. */ +-export declare interface RagRetrievalConfigFilter { +- /** Optional. String for metadata filtering. */ +- metadataFilter?: string; +- /** Optional. Only returns contexts with vector distance smaller than the threshold. */ +- vectorDistanceThreshold?: number; +- /** Optional. Only returns contexts with vector similarity larger than the threshold. */ +- vectorSimilarityThreshold?: number; +-} +- +-/** Config for Hybrid Search. */ +-export declare interface RagRetrievalConfigHybridSearch { +- /** Optional. Alpha value controls the weight between dense and sparse vector search results. The range is [0, 1], while 0 means sparse vector search only and 1 means dense vector search only. The default value is 0.5 which balances sparse and dense vector search equally. */ +- alpha?: number; +-} +- +-/** Config for ranking and reranking. */ +-export declare interface RagRetrievalConfigRanking { +- /** Optional. Config for LlmRanker. */ +- llmRanker?: RagRetrievalConfigRankingLlmRanker; +- /** Optional. Config for Rank Service. */ +- rankService?: RagRetrievalConfigRankingRankService; +-} +- +-/** Config for LlmRanker. */ +-export declare interface RagRetrievalConfigRankingLlmRanker { +- /** Optional. The model name used for ranking. Format: `gemini-1.5-pro` */ +- modelName?: string; +-} +- +-/** Config for Rank Service. */ +-export declare interface RagRetrievalConfigRankingRankService { +- /** Optional. The model name of the rank service. Format: `semantic-ranker-512@latest` */ +- modelName?: string; +-} +- +-/** A raw reference image. +- +- A raw reference image represents the base image to edit, provided by the user. +- It can optionally be provided in addition to a mask reference image or +- a style reference image. +- */ +-export declare class RawReferenceImage { +- /** The reference image for the editing operation. */ +- referenceImage?: Image_2; +- /** The id of the reference image. */ +- referenceId?: number; +- /** The type of the reference image. Only set by the SDK. */ +- referenceType?: string; +- /** Internal method to convert to ReferenceImageAPIInternal. */ +- toReferenceImageAPI(): any; +-} +- +-/** Marks the end of user activity. +- +- This can only be sent if automatic (i.e. server-side) activity detection is +- disabled. +- */ +-export declare interface RealtimeInputConfig { +- /** If not set, automatic activity detection is enabled by default. If automatic voice detection is disabled, the client must send activity signals. */ +- automaticActivityDetection?: AutomaticActivityDetection; +- /** Defines what effect activity has. */ +- activityHandling?: ActivityHandling; +- /** Defines which input is included in the user's turn. */ +- turnCoverage?: TurnCoverage; +-} +- +-export declare type ReferenceImage = RawReferenceImage | MaskReferenceImage | ControlReferenceImage | StyleReferenceImage | SubjectReferenceImage; +- +-/** Represents a recorded session. */ +-export declare interface ReplayFile { +- replayId?: string; +- interactions?: ReplayInteraction[]; +-} +- +-/** Represents a single interaction, request and response in a replay. */ +-export declare interface ReplayInteraction { +- request?: ReplayRequest; +- response?: ReplayResponse; +-} +- +-/** Represents a single request in a replay. */ +-export declare interface ReplayRequest { +- method?: string; +- url?: string; +- headers?: Record; +- bodySegments?: Record[]; +-} +- +-/** Represents a single response in a replay. */ +-export declare class ReplayResponse { +- statusCode?: number; +- headers?: Record; +- bodySegments?: Record[]; +- sdkResponseSegments?: Record[]; +-} +- +-/** Defines a retrieval tool that model can call to access external knowledge. */ +-export declare interface Retrieval { +- /** Optional. Deprecated. This option is no longer supported. */ +- disableAttribution?: boolean; +- /** Set to use data source powered by Vertex AI Search. */ +- vertexAiSearch?: VertexAISearch; +- /** Set to use data source powered by Vertex RAG store. User data is uploaded via the VertexRagDataService. */ +- vertexRagStore?: VertexRagStore; +-} +- +-/** Retrieval config. +- */ +-export declare interface RetrievalConfig { +- /** Optional. The location of the user. */ +- latLng?: LatLng; +-} +- +-/** Metadata related to retrieval in the grounding flow. */ +-export declare interface RetrievalMetadata { +- /** Optional. Score indicating how likely information from Google Search could help answer the prompt. The score is in the range `[0, 1]`, where 0 is the least likely and 1 is the most likely. This score is only populated when Google Search grounding and dynamic retrieval is enabled. It will be compared to the threshold to determine whether to trigger Google Search. */ +- googleSearchDynamicRetrievalScore?: number; +-} +- +-/** Safety attributes of a GeneratedImage or the user-provided prompt. */ +-export declare interface SafetyAttributes { +- /** List of RAI categories. +- */ +- categories?: string[]; +- /** List of scores of each categories. +- */ +- scores?: number[]; +- /** Internal use only. +- */ +- contentType?: string; +-} +- +-/** Enum that controls the safety filter level for objectionable content. */ +-export declare enum SafetyFilterLevel { +- BLOCK_LOW_AND_ABOVE = "BLOCK_LOW_AND_ABOVE", +- BLOCK_MEDIUM_AND_ABOVE = "BLOCK_MEDIUM_AND_ABOVE", +- BLOCK_ONLY_HIGH = "BLOCK_ONLY_HIGH", +- BLOCK_NONE = "BLOCK_NONE" +-} +- +-/** Safety rating corresponding to the generated content. */ +-export declare interface SafetyRating { +- /** Output only. Indicates whether the content was filtered out because of this rating. */ +- blocked?: boolean; +- /** Output only. Harm category. */ +- category?: HarmCategory; +- /** Output only. Harm probability levels in the content. */ +- probability?: HarmProbability; +- /** Output only. Harm probability score. */ +- probabilityScore?: number; +- /** Output only. Harm severity levels in the content. */ +- severity?: HarmSeverity; +- /** Output only. Harm severity score. */ +- severityScore?: number; +-} +- +-/** Safety settings. */ +-export declare interface SafetySetting { +- /** Determines if the harm block method uses probability or probability +- and severity scores. */ +- method?: HarmBlockMethod; +- /** Required. Harm category. */ +- category?: HarmCategory; +- /** Required. The harm block threshold. */ +- threshold?: HarmBlockThreshold; +-} +- +-/** Scale of the generated music. */ +-export declare enum Scale { +- /** +- * Default value. This value is unused. +- */ +- SCALE_UNSPECIFIED = "SCALE_UNSPECIFIED", +- /** +- * C major or A minor. +- */ +- C_MAJOR_A_MINOR = "C_MAJOR_A_MINOR", +- /** +- * Db major or Bb minor. +- */ +- D_FLAT_MAJOR_B_FLAT_MINOR = "D_FLAT_MAJOR_B_FLAT_MINOR", +- /** +- * D major or B minor. +- */ +- D_MAJOR_B_MINOR = "D_MAJOR_B_MINOR", +- /** +- * Eb major or C minor +- */ +- E_FLAT_MAJOR_C_MINOR = "E_FLAT_MAJOR_C_MINOR", +- /** +- * E major or Db minor. +- */ +- E_MAJOR_D_FLAT_MINOR = "E_MAJOR_D_FLAT_MINOR", +- /** +- * F major or D minor. +- */ +- F_MAJOR_D_MINOR = "F_MAJOR_D_MINOR", +- /** +- * Gb major or Eb minor. +- */ +- G_FLAT_MAJOR_E_FLAT_MINOR = "G_FLAT_MAJOR_E_FLAT_MINOR", +- /** +- * G major or E minor. +- */ +- G_MAJOR_E_MINOR = "G_MAJOR_E_MINOR", +- /** +- * Ab major or F minor. +- */ +- A_FLAT_MAJOR_F_MINOR = "A_FLAT_MAJOR_F_MINOR", +- /** +- * A major or Gb minor. +- */ +- A_MAJOR_G_FLAT_MINOR = "A_MAJOR_G_FLAT_MINOR", +- /** +- * Bb major or G minor. +- */ +- B_FLAT_MAJOR_G_MINOR = "B_FLAT_MAJOR_G_MINOR", +- /** +- * B major or Ab minor. +- */ +- B_MAJOR_A_FLAT_MINOR = "B_MAJOR_A_FLAT_MINOR" +-} +- +-/** Schema is used to define the format of input/output data. Represents a select subset of an [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema-object). More fields may be added in the future as needed. */ +-export declare interface Schema { +- /** Optional. The value should be validated against any (one or more) of the subschemas in the list. */ +- anyOf?: Schema[]; +- /** Optional. Default value of the data. */ +- default?: unknown; +- /** Optional. The description of the data. */ +- description?: string; +- /** Optional. Possible values of the element of primitive type with enum format. Examples: 1. We can define direction as : {type:STRING, format:enum, enum:["EAST", NORTH", "SOUTH", "WEST"]} 2. We can define apartment number as : {type:INTEGER, format:enum, enum:["101", "201", "301"]} */ +- enum?: string[]; +- /** Optional. Example of the object. Will only populated when the object is the root. */ +- example?: unknown; +- /** Optional. The format of the data. Supported formats: for NUMBER type: "float", "double" for INTEGER type: "int32", "int64" for STRING type: "email", "byte", etc */ +- format?: string; +- /** Optional. SCHEMA FIELDS FOR TYPE ARRAY Schema of the elements of Type.ARRAY. */ +- items?: Schema; +- /** Optional. Maximum number of the elements for Type.ARRAY. */ +- maxItems?: string; +- /** Optional. Maximum length of the Type.STRING */ +- maxLength?: string; +- /** Optional. Maximum number of the properties for Type.OBJECT. */ +- maxProperties?: string; +- /** Optional. Maximum value of the Type.INTEGER and Type.NUMBER */ +- maximum?: number; +- /** Optional. Minimum number of the elements for Type.ARRAY. */ +- minItems?: string; +- /** Optional. SCHEMA FIELDS FOR TYPE STRING Minimum length of the Type.STRING */ +- minLength?: string; +- /** Optional. Minimum number of the properties for Type.OBJECT. */ +- minProperties?: string; +- /** Optional. SCHEMA FIELDS FOR TYPE INTEGER and NUMBER Minimum value of the Type.INTEGER and Type.NUMBER */ +- minimum?: number; +- /** Optional. Indicates if the value may be null. */ +- nullable?: boolean; +- /** Optional. Pattern of the Type.STRING to restrict a string to a regular expression. */ +- pattern?: string; +- /** Optional. SCHEMA FIELDS FOR TYPE OBJECT Properties of Type.OBJECT. */ +- properties?: Record; +- /** Optional. The order of the properties. Not a standard field in open api spec. Only used to support the order of the properties. */ +- propertyOrdering?: string[]; +- /** Optional. Required properties of Type.OBJECT. */ +- required?: string[]; +- /** Optional. The title of the Schema. */ +- title?: string; +- /** Optional. The type of the data. */ +- type?: Type; +-} +- +-export declare type SchemaUnion = Schema | unknown; +- +-/** Google search entry point. */ +-export declare interface SearchEntryPoint { +- /** Optional. Web content snippet that can be embedded in a web page or an app webview. */ +- renderedContent?: string; +- /** Optional. Base64 encoded JSON representing array of tuple. */ +- sdkBlob?: string; +-} +- +-/** Segment of the content. */ +-export declare interface Segment { +- /** Output only. End index in the given Part, measured in bytes. Offset from the start of the Part, exclusive, starting at zero. */ +- endIndex?: number; +- /** Output only. The index of a Part object within its parent Content object. */ +- partIndex?: number; +- /** Output only. Start index in the given Part, measured in bytes. Offset from the start of the Part, inclusive, starting at zero. */ +- startIndex?: number; +- /** Output only. The text corresponding to the segment from the response. */ +- text?: string; +-} +- +-/** Parameters for sending a message within a chat session. +- +- These parameters are used with the `chat.sendMessage()` method. +- */ +-export declare interface SendMessageParameters { +- /** The message to send to the model. +- +- The SDK will combine all parts into a single 'user' content to send to +- the model. +- */ +- message: PartListUnion; +- /** Config for this specific request. +- +- Please note that the per-request config does not change the chat level +- config, nor inherit from it. If you intend to use some values from the +- chat's default config, you must explicitly copy them into this per-request +- config. +- */ +- config?: GenerateContentConfig; +-} +- +-/** +- Represents a connection to the API. +- +- @experimental +- */ +-export declare class Session { +- readonly conn: WebSocket_2; +- private readonly apiClient; +- constructor(conn: WebSocket_2, apiClient: ApiClient); +- private tLiveClientContent; +- private tLiveClienttToolResponse; +- /** +- Send a message over the established connection. +- +- @param params - Contains two **optional** properties, `turns` and +- `turnComplete`. +- +- - `turns` will be converted to a `Content[]` +- - `turnComplete: true` [default] indicates that you are done sending +- content and expect a response. If `turnComplete: false`, the server +- will wait for additional messages before starting generation. +- +- @experimental +- +- @remarks +- There are two ways to send messages to the live API: +- `sendClientContent` and `sendRealtimeInput`. +- +- `sendClientContent` messages are added to the model context **in order**. +- Having a conversation using `sendClientContent` messages is roughly +- equivalent to using the `Chat.sendMessageStream`, except that the state of +- the `chat` history is stored on the API server instead of locally. +- +- Because of `sendClientContent`'s order guarantee, the model cannot respons +- as quickly to `sendClientContent` messages as to `sendRealtimeInput` +- messages. This makes the biggest difference when sending objects that have +- significant preprocessing time (typically images). +- +- The `sendClientContent` message sends a `Content[]` +- which has more options than the `Blob` sent by `sendRealtimeInput`. +- +- So the main use-cases for `sendClientContent` over `sendRealtimeInput` are: +- +- - Sending anything that can't be represented as a `Blob` (text, +- `sendClientContent({turns="Hello?"}`)). +- - Managing turns when not using audio input and voice activity detection. +- (`sendClientContent({turnComplete:true})` or the short form +- `sendClientContent()`) +- - Prefilling a conversation context +- ``` +- sendClientContent({ +- turns: [ +- Content({role:user, parts:...}), +- Content({role:user, parts:...}), +- ... +- ] +- }) +- ``` +- @experimental +- */ +- sendClientContent(params: types.LiveSendClientContentParameters): void; +- /** +- Send a realtime message over the established connection. +- +- @param params - Contains one property, `media`. +- +- - `media` will be converted to a `Blob` +- +- @experimental +- +- @remarks +- Use `sendRealtimeInput` for realtime audio chunks and video frames (images). +- +- With `sendRealtimeInput` the api will respond to audio automatically +- based on voice activity detection (VAD). +- +- `sendRealtimeInput` is optimized for responsivness at the expense of +- deterministic ordering guarantees. Audio and video tokens are to the +- context when they become available. +- +- Note: The Call signature expects a `Blob` object, but only a subset +- of audio and image mimetypes are allowed. +- */ +- sendRealtimeInput(params: types.LiveSendRealtimeInputParameters): void; +- /** +- Send a function response message over the established connection. +- +- @param params - Contains property `functionResponses`. +- +- - `functionResponses` will be converted to a `functionResponses[]` +- +- @remarks +- Use `sendFunctionResponse` to reply to `LiveServerToolCall` from the server. +- +- Use {@link types.LiveConnectConfig#tools} to configure the callable functions. +- +- @experimental +- */ +- sendToolResponse(params: types.LiveSendToolResponseParameters): void; +- /** +- Terminates the WebSocket connection. +- +- @experimental +- +- @example +- ```ts +- let model: string; +- if (GOOGLE_GENAI_USE_VERTEXAI) { +- model = 'gemini-2.0-flash-live-preview-04-09'; +- } else { +- model = 'gemini-2.0-flash-live-001'; +- } +- const session = await ai.live.connect({ +- model: model, +- config: { +- responseModalities: [Modality.AUDIO], +- } +- }); +- +- session.close(); +- ``` +- */ +- close(): void; +-} +- +-/** Configuration of session resumption mechanism. +- +- Included in `LiveConnectConfig.session_resumption`. If included server +- will send `LiveServerSessionResumptionUpdate` messages. +- */ +-export declare interface SessionResumptionConfig { +- /** Session resumption handle of previous session (session to restore). +- +- If not present new session will be started. */ +- handle?: string; +- /** If set the server will send `last_consumed_client_message_index` in the `session_resumption_update` messages to allow for transparent reconnections. */ +- transparent?: boolean; +-} +- +-/** +- * Overrides the base URLs for the Gemini API and Vertex AI API. +- * +- * @remarks This function should be called before initializing the SDK. If the +- * base URLs are set after initializing the SDK, the base URLs will not be +- * updated. Base URLs provided in the HttpOptions will also take precedence over +- * URLs set here. +- * +- * @example +- * ```ts +- * import {GoogleGenAI, setDefaultBaseUrls} from '@google/genai'; +- * // Override the base URL for the Gemini API. +- * setDefaultBaseUrls({geminiUrl:'https://gemini.google.com'}); +- * +- * // Override the base URL for the Vertex AI API. +- * setDefaultBaseUrls({vertexUrl: 'https://vertexai.googleapis.com'}); +- * +- * const ai = new GoogleGenAI({apiKey: 'GEMINI_API_KEY'}); +- * ``` +- */ +-export declare function setDefaultBaseUrls(baseUrlParams: BaseUrlParameters): void; +- +-/** Context window will be truncated by keeping only suffix of it. +- +- Context window will always be cut at start of USER role turn. System +- instructions and `BidiGenerateContentSetup.prefix_turns` will not be +- subject to the sliding window mechanism, they will always stay at the +- beginning of context window. +- */ +-export declare interface SlidingWindow { +- /** Session reduction target -- how many tokens we should keep. Window shortening operation has some latency costs, so we should avoid running it on every turn. Should be < trigger_tokens. If not set, trigger_tokens/2 is assumed. */ +- targetTokens?: string; +-} +- +-/** The configuration for the speaker to use. */ +-export declare interface SpeakerVoiceConfig { +- /** The name of the speaker to use. Should be the same as in the +- prompt. */ +- speaker?: string; +- /** The configuration for the voice to use. */ +- voiceConfig?: VoiceConfig; +-} +- +-/** The speech generation configuration. */ +-export declare interface SpeechConfig { +- /** The configuration for the speaker to use. +- */ +- voiceConfig?: VoiceConfig; +- /** The configuration for the multi-speaker setup. +- It is mutually exclusive with the voice_config field. +- */ +- multiSpeakerVoiceConfig?: MultiSpeakerVoiceConfig; +- /** Language code (ISO 639. e.g. en-US) for the speech synthesization. +- Only available for Live API. +- */ +- languageCode?: string; +-} +- +-export declare type SpeechConfigUnion = SpeechConfig | string; +- +-/** Start of speech sensitivity. */ +-export declare enum StartSensitivity { +- /** +- * The default is START_SENSITIVITY_LOW. +- */ +- START_SENSITIVITY_UNSPECIFIED = "START_SENSITIVITY_UNSPECIFIED", +- /** +- * Automatic detection will detect the start of speech more often. +- */ +- START_SENSITIVITY_HIGH = "START_SENSITIVITY_HIGH", +- /** +- * Automatic detection will detect the start of speech less often. +- */ +- START_SENSITIVITY_LOW = "START_SENSITIVITY_LOW" +-} +- +-/** Configuration for a Style reference image. */ +-export declare interface StyleReferenceConfig { +- /** A text description of the style to use for the generated image. */ +- styleDescription?: string; +-} +- +-/** A style reference image. +- +- This encapsulates a style reference image provided by the user, and +- additionally optional config parameters for the style reference image. +- +- A raw reference image can also be provided as a destination for the style to +- be applied to. +- */ +-export declare class StyleReferenceImage { +- /** The reference image for the editing operation. */ +- referenceImage?: Image_2; +- /** The id of the reference image. */ +- referenceId?: number; +- /** The type of the reference image. Only set by the SDK. */ +- referenceType?: string; +- /** Configuration for the style reference image. */ +- config?: StyleReferenceConfig; +- /** Internal method to convert to ReferenceImageAPIInternal. */ +- toReferenceImageAPI(): any; +-} +- +-/** Configuration for a Subject reference image. */ +-export declare interface SubjectReferenceConfig { +- /** The subject type of a subject reference image. */ +- subjectType?: SubjectReferenceType; +- /** Subject description for the image. */ +- subjectDescription?: string; +-} +- +-/** A subject reference image. +- +- This encapsulates a subject reference image provided by the user, and +- additionally optional config parameters for the subject reference image. +- +- A raw reference image can also be provided as a destination for the subject to +- be applied to. +- */ +-export declare class SubjectReferenceImage { +- /** The reference image for the editing operation. */ +- referenceImage?: Image_2; +- /** The id of the reference image. */ +- referenceId?: number; +- /** The type of the reference image. Only set by the SDK. */ +- referenceType?: string; +- /** Configuration for the subject reference image. */ +- config?: SubjectReferenceConfig; +- toReferenceImageAPI(): any; +-} +- +-/** Enum representing the subject type of a subject reference image. */ +-export declare enum SubjectReferenceType { +- SUBJECT_TYPE_DEFAULT = "SUBJECT_TYPE_DEFAULT", +- SUBJECT_TYPE_PERSON = "SUBJECT_TYPE_PERSON", +- SUBJECT_TYPE_ANIMAL = "SUBJECT_TYPE_ANIMAL", +- SUBJECT_TYPE_PRODUCT = "SUBJECT_TYPE_PRODUCT" +-} +- +-/** Hyperparameters for SFT. */ +-export declare interface SupervisedHyperParameters { +- /** Optional. Adapter size for tuning. */ +- adapterSize?: AdapterSize; +- /** Optional. Number of complete passes the model makes over the entire training dataset during training. */ +- epochCount?: string; +- /** Optional. Multiplier for adjusting the default learning rate. */ +- learningRateMultiplier?: number; +-} +- +-/** Dataset distribution for Supervised Tuning. */ +-export declare interface SupervisedTuningDatasetDistribution { +- /** Output only. Sum of a given population of values that are billable. */ +- billableSum?: string; +- /** Output only. Defines the histogram bucket. */ +- buckets?: SupervisedTuningDatasetDistributionDatasetBucket[]; +- /** Output only. The maximum of the population values. */ +- max?: number; +- /** Output only. The arithmetic mean of the values in the population. */ +- mean?: number; +- /** Output only. The median of the values in the population. */ +- median?: number; +- /** Output only. The minimum of the population values. */ +- min?: number; +- /** Output only. The 5th percentile of the values in the population. */ +- p5?: number; +- /** Output only. The 95th percentile of the values in the population. */ +- p95?: number; +- /** Output only. Sum of a given population of values. */ +- sum?: string; +-} +- +-/** Dataset bucket used to create a histogram for the distribution given a population of values. */ +-export declare interface SupervisedTuningDatasetDistributionDatasetBucket { +- /** Output only. Number of values in the bucket. */ +- count?: number; +- /** Output only. Left bound of the bucket. */ +- left?: number; +- /** Output only. Right bound of the bucket. */ +- right?: number; +-} +- +-/** Tuning data statistics for Supervised Tuning. */ +-export declare interface SupervisedTuningDataStats { +- /** Output only. Number of billable characters in the tuning dataset. */ +- totalBillableCharacterCount?: string; +- /** Output only. Number of billable tokens in the tuning dataset. */ +- totalBillableTokenCount?: string; +- /** The number of examples in the dataset that have been truncated by any amount. */ +- totalTruncatedExampleCount?: string; +- /** Output only. Number of tuning characters in the tuning dataset. */ +- totalTuningCharacterCount?: string; +- /** A partial sample of the indices (starting from 1) of the truncated examples. */ +- truncatedExampleIndices?: string[]; +- /** Output only. Number of examples in the tuning dataset. */ +- tuningDatasetExampleCount?: string; +- /** Output only. Number of tuning steps for this Tuning Job. */ +- tuningStepCount?: string; +- /** Output only. Sample user messages in the training dataset uri. */ +- userDatasetExamples?: Content[]; +- /** Output only. Dataset distributions for the user input tokens. */ +- userInputTokenDistribution?: SupervisedTuningDatasetDistribution; +- /** Output only. Dataset distributions for the messages per example. */ +- userMessagePerExampleDistribution?: SupervisedTuningDatasetDistribution; +- /** Output only. Dataset distributions for the user output tokens. */ +- userOutputTokenDistribution?: SupervisedTuningDatasetDistribution; +-} +- +-/** Tuning Spec for Supervised Tuning for first party models. */ +-export declare interface SupervisedTuningSpec { +- /** Optional. Hyperparameters for SFT. */ +- hyperParameters?: SupervisedHyperParameters; +- /** Required. Cloud Storage path to file containing training dataset for tuning. The dataset must be formatted as a JSONL file. */ +- trainingDatasetUri?: string; +- /** Optional. Cloud Storage path to file containing validation dataset for tuning. The dataset must be formatted as a JSONL file. */ +- validationDatasetUri?: string; +- /** Optional. If set to true, disable intermediate checkpoints for SFT and only the last checkpoint will be exported. */ +- exportLastCheckpointOnly?: boolean; +-} +- +-export declare interface TestTableFile { +- comment?: string; +- testMethod?: string; +- parameterNames?: string[]; +- testTable?: TestTableItem[]; +-} +- +-export declare interface TestTableItem { +- /** The name of the test. This is used to derive the replay id. */ +- name?: string; +- /** The parameters to the test. Use pydantic models. */ +- parameters?: Record; +- /** Expects an exception for MLDev matching the string. */ +- exceptionIfMldev?: string; +- /** Expects an exception for Vertex matching the string. */ +- exceptionIfVertex?: string; +- /** Use if you don't want to use the default replay id which is derived from the test name. */ +- overrideReplayId?: string; +- /** True if the parameters contain an unsupported union type. This test will be skipped for languages that do not support the union type. */ +- hasUnion?: boolean; +- /** When set to a reason string, this test will be skipped in the API mode. Use this flag for tests that can not be reproduced with the real API. E.g. a test that deletes a resource. */ +- skipInApiMode?: string; +- /** Keys to ignore when comparing the request and response. This is useful for tests that are not deterministic. */ +- ignoreKeys?: string[]; +-} +- +-/** The thinking features configuration. */ +-export declare interface ThinkingConfig { +- /** Indicates whether to include thoughts in the response. If true, thoughts are returned only if the model supports thought and thoughts are available. +- */ +- includeThoughts?: boolean; +- /** Indicates the thinking budget in tokens. +- */ +- thinkingBudget?: number; +-} +- +-/** Tokens info with a list of tokens and the corresponding list of token ids. */ +-export declare interface TokensInfo { +- /** Optional. Optional fields for the role from the corresponding Content. */ +- role?: string; +- /** A list of token ids from the input. */ +- tokenIds?: string[]; +- /** A list of tokens from the input. */ +- tokens?: string[]; +-} +- +-/** Tool details of a tool that the model may use to generate a response. */ +-export declare interface Tool { +- /** List of function declarations that the tool supports. */ +- functionDeclarations?: FunctionDeclaration[]; +- /** Optional. Retrieval tool type. System will always execute the provided retrieval tool(s) to get external knowledge to answer the prompt. Retrieval results are presented to the model for generation. */ +- retrieval?: Retrieval; +- /** Optional. Google Search tool type. Specialized retrieval tool +- that is powered by Google Search. */ +- googleSearch?: GoogleSearch; +- /** Optional. GoogleSearchRetrieval tool type. Specialized retrieval tool that is powered by Google search. */ +- googleSearchRetrieval?: GoogleSearchRetrieval; +- /** Optional. Enterprise web search tool type. Specialized retrieval +- tool that is powered by Vertex AI Search and Sec4 compliance. */ +- enterpriseWebSearch?: EnterpriseWebSearch; +- /** Optional. Google Maps tool type. Specialized retrieval tool +- that is powered by Google Maps. */ +- googleMaps?: GoogleMaps; +- /** Optional. Tool to support URL context retrieval. */ +- urlContext?: UrlContext; +- /** Optional. CodeExecution tool type. Enables the model to execute code as part of generation. This field is only used by the Gemini Developer API services. */ +- codeExecution?: ToolCodeExecution; +-} +- +-/** Tool that executes code generated by the model, and automatically returns the result to the model. See also [ExecutableCode]and [CodeExecutionResult] which are input and output to this tool. */ +-export declare interface ToolCodeExecution { +-} +- +-/** Tool config. +- +- This config is shared for all tools provided in the request. +- */ +-export declare interface ToolConfig { +- /** Optional. Function calling config. */ +- functionCallingConfig?: FunctionCallingConfig; +- /** Optional. Retrieval config. */ +- retrievalConfig?: RetrievalConfig; +-} +- +-export declare type ToolListUnion = ToolUnion[]; +- +-export declare type ToolUnion = Tool | CallableTool; +- +-/** Output only. Traffic type. This shows whether a request consumes Pay-As-You-Go or Provisioned Throughput quota. */ +-export declare enum TrafficType { +- /** +- * Unspecified request traffic type. +- */ +- TRAFFIC_TYPE_UNSPECIFIED = "TRAFFIC_TYPE_UNSPECIFIED", +- /** +- * Type for Pay-As-You-Go traffic. +- */ +- ON_DEMAND = "ON_DEMAND", +- /** +- * Type for Provisioned Throughput traffic. +- */ +- PROVISIONED_THROUGHPUT = "PROVISIONED_THROUGHPUT" +-} +- +-/** Audio transcription in Server Conent. */ +-export declare interface Transcription { +- /** Transcription text. +- */ +- text?: string; +- /** The bool indicates the end of the transcription. +- */ +- finished?: boolean; +-} +- +-export declare interface TunedModel { +- /** Output only. The resource name of the TunedModel. Format: `projects/{project}/locations/{location}/models/{model}`. */ +- model?: string; +- /** Output only. A resource name of an Endpoint. Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`. */ +- endpoint?: string; +- /** The checkpoints associated with this TunedModel. +- This field is only populated for tuning jobs that enable intermediate +- checkpoints. */ +- checkpoints?: TunedModelCheckpoint[]; +-} +- +-/** TunedModelCheckpoint for the Tuned Model of a Tuning Job. */ +-export declare interface TunedModelCheckpoint { +- /** The ID of the checkpoint. +- */ +- checkpointId?: string; +- /** The epoch of the checkpoint. +- */ +- epoch?: string; +- /** The step of the checkpoint. +- */ +- step?: string; +- /** The Endpoint resource name that the checkpoint is deployed to. +- Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`. +- */ +- endpoint?: string; +-} +- +-/** A tuned machine learning model. */ +-export declare interface TunedModelInfo { +- /** ID of the base model that you want to tune. */ +- baseModel?: string; +- /** Date and time when the base model was created. */ +- createTime?: string; +- /** Date and time when the base model was last updated. */ +- updateTime?: string; +-} +- +-/** Supervised fine-tuning training dataset. */ +-export declare interface TuningDataset { +- /** GCS URI of the file containing training dataset in JSONL format. */ +- gcsUri?: string; +- /** Inline examples with simple input/output text. */ +- examples?: TuningExample[]; +-} +- +-/** The tuning data statistic values for TuningJob. */ +-export declare interface TuningDataStats { +- /** Output only. Statistics for distillation. */ +- distillationDataStats?: DistillationDataStats; +- /** The SFT Tuning data stats. */ +- supervisedTuningDataStats?: SupervisedTuningDataStats; +-} +- +-export declare interface TuningExample { +- /** Text model input. */ +- textInput?: string; +- /** The expected model output. */ +- output?: string; +-} +- +-/** A tuning job. */ +-export declare interface TuningJob { +- /** Output only. Identifier. Resource name of a TuningJob. Format: `projects/{project}/locations/{location}/tuningJobs/{tuning_job}` */ +- name?: string; +- /** Output only. The detailed state of the job. */ +- state?: JobState; +- /** Output only. Time when the TuningJob was created. */ +- createTime?: string; +- /** Output only. Time when the TuningJob for the first time entered the `JOB_STATE_RUNNING` state. */ +- startTime?: string; +- /** Output only. Time when the TuningJob entered any of the following JobStates: `JOB_STATE_SUCCEEDED`, `JOB_STATE_FAILED`, `JOB_STATE_CANCELLED`, `JOB_STATE_EXPIRED`. */ +- endTime?: string; +- /** Output only. Time when the TuningJob was most recently updated. */ +- updateTime?: string; +- /** Output only. Only populated when job's state is `JOB_STATE_FAILED` or `JOB_STATE_CANCELLED`. */ +- error?: GoogleRpcStatus; +- /** Optional. The description of the TuningJob. */ +- description?: string; +- /** The base model that is being tuned, e.g., "gemini-1.0-pro-002". . */ +- baseModel?: string; +- /** Output only. The tuned model resources associated with this TuningJob. */ +- tunedModel?: TunedModel; +- /** Tuning Spec for Supervised Fine Tuning. */ +- supervisedTuningSpec?: SupervisedTuningSpec; +- /** Output only. The tuning data statistics associated with this TuningJob. */ +- tuningDataStats?: TuningDataStats; +- /** Customer-managed encryption key options for a TuningJob. If this is set, then all resources created by the TuningJob will be encrypted with the provided encryption key. */ +- encryptionSpec?: EncryptionSpec; +- /** Tuning Spec for open sourced and third party Partner models. */ +- partnerModelTuningSpec?: PartnerModelTuningSpec; +- /** Tuning Spec for Distillation. */ +- distillationSpec?: DistillationSpec; +- /** Output only. The Experiment associated with this TuningJob. */ +- experiment?: string; +- /** Optional. The labels with user-defined metadata to organize TuningJob and generated resources such as Model and Endpoint. Label keys and values can be no longer than 64 characters (Unicode codepoints), can only contain lowercase letters, numeric characters, underscores and dashes. International characters are allowed. See https://goo.gl/xmQnxf for more information and examples of labels. */ +- labels?: Record; +- /** Output only. The resource name of the PipelineJob associated with the TuningJob. Format: `projects/{project}/locations/{location}/pipelineJobs/{pipeline_job}`. */ +- pipelineJob?: string; +- /** Optional. The display name of the TunedModel. The name can be up to 128 characters long and can consist of any UTF-8 characters. */ +- tunedModelDisplayName?: string; +-} +- +-declare class Tunings extends BaseModule { +- private readonly apiClient; +- constructor(apiClient: ApiClient); +- /** +- * Gets a TuningJob. +- * +- * @param name - The resource name of the tuning job. +- * @return - A TuningJob object. +- * +- * @experimental - The SDK's tuning implementation is experimental, and may +- * change in future versions. +- */ +- get: (params: types.GetTuningJobParameters) => Promise; +- /** +- * Lists tuning jobs. +- * +- * @param config - The configuration for the list request. +- * @return - A list of tuning jobs. +- * +- * @experimental - The SDK's tuning implementation is experimental, and may +- * change in future versions. +- */ +- list: (params?: types.ListTuningJobsParameters) => Promise>; +- /** +- * Creates a supervised fine-tuning job. +- * +- * @param params - The parameters for the tuning job. +- * @return - A TuningJob operation. +- * +- * @experimental - The SDK's tuning implementation is experimental, and may +- * change in future versions. +- */ +- tune: (params: types.CreateTuningJobParameters) => Promise; +- private getInternal; +- private listInternal; +- private tuneInternal; +- private tuneMldevInternal; +-} +- +-export declare interface TuningValidationDataset { +- /** GCS URI of the file containing validation dataset in JSONL format. */ +- gcsUri?: string; +-} +- +-/** Options about which input is included in the user's turn. */ +-export declare enum TurnCoverage { +- /** +- * If unspecified, the default behavior is `TURN_INCLUDES_ONLY_ACTIVITY`. +- */ +- TURN_COVERAGE_UNSPECIFIED = "TURN_COVERAGE_UNSPECIFIED", +- /** +- * The users turn only includes activity since the last turn, excluding inactivity (e.g. silence on the audio stream). This is the default behavior. +- */ +- TURN_INCLUDES_ONLY_ACTIVITY = "TURN_INCLUDES_ONLY_ACTIVITY", +- /** +- * The users turn includes all realtime input since the last turn, including inactivity (e.g. silence on the audio stream). +- */ +- TURN_INCLUDES_ALL_INPUT = "TURN_INCLUDES_ALL_INPUT" +-} +- +-/** Optional. The type of the data. */ +-export declare enum Type { +- /** +- * Not specified, should not be used. +- */ +- TYPE_UNSPECIFIED = "TYPE_UNSPECIFIED", +- /** +- * OpenAPI string type +- */ +- STRING = "STRING", +- /** +- * OpenAPI number type +- */ +- NUMBER = "NUMBER", +- /** +- * OpenAPI integer type +- */ +- INTEGER = "INTEGER", +- /** +- * OpenAPI boolean type +- */ +- BOOLEAN = "BOOLEAN", +- /** +- * OpenAPI array type +- */ +- ARRAY = "ARRAY", +- /** +- * OpenAPI object type +- */ +- OBJECT = "OBJECT" +-} +- +-declare namespace types { +- export { +- createPartFromUri, +- createPartFromText, +- createPartFromFunctionCall, +- createPartFromFunctionResponse, +- createPartFromBase64, +- createPartFromCodeExecutionResult, +- createPartFromExecutableCode, +- createUserContent, +- createModelContent, +- Outcome, +- Language, +- HarmCategory, +- HarmBlockMethod, +- HarmBlockThreshold, +- Type, +- Mode, +- AuthType, +- FinishReason, +- HarmProbability, +- HarmSeverity, +- BlockedReason, +- TrafficType, +- Modality, +- MediaResolution, +- JobState, +- AdapterSize, +- FeatureSelectionPreference, +- Behavior, +- DynamicRetrievalConfigMode, +- FunctionCallingConfigMode, +- UrlRetrievalStatus, +- SafetyFilterLevel, +- PersonGeneration, +- ImagePromptLanguage, +- MaskReferenceMode, +- ControlReferenceType, +- SubjectReferenceType, +- EditMode, +- FileState, +- FileSource, +- MediaModality, +- StartSensitivity, +- EndSensitivity, +- ActivityHandling, +- TurnCoverage, +- FunctionResponseScheduling, +- Scale, +- MusicGenerationMode, +- LiveMusicPlaybackControl, +- VideoMetadata, +- Blob_2 as Blob, +- CodeExecutionResult, +- ExecutableCode, +- FileData, +- FunctionCall, +- FunctionResponse, +- Part, +- Content, +- HttpOptions, +- ModelSelectionConfig, +- SafetySetting, +- Schema, +- FunctionDeclaration, +- Interval, +- GoogleSearch, +- DynamicRetrievalConfig, +- GoogleSearchRetrieval, +- EnterpriseWebSearch, +- ApiKeyConfig, +- AuthConfigGoogleServiceAccountConfig, +- AuthConfigHttpBasicAuthConfig, +- AuthConfigOauthConfig, +- AuthConfigOidcConfig, +- AuthConfig, +- GoogleMaps, +- UrlContext, +- VertexAISearch, +- VertexRagStoreRagResource, +- RagRetrievalConfigFilter, +- RagRetrievalConfigHybridSearch, +- RagRetrievalConfigRankingLlmRanker, +- RagRetrievalConfigRankingRankService, +- RagRetrievalConfigRanking, +- RagRetrievalConfig, +- VertexRagStore, +- Retrieval, +- ToolCodeExecution, +- Tool, +- FunctionCallingConfig, +- LatLng, +- RetrievalConfig, +- ToolConfig, +- PrebuiltVoiceConfig, +- VoiceConfig, +- SpeakerVoiceConfig, +- MultiSpeakerVoiceConfig, +- SpeechConfig, +- AutomaticFunctionCallingConfig, +- ThinkingConfig, +- GenerationConfigRoutingConfigAutoRoutingMode, +- GenerationConfigRoutingConfigManualRoutingMode, +- GenerationConfigRoutingConfig, +- GenerateContentConfig, +- GenerateContentParameters, +- GoogleTypeDate, +- Citation, +- CitationMetadata, +- UrlMetadata, +- UrlContextMetadata, +- GroundingChunkRetrievedContext, +- GroundingChunkWeb, +- GroundingChunk, +- Segment, +- GroundingSupport, +- RetrievalMetadata, +- SearchEntryPoint, +- GroundingMetadata, +- LogprobsResultCandidate, +- LogprobsResultTopCandidates, +- LogprobsResult, +- SafetyRating, +- Candidate, +- GenerateContentResponsePromptFeedback, +- ModalityTokenCount, +- GenerateContentResponseUsageMetadata, +- GenerateContentResponse, +- ReferenceImage, +- EditImageParameters, +- EmbedContentConfig, +- EmbedContentParameters, +- ContentEmbeddingStatistics, +- ContentEmbedding, +- EmbedContentMetadata, +- EmbedContentResponse, +- GenerateImagesConfig, +- GenerateImagesParameters, +- Image_2 as Image, +- SafetyAttributes, +- GeneratedImage, +- GenerateImagesResponse, +- MaskReferenceConfig, +- ControlReferenceConfig, +- StyleReferenceConfig, +- SubjectReferenceConfig, +- EditImageConfig, +- EditImageResponse, +- UpscaleImageResponse, +- GetModelConfig, +- GetModelParameters, +- Endpoint, +- TunedModelInfo, +- Checkpoint, +- Model, +- ListModelsConfig, +- ListModelsParameters, +- ListModelsResponse, +- UpdateModelConfig, +- UpdateModelParameters, +- DeleteModelConfig, +- DeleteModelParameters, +- DeleteModelResponse, +- GenerationConfig, +- CountTokensConfig, +- CountTokensParameters, +- CountTokensResponse, +- ComputeTokensConfig, +- ComputeTokensParameters, +- TokensInfo, +- ComputeTokensResponse, +- GenerateVideosConfig, +- GenerateVideosParameters, +- Video, +- GeneratedVideo, +- GenerateVideosResponse, +- GenerateVideosOperation, +- GetTuningJobConfig, +- GetTuningJobParameters, +- TunedModelCheckpoint, +- TunedModel, +- GoogleRpcStatus, +- SupervisedHyperParameters, +- SupervisedTuningSpec, +- DatasetDistributionDistributionBucket, +- DatasetDistribution, +- DatasetStats, +- DistillationDataStats, +- SupervisedTuningDatasetDistributionDatasetBucket, +- SupervisedTuningDatasetDistribution, +- SupervisedTuningDataStats, +- TuningDataStats, +- EncryptionSpec, +- PartnerModelTuningSpec, +- DistillationHyperParameters, +- DistillationSpec, +- TuningJob, +- ListTuningJobsConfig, +- ListTuningJobsParameters, +- ListTuningJobsResponse, +- TuningExample, +- TuningDataset, +- TuningValidationDataset, +- CreateTuningJobConfig, +- CreateTuningJobParameters, +- Operation, +- CreateCachedContentConfig, +- CreateCachedContentParameters, +- CachedContentUsageMetadata, +- CachedContent, +- GetCachedContentConfig, +- GetCachedContentParameters, +- DeleteCachedContentConfig, +- DeleteCachedContentParameters, +- DeleteCachedContentResponse, +- UpdateCachedContentConfig, +- UpdateCachedContentParameters, +- ListCachedContentsConfig, +- ListCachedContentsParameters, +- ListCachedContentsResponse, +- ListFilesConfig, +- ListFilesParameters, +- FileStatus, +- File_2 as File, +- ListFilesResponse, +- CreateFileConfig, +- CreateFileParameters, +- HttpResponse, +- LiveCallbacks, +- CreateFileResponse, +- GetFileConfig, +- GetFileParameters, +- DeleteFileConfig, +- DeleteFileParameters, +- DeleteFileResponse, +- GetOperationConfig, +- GetOperationParameters, +- FetchPredictOperationConfig, +- FetchPredictOperationParameters, +- TestTableItem, +- TestTableFile, +- ReplayRequest, +- ReplayResponse, +- ReplayInteraction, +- ReplayFile, +- UploadFileConfig, +- DownloadFileConfig, +- DownloadFileParameters, +- UpscaleImageConfig, +- UpscaleImageParameters, +- RawReferenceImage, +- MaskReferenceImage, +- ControlReferenceImage, +- StyleReferenceImage, +- SubjectReferenceImage, +- LiveServerSetupComplete, +- Transcription, +- LiveServerContent, +- LiveServerToolCall, +- LiveServerToolCallCancellation, +- UsageMetadata, +- LiveServerGoAway, +- LiveServerSessionResumptionUpdate, +- LiveServerMessage, +- AutomaticActivityDetection, +- RealtimeInputConfig, +- SessionResumptionConfig, +- SlidingWindow, +- ContextWindowCompressionConfig, +- AudioTranscriptionConfig, +- ProactivityConfig, +- LiveClientSetup, +- LiveClientContent, +- ActivityStart, +- ActivityEnd, +- LiveClientRealtimeInput, +- LiveSendRealtimeInputParameters, +- LiveClientToolResponse, +- LiveClientMessage, +- LiveConnectConfig, +- LiveConnectParameters, +- CreateChatParameters, +- SendMessageParameters, +- LiveSendClientContentParameters, +- LiveSendToolResponseParameters, +- LiveMusicClientSetup, +- WeightedPrompt, +- LiveMusicClientContent, +- LiveMusicGenerationConfig, +- LiveMusicClientMessage, +- LiveMusicServerSetupComplete, +- LiveMusicSourceMetadata, +- AudioChunk, +- LiveMusicServerContent, +- LiveMusicFilteredPrompt, +- LiveMusicServerMessage, +- LiveMusicCallbacks, +- UploadFileParameters, +- CallableTool, +- CallableToolConfig, +- LiveMusicConnectParameters, +- LiveMusicSetConfigParameters, +- LiveMusicSetWeightedPromptsParameters, +- LiveEphemeralParameters, +- CreateAuthTokenConfig, +- OperationGetParameters, +- BlobImageUnion, +- PartUnion, +- PartListUnion, +- ContentUnion, +- ContentListUnion, +- SchemaUnion, +- SpeechConfigUnion, +- ToolUnion, +- ToolListUnion, +- DownloadableFileUnion +- } +-} +- +-/** Optional parameters for caches.update method. */ +-export declare interface UpdateCachedContentConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +- /** The TTL for this resource. The expiration time is computed: now + TTL. It is a duration string, with up to nine fractional digits, terminated by 's'. Example: "3.5s". */ +- ttl?: string; +- /** Timestamp of when this resource is considered expired. Uses RFC 3339 format, Example: 2014-10-02T15:01:23Z. */ +- expireTime?: string; +-} +- +-export declare interface UpdateCachedContentParameters { +- /** The server-generated resource name of the cached content. +- */ +- name: string; +- /** Configuration that contains optional parameters. +- */ +- config?: UpdateCachedContentConfig; +-} +- +-/** Configuration for updating a tuned model. */ +-export declare interface UpdateModelConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +- displayName?: string; +- description?: string; +- defaultCheckpointId?: string; +-} +- +-/** Configuration for updating a tuned model. */ +-export declare interface UpdateModelParameters { +- model: string; +- config?: UpdateModelConfig; +-} +- +-declare interface Uploader { +- /** +- * Uploads a file to the given upload url. +- * +- * @param file The file to upload. file is in string type or a Blob. +- * @param uploadUrl The upload URL as a string is where the file will be +- * uploaded to. The uploadUrl must be a url that was returned by the +- * https://generativelanguage.googleapis.com/upload/v1beta/files endpoint +- * @param apiClient The ApiClient to use for uploading. +- * @return A Promise that resolves to types.File. +- */ +- upload(file: string | Blob, uploadUrl: string, apiClient: ApiClient): Promise; +- /** +- * Returns the file's mimeType and the size of a given file. If the file is a +- * string path, the file type is determined by the file extension. If the +- * file's type cannot be determined, the type will be set to undefined. +- * +- * @param file The file to get the stat for. Can be a string path or a Blob. +- * @return A Promise that resolves to the file stat of the given file. +- */ +- stat(file: string | Blob): Promise; +-} +- +-/** Used to override the default configuration. */ +-export declare interface UploadFileConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +- /** The name of the file in the destination (e.g., 'files/sample-image'. If not provided one will be generated. */ +- name?: string; +- /** mime_type: The MIME type of the file. If not provided, it will be inferred from the file extension. */ +- mimeType?: string; +- /** Optional display name of the file. */ +- displayName?: string; +-} +- +-/** Parameters for the upload file method. */ +-export declare interface UploadFileParameters { +- /** The string path to the file to be uploaded or a Blob object. */ +- file: string | globalThis.Blob; +- /** Configuration that contains optional parameters. */ +- config?: UploadFileConfig; +-} +- +-/** Configuration for upscaling an image. +- +- For more information on this configuration, refer to +- the `Imagen API reference documentation +- `_. +- */ +-export declare interface UpscaleImageConfig { +- /** Used to override HTTP request options. */ +- httpOptions?: HttpOptions; +- /** Abort signal which can be used to cancel the request. +- +- NOTE: AbortSignal is a client-only operation. Using it to cancel an +- operation will not cancel the request in the service. You will still +- be charged usage for any applicable operations. +- */ +- abortSignal?: AbortSignal; +- /** Whether to include a reason for filtered-out images in the +- response. */ +- includeRaiReason?: boolean; +- /** The image format that the output should be saved as. */ +- outputMimeType?: string; +- /** The level of compression if the ``output_mime_type`` is +- ``image/jpeg``. */ +- outputCompressionQuality?: number; +-} +- +-/** User-facing config UpscaleImageParameters. */ +-export declare interface UpscaleImageParameters { +- /** The model to use. */ +- model: string; +- /** The input image to upscale. */ +- image: Image_2; +- /** The factor to upscale the image (x2 or x4). */ +- upscaleFactor: string; +- /** Configuration for upscaling. */ +- config?: UpscaleImageConfig; +-} +- +-export declare class UpscaleImageResponse { +- /** Generated images. */ +- generatedImages?: GeneratedImage[]; +-} +- +-/** Tool to support URL context retrieval. */ +-export declare interface UrlContext { +-} +- +-/** Metadata related to url context retrieval tool. */ +-export declare interface UrlContextMetadata { +- /** List of url context. */ +- urlMetadata?: UrlMetadata[]; +-} +- +-/** Context for a single url retrieval. */ +-export declare interface UrlMetadata { +- /** The URL retrieved by the tool. */ +- retrievedUrl?: string; +- /** Status of the url retrieval. */ +- urlRetrievalStatus?: UrlRetrievalStatus; +-} +- +-/** Status of the url retrieval. */ +-export declare enum UrlRetrievalStatus { +- /** +- * Default value. This value is unused +- */ +- URL_RETRIEVAL_STATUS_UNSPECIFIED = "URL_RETRIEVAL_STATUS_UNSPECIFIED", +- /** +- * Url retrieval is successful. +- */ +- URL_RETRIEVAL_STATUS_SUCCESS = "URL_RETRIEVAL_STATUS_SUCCESS", +- /** +- * Url retrieval is failed due to error. +- */ +- URL_RETRIEVAL_STATUS_ERROR = "URL_RETRIEVAL_STATUS_ERROR" +-} +- +-/** Usage metadata about response(s). */ +-export declare interface UsageMetadata { +- /** Number of tokens in the prompt. When `cached_content` is set, this is still the total effective prompt size meaning this includes the number of tokens in the cached content. */ +- promptTokenCount?: number; +- /** Number of tokens in the cached part of the prompt (the cached content). */ +- cachedContentTokenCount?: number; +- /** Total number of tokens across all the generated response candidates. */ +- responseTokenCount?: number; +- /** Number of tokens present in tool-use prompt(s). */ +- toolUsePromptTokenCount?: number; +- /** Number of tokens of thoughts for thinking models. */ +- thoughtsTokenCount?: number; +- /** Total token count for prompt, response candidates, and tool-use prompts(if present). */ +- totalTokenCount?: number; +- /** List of modalities that were processed in the request input. */ +- promptTokensDetails?: ModalityTokenCount[]; +- /** List of modalities that were processed in the cache input. */ +- cacheTokensDetails?: ModalityTokenCount[]; +- /** List of modalities that were returned in the response. */ +- responseTokensDetails?: ModalityTokenCount[]; +- /** List of modalities that were processed in the tool-use prompt. */ +- toolUsePromptTokensDetails?: ModalityTokenCount[]; +- /** Traffic type. This shows whether a request consumes Pay-As-You-Go +- or Provisioned Throughput quota. */ +- trafficType?: TrafficType; +-} +- +-/** Retrieve from Vertex AI Search datastore or engine for grounding. datastore and engine are mutually exclusive. See https://cloud.google.com/products/agent-builder */ +-export declare interface VertexAISearch { +- /** Optional. Fully-qualified Vertex AI Search data store resource ID. Format: `projects/{project}/locations/{location}/collections/{collection}/dataStores/{dataStore}` */ +- datastore?: string; +- /** Optional. Fully-qualified Vertex AI Search engine resource ID. Format: `projects/{project}/locations/{location}/collections/{collection}/engines/{engine}` */ +- engine?: string; +-} +- +-/** Retrieve from Vertex RAG Store for grounding. */ +-export declare interface VertexRagStore { +- /** Optional. Deprecated. Please use rag_resources instead. */ +- ragCorpora?: string[]; +- /** Optional. The representation of the rag source. It can be used to specify corpus only or ragfiles. Currently only support one corpus or multiple files from one corpus. In the future we may open up multiple corpora support. */ +- ragResources?: VertexRagStoreRagResource[]; +- /** Optional. The retrieval config for the Rag query. */ +- ragRetrievalConfig?: RagRetrievalConfig; +- /** Optional. Number of top k results to return from the selected corpora. */ +- similarityTopK?: number; +- /** Optional. Only return results with vector distance smaller than the threshold. */ +- vectorDistanceThreshold?: number; +-} +- +-/** The definition of the Rag resource. */ +-export declare interface VertexRagStoreRagResource { +- /** Optional. RagCorpora resource name. Format: `projects/{project}/locations/{location}/ragCorpora/{rag_corpus}` */ +- ragCorpus?: string; +- /** Optional. rag_file_id. The files should be in the same rag_corpus set in rag_corpus field. */ +- ragFileIds?: string[]; +-} +- +-/** A generated video. */ +-export declare interface Video { +- /** Path to another storage. */ +- uri?: string; +- /** Video bytes. */ +- videoBytes?: string; +- /** Video encoding, for example "video/mp4". */ +- mimeType?: string; +-} +- +-/** Describes how the video in the Part should be used by the model. */ +-export declare interface VideoMetadata { +- /** The frame rate of the video sent to the model. If not specified, the +- default value will be 1.0. The fps range is (0.0, 24.0]. */ +- fps?: number; +- /** Optional. The end offset of the video. */ +- endOffset?: string; +- /** Optional. The start offset of the video. */ +- startOffset?: string; +-} +- +-/** The configuration for the voice to use. */ +-export declare interface VoiceConfig { +- /** The configuration for the speaker to use. +- */ +- prebuiltVoiceConfig?: PrebuiltVoiceConfig; +-} +- +-declare interface WebSocket_2 { +- /** +- * Connects the socket to the server. +- */ +- connect(): void; +- /** +- * Sends a message to the server. +- */ +- send(message: string): void; +- /** +- * Closes the socket connection. +- */ +- close(): void; +-} +- +-/** +- * @license +- * Copyright 2025 Google LLC +- * SPDX-License-Identifier: Apache-2.0 +- */ +-declare interface WebSocketCallbacks { +- onopen: () => void; +- onerror: (e: any) => void; +- onmessage: (e: any) => void; +- onclose: (e: any) => void; +-} +- +-declare interface WebSocketFactory { +- /** +- * Returns a new WebSocket instance. +- */ +- create(url: string, headers: Record, callbacks: WebSocketCallbacks): WebSocket_2; +-} +- +-/** Maps a prompt to a relative weight to steer music generation. */ +-export declare interface WeightedPrompt { +- /** Text prompt. */ +- text?: string; +- /** Weight of the prompt. The weight is used to control the relative +- importance of the prompt. Higher weights are more important than lower +- weights. +- +- Weight must not be 0. Weights of all weighted_prompts in this +- LiveMusicClientContent message will be normalized. */ +- weight?: number; +-} +- +-export { } diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md index 9338bc035d..408057252b 100644 --- a/CONTRIBUTING.md +++ b/CONTRIBUTING.md @@ -1,4 +1,4 @@ -[中文](./docs/CONTRIBUTING.zh.md) | [English](./CONTRIBUTING.md) +[中文](docs/CONTRIBUTING.zh.md) | [English](CONTRIBUTING.md) # Cherry Studio Contributor Guide @@ -58,6 +58,10 @@ git commit --signoff -m "Your commit message" Maintainers are here to help you implement your use case within a reasonable timeframe. They will do their best to review your code and provide constructive feedback promptly. However, if you get stuck during the review process or feel your Pull Request is not receiving the attention it deserves, please contact us via comments in the Issue or through the [Community](README.md#-community). +### Participating in the Test Plan + +The Test Plan aims to provide users with a more stable application experience and faster iteration speed. For details, please refer to the [Test Plan](docs/testplan-en.md). + ### Other Suggestions - **Contact Developers**: Before submitting a PR, you can contact the developers first to discuss or get help. diff --git a/README.md b/README.md index 097f0b5756..3594915f34 100644 --- a/README.md +++ b/README.md @@ -47,6 +47,7 @@
[![][github-release-shield]][github-release-link] +[![][github-nightly-shield]][github-nightly-link] [![][github-contributors-shield]][github-contributors-link] [![][license-shield]][license-link] [![][commercial-shield]][commercial-link] @@ -182,7 +183,7 @@ Refer to the [Branching Strategy](docs/branching-strategy-en.md) for contributio 3. **Submit Changes**: Commit and push your changes. 4. **Open a Pull Request**: Describe your changes and reasons. -For more detailed guidelines, please refer to our [Contributing Guide](./CONTRIBUTING.md). +For more detailed guidelines, please refer to our [Contributing Guide](CONTRIBUTING.md). Thank you for your support and contributions! @@ -287,7 +288,7 @@ We believe the Enterprise Edition will become your team's AI productivity engine -[deepwiki-shield]: https://img.shields.io/badge/Deepwiki-CherryHQ-0088CC +[deepwiki-shield]: https://img.shields.io/badge/Deepwiki-CherryHQ-0088CC?logo=data:image/svg+xml;base64,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 [deepwiki-link]: https://deepwiki.com/CherryHQ/cherry-studio [twitter-shield]: https://img.shields.io/badge/Twitter-CherryStudioApp-0088CC?logo=x [twitter-link]: https://twitter.com/CherryStudioHQ @@ -298,9 +299,11 @@ We believe the Enterprise Edition will become your team's AI productivity engine -[github-release-shield]: https://img.shields.io/github/v/release/CherryHQ/cherry-studio +[github-release-shield]: https://img.shields.io/github/v/release/CherryHQ/cherry-studio?logo=github [github-release-link]: https://github.com/CherryHQ/cherry-studio/releases -[github-contributors-shield]: https://img.shields.io/github/contributors/CherryHQ/cherry-studio +[github-nightly-shield]: https://img.shields.io/github/actions/workflow/status/CherryHQ/cherry-studio/nightly-build.yml?label=nightly%20build&logo=github +[github-nightly-link]: https://github.com/CherryHQ/cherry-studio/actions/workflows/nightly-build.yml +[github-contributors-shield]: https://img.shields.io/github/contributors/CherryHQ/cherry-studio?logo=github [github-contributors-link]: https://github.com/CherryHQ/cherry-studio/graphs/contributors diff --git a/docs/CONTRIBUTING.zh.md b/docs/CONTRIBUTING.zh.md index 30b1983d18..7574990cd4 100644 --- a/docs/CONTRIBUTING.zh.md +++ b/docs/CONTRIBUTING.zh.md @@ -1,6 +1,6 @@ # Cherry Studio 贡献者指南 -[**English**](../CONTRIBUTING.md) | [**中文**](./CONTRIBUTING.zh.md) +[**English**](../CONTRIBUTING.md) | [**中文**](CONTRIBUTING.zh.md) 欢迎来到 Cherry Studio 的贡献者社区！我们致力于将 Cherry Studio 打造成一个长期提供价值的项目，并希望邀请更多的开发者加入我们的行列。无论您是经验丰富的开发者还是刚刚起步的初学者，您的贡献都将帮助我们更好地服务用户，提升软件质量。 @@ -24,7 +24,7 @@ ## 开始之前 -请确保阅读了[行为准则](CODE_OF_CONDUCT.md)和[LICENSE](LICENSE)。 +请确保阅读了[行为准则](../CODE_OF_CONDUCT.md)和[LICENSE](../LICENSE)。 ## 开始贡献 @@ -32,7 +32,7 @@ ### 测试 -未经测试的功能等同于不存在。为确保代码真正有效，应通过单元测试和功能测试覆盖相关流程。因此，在考虑贡献时，也请考虑可测试性。所有测试均可本地运行，无需依赖 CI。请参阅[开发者指南](docs/dev.md#test)中的“Test”部分。 +未经测试的功能等同于不存在。为确保代码真正有效，应通过单元测试和功能测试覆盖相关流程。因此，在考虑贡献时，也请考虑可测试性。所有测试均可本地运行，无需依赖 CI。请参阅[开发者指南](dev.md#test)中的“Test”部分。 ### 拉取请求的自动化测试 @@ -60,7 +60,11 @@ git commit --signoff -m "Your commit message" ### 获取代码审查/合并 -维护者在此帮助您在合理时间内实现您的用例。他们会尽力在合理时间内审查您的代码并提供建设性反馈。但如果您在审查过程中受阻，或认为您的 Pull Request 未得到应有的关注，请通过 Issue 中的评论或者[社群](README.md#-community)联系我们 +维护者在此帮助您在合理时间内实现您的用例。他们会尽力在合理时间内审查您的代码并提供建设性反馈。但如果您在审查过程中受阻，或认为您的 Pull Request 未得到应有的关注，请通过 Issue 中的评论或者[社群](README.zh.md#-community)联系我们 + +### 参与测试计划 + +测试计划旨在为用户提供更稳定的应用体验和更快的迭代速度，详细情况请参阅[测试计划](testplan-zh.md)。 ### 其他建议 diff --git a/docs/README.zh.md b/docs/README.zh.md index 7c1cccaa1f..774db66627 100644 --- a/docs/README.zh.md +++ b/docs/README.zh.md @@ -190,7 +190,7 @@ https://docs.cherry-ai.com 3. **提交更改**：提交并推送您的更改 4. **打开 Pull Request**：描述您的更改和原因 -有关更详细的指南，请参阅我们的 [贡献指南](./CONTRIBUTING.zh.md) +有关更详细的指南，请参阅我们的 [贡献指南](CONTRIBUTING.zh.md) 感谢您的支持和贡献！ diff --git a/docs/branching-strategy-en.md b/docs/branching-strategy-en.md index f3b7ddf508..8e646249ad 100644 --- a/docs/branching-strategy-en.md +++ b/docs/branching-strategy-en.md @@ -16,6 +16,8 @@ Cherry Studio implements a structured branching strategy to maintain code qualit - Only accepts documentation updates and bug fixes - Thoroughly tested before production deployment +For details about the `testplan` branch used in the Test Plan, please refer to the [Test Plan](testplan-en.md). + ## Contributing Branches When contributing to Cherry Studio, please follow these guidelines: diff --git a/docs/branching-strategy-zh.md b/docs/branching-strategy-zh.md index b1379537a5..36b7ca263d 100644 --- a/docs/branching-strategy-zh.md +++ b/docs/branching-strategy-zh.md @@ -16,6 +16,8 @@ Cherry Studio 采用结构化的分支策略来维护代码质量并简化开发 - 只接受文档更新和 bug 修复 - 经过完整测试后可以发布到生产环境 +关于测试计划所使用的`testplan`分支，请查阅[测试计划](testplan-zh.md)。 + ## 贡献分支 在为 Cherry Studio 贡献代码时，请遵循以下准则： diff --git a/docs/technical/db.settings.md b/docs/technical/db.settings.md new file mode 100644 index 0000000000..1d63098851 --- /dev/null +++ b/docs/technical/db.settings.md @@ -0,0 +1,11 @@ +# 数据库设置字段 + +此文档包含部分字段的数据类型说明。 + +## 字段 + +| 字段名 | 类型 | 说明 | +| ------------------------------ | ------------------------------ | ------------ | +| `translate:target:language` | `LanguageCode` | 翻译目标语言 | +| `translate:source:language` | `LanguageCode` | 翻译源语言 | +| `translate:bidirectional:pair` | `[LanguageCode, LanguageCode]` | 双向翻译对 | diff --git a/docs/testplan-en.md b/docs/testplan-en.md new file mode 100644 index 0000000000..0f7cd41473 --- /dev/null +++ b/docs/testplan-en.md @@ -0,0 +1,99 @@ +# Test Plan + +To provide users with a more stable application experience and faster iteration speed, Cherry Studio has launched the "Test Plan". + +## User Guide + +The Test Plan is divided into the RC channel and the Beta channel, with the following differences: + +- **RC (Release Candidate)**: The features are stable, with fewer bugs, and it is close to the official release. +- **Beta**: Features may change at any time, and there may be more bugs, but users can experience future features earlier. + +Users can enable the "Test Plan" and select the version channel in the software's `Settings` > `About`. Please note that the versions in the "Test Plan" cannot guarantee data consistency, so be sure to back up your data before using them. + +Users are welcome to submit issues or provide feedback through other channels for any bugs encountered during testing. Your feedback is very important to us. + +## Developer Guide + +### Participating in the Test Plan + +Developers should submit `PRs` according to the [Contributor Guide](../CONTRIBUTING.md) (and ensure the target branch is `main`). The repository maintainers will evaluate whether the `PR` should be included in the Test Plan based on factors such as the impact of the feature on the application, its importance, and whether broader testing is needed. + +If the `PR` is added to the Test Plan, the repository maintainers will: + +- Notify the `PR` submitter. +- Set the PR to `draft` status (to avoid accidental merging into `main` before testing is complete). +- Set the `milestone` to the specific Test Plan version. +- Modify the `PR` title. + +During participation in the Test Plan, `PR` submitters should: + +- Keep the `PR` branch synchronized with the latest `main` (i.e., the `PR` branch should always be based on the latest `main` code). +- Ensure the `PR` branch is conflict-free. +- Actively respond to comments & reviews and fix bugs. +- Enable maintainers to modify the `PR` branch to allow for bug fixes at any time. + +Inclusion in the Test Plan does not guarantee the final merging of the `PR`. It may be shelved due to immature features or poor testing feedback. + +### Test Plan Lead + +A maintainer will be assigned as the lead for a specific version (e.g., `1.5.0-rc`). The responsibilities of the Test Plan lead include: + +- Determining whether a `PR` meets the Test Plan requirements and deciding whether it should be included in the current Test Plan. +- Modifying the status of `PRs` added to the Test Plan and communicating relevant matters with the `PR` submitter. +- Before the Test Plan release, merging the branches of `PRs` added to the Test Plan (using squash merge) into the corresponding version branch of `testplan` and resolving conflicts. +- Ensuring the `testplan` branch is synchronized with the latest `main`. +- Overseeing the Test Plan release. + +## In-Depth Understanding + +### About `PRs` + +A `PR` is a collection of a specific branch (and commits), comments, reviews, and other information, and it is the **smallest management unit** of the Test Plan. + +Compared to submitting all features to a single branch, the Test Plan manages features through `PRs`, which offers greater flexibility and efficiency: + +- Features can be added or removed between different versions of the Test Plan without cumbersome `revert` operations. +- Clear feature boundaries and responsibilities are established. Bug fixes are completed within their respective `PRs`, isolating cross-impact and better tracking progress. +- The `PR` submitter is responsible for resolving conflicts with the latest `main`. The Test Plan lead is responsible for resolving conflicts between `PR` branches. However, since features added to the Test Plan are relatively independent (in other words, if a feature has broad implications, it should be independently included in the Test Plan), conflicts are generally few or simple. + +### The `testplan` Branch + +The `testplan` branch is a **temporary** branch used for Test Plan releases. + +Note: + +- **Do not develop based on this branch**. It may change or even be deleted at any time, and there is no guarantee of commit completeness or order. +- **Do not submit `commits` or `PRs` to this branch**, as they will not be retained. +- The `testplan` branch is always based on the latest `main` branch (not on a released version), with features added on top. + +#### RC Branch + +Branch name: `testplan/rc/x.y.z` + +Used for RC releases, where `x.y.z` is the target version number. Note that whether it is rc.1 or rc.5, as long as the major version number is `x.y.z`, it is completed in this branch. + +Generally, the version number for releases from this branch is named `x.y.z-rc.n`. + +#### Beta Branch + +Branch name: `testplan/beta/x.y.z` + +Used for Beta releases, where `x.y.z` is the target version number. Note that whether it is beta.1 or beta.5, as long as the major version number is `x.y.z`, it is completed in this branch. + +Generally, the version number for releases from this branch is named `x.y.z-beta.n`. + +### Version Rules + +The application version number for the Test Plan is: `x.y.z-CHA.n`, where: + +- `x.y.z` is the conventional version number, referred to here as the **target version number**. +- `CHA` is the channel code (Channel), currently divided into `rc` and `beta`. +- `n` is the release number, starting from `1`. + +Examples of complete version numbers: `1.5.0-rc.3`, `1.5.1-beta.1`, `1.6.0-beta.6`. + +The **target version number** of the Test Plan points to the official version number where these features are expected to be added. For example: + +- `1.5.0-rc.3` means this is a preview of the `1.5.0` official release (the current latest official release is `1.4.9`, and `1.5.0` has not yet been officially released). +- `1.5.1-beta.1` means this is a beta version of the `1.5.1` official release (the current latest official release is `1.5.0`, and `1.5.1` has not yet been officially released). diff --git a/docs/testplan-zh.md b/docs/testplan-zh.md new file mode 100644 index 0000000000..ed4913d4a4 --- /dev/null +++ b/docs/testplan-zh.md @@ -0,0 +1,99 @@ +# 测试计划 + +为了给用户提供更稳定的应用体验，并提供更快的迭代速度，Cherry Studio推出“测试计划”。 + +## 用户指南 + +测试计划分为RC版通道和Beta版通道吗，区别在于： + +- **RC版（预览版）**：RC即Release Candidate，功能已经稳定，BUG较少，接近正式版 +- **Beta版（测试版）**：功能可能随时变化，BUG较多，可以较早体验未来功能 + +用户可以在软件的`设置`-`关于`中，开启“测试计划”并选择版本通道。请注意“测试计划”的版本无法保证数据的一致性，请使用前一定要备份数据。 + +用户在测试过程中发现的BUG，欢迎提交issue或通过其他渠道反馈。用户的反馈对我们非常重要。 + +## 开发者指南 + +### 参与测试计划 + +开发者按照[贡献者指南](CONTRIBUTING.zh.md)要求正常提交`PR`（并注意提交target为`main`）。仓库维护者会综合考虑（例如该功能对应用的影响程度，功能的重要性，是否需要更广泛的测试等），决定该`PR`是否应加入测试计划。 + +若该`PR`加入测试计划，仓库维护者会做如下操作： + +- 通知`PR`提交人 +- 设置PR为`draft`状态（避免在测试完成前意外并入`main`） +- `milestone`设置为具体测试计划版本 +- 修改`PR`标题 + +`PR`提交人在参与测试计划过程中，应做到： + +- 保持`PR`分支与最新`main`同步（即`PR`分支总是应基于最新`main`代码） +- 保持`PR`分支为无冲突状态 +- 积极响应 comments & reviews，修复bug +- 开启维护者可以修改`PR`分支的权限，以便维护者能随时修改BUG + +加入测试计划并不保证`PR`的最终合并，也有可能由于功能不成熟或测试反馈不佳而搁置 + +### 测试计划负责人 + +某个维护者会被指定为某个版本期间（例如`1.5.0-rc`）的测试计划负责人。测试计划负责人的工作为： + +- 判断某个`PR`是否符合测试计划要求，并决定是否应合入当期测试计划 +- 修改加入测试计划的`PR`状态，并与`PR`提交人沟通相关事宜 +- 在测试计划发版前，将加入测试计划的`PR`分支逐一合并（采用squash merge）至`testplan`对应版本分支，并解决冲突 +- 保证`testplan`分支与最新`main`同步 +- 负责测试计划发版 + +## 深入理解 + +### 关于`PR` + +`PR`是特定分支（及commits）、comments、reviews等各种信息的集合，也是测试计划的**最小管理单元**。 + +相比将所有功能都提交到某个分支，测试计划通过`PR`来管理功能，这可以带来极大的灵活度和效率： + +- 测试计划的各个版本间，可以随意增减功能，而无需繁琐的`revert`操作 +- 明确了功能边界和负责人，bug修复在各自`PR`中完成，隔离了交叉影响，也能更好观察进度 +- `PR`提交人负责与最新`main`之间的冲突；测试计划负责人负责各`PR`分支之间的冲突，但因加入测试计划的各功能相对比较独立（话句话说，如果功能牵涉较广，则应独立上测试计划），冲突一般比较少或简单。 + +### `testplan`分支 + +`testplan`分支是用于测试计划发版所用的**临时**分支。 + +注意： + +- **请勿基于该分支开发**。该分支随时会变化甚至删除，且并不保证commit的完整和顺序。 +- **请勿向该分支提交`commit`及`PR`**，将不会得到保留 +- `testplan`分支总是基于最新`main`分支（而不是基于已发布版本），在其之上添加功能 + +#### RC版分支 + +分支名称：`testplan/rc/x.y.z` + +用于RC版的发版，x.y.z为目标版本号，注意无论是rc.1还是rc.5，只要主版本号为x.y.z，都在该分支完成。 + +一般而言，该分支发版的版本号命名为`x.y.z-rc.n` + +#### Beta版分支 + +分支名称：`testplan/beta/x.y.z` + +用于Beta版的发版，x.y.z为目标版本号，注意无论是beta.1还是beta.5，只要主版本号为x.y.z，都在该分支完成。 + +一般而言，该分支发版的版本号命名为`x.y.z-beta.n` + +### 版本规则 + +测试计划的应用版本号为：`x.y.z-CHA.n`，其中： + +- `x.y.z`为一般意义上的版本号，在这里称为**目标版本号** +- `CHA`为通道号（Channel），现在分为`rc`和`beta` +- `n`为发版编号，从`1`计数 + +完整的版本号举例：`1.5.0-rc.3`、`1.5.1-beta.1`、`1.6.0-beta.6` + +测试计划的**目标版本号**指向希望添加这些功能的正式版版本号。例如： + +- `1.5.0-rc.3`是指，这是`1.5.0`正式版的预览版（当前最新正式版是`1.4.9`，而`1.5.0`正式版还未发布） +- `1.5.1-beta.1`是指，这是`1.5.1`正式版的测试版（当前最新正式版是`1.5.0`，而`1.5.1`正式版还未发布） diff --git a/electron-builder.yml b/electron-builder.yml index 170c026f49..2aa58ecb8d 100644 --- a/electron-builder.yml +++ b/electron-builder.yml @@ -117,9 +117,9 @@ afterSign: scripts/notarize.js artifactBuildCompleted: scripts/artifact-build-completed.js releaseInfo: releaseNotes: | - 划词助手：支持 macOS 系统 - 文档处理：增加 MinerU、Doc2x，Mistral 等服务商支持 - 知识库：新的知识库界面，增加扫描版 PDF 支持 - OCR：macOS 增加系统 OCR 支持 - 服务商：支持一键添加服务商，新增 PH8 大模型开放平台, 支持 PPIO OAuth 登录 - 修复：Linux下数据目录移动问题 + 服务商：新增 NewAPI 服务商支持 + 绘图：新增 NewAPI 绘图服务商支持 + 备份：支持 s3 兼容存储备份 + 服务商：支持多个密钥管理，支持配置自定义请求头 + 设置：支持禁用硬件加速 + 其他：性能优化和错误改进 diff --git a/eslint.config.mjs b/eslint.config.mjs index 33e6ae8757..e0a893527e 100644 --- a/eslint.config.mjs +++ b/eslint.config.mjs @@ -26,7 +26,7 @@ export default defineConfig([ 'simple-import-sort/exports': 'error', 'unused-imports/no-unused-imports': 'error', '@eslint-react/no-prop-types': 'error', - 'prettier/prettier': ['error', { endOfLine: 'auto' }] + 'prettier/prettier': ['error'] } }, // Configuration for ensuring compatibility with the original ESLint(8.x) rules diff --git a/package.json b/package.json index d5cbc6e707..41f1f58bf7 100644 --- a/package.json +++ b/package.json @@ -1,6 +1,6 @@ { "name": "CherryStudio", - "version": "1.4.8", + "version": "1.4.9", "private": true, "description": "A powerful AI assistant for producer.", "main": "./out/main/index.js", @@ -55,20 +55,23 @@ "test:lint": "eslint . --ext .js,.jsx,.cjs,.mjs,.ts,.tsx,.cts,.mts", "format": "prettier --write .", "lint": "eslint . --ext .js,.jsx,.cjs,.mjs,.ts,.tsx,.cts,.mts --fix", - "prepare": "husky" + "prepare": "git config blame.ignoreRevsFile .git-blame-ignore-revs && husky" }, "dependencies": { + "@aws-sdk/client-s3": "^3.840.0", "@cherrystudio/pdf-to-img-napi": "^0.0.1", "@libsql/client": "0.14.0", "@libsql/win32-x64-msvc": "^0.4.7", "@strongtz/win32-arm64-msvc": "^0.4.7", + "iconv-lite": "^0.6.3", + "jschardet": "^3.1.4", "jsdom": "26.1.0", "macos-release": "^3.4.0", "node-stream-zip": "^1.15.0", "notion-helper": "^1.3.22", "os-proxy-config": "^1.1.2", "pdfjs-dist": "4.10.38", - "selection-hook": "^1.0.4", + "selection-hook": "^1.0.5", "turndown": "7.2.0" }, "devDependencies": { @@ -104,7 +107,7 @@ "@langchain/community": "^0.3.36", "@langchain/ollama": "^0.2.1", "@mistralai/mistralai": "^1.6.0", - "@modelcontextprotocol/sdk": "^1.11.4", + "@modelcontextprotocol/sdk": "^1.12.3", "@mozilla/readability": "^0.6.0", "@notionhq/client": "^2.2.15", "@playwright/test": "^1.52.0", diff --git a/packages/shared/IpcChannel.ts b/packages/shared/IpcChannel.ts index 7dd60bab06..78208133d5 100644 --- a/packages/shared/IpcChannel.ts +++ b/packages/shared/IpcChannel.ts @@ -36,6 +36,7 @@ export enum IpcChannel { App_MacRequestProcessTrust = 'app:mac-request-process-trust', App_QuoteToMain = 'app:quote-to-main', + App_SetDisableHardwareAcceleration = 'app:set-disable-hardware-acceleration', Notification_Send = 'notification:send', Notification_OnClick = 'notification:on-click', @@ -73,6 +74,8 @@ export enum IpcChannel { Mcp_ServersChanged = 'mcp:servers-changed', Mcp_ServersUpdated = 'mcp:servers-updated', Mcp_CheckConnectivity = 'mcp:check-connectivity', + Mcp_SetProgress = 'mcp:set-progress', + Mcp_AbortTool = 'mcp:abort-tool', // Python Python_Execute = 'python:execute', @@ -164,6 +167,16 @@ export enum IpcChannel { Backup_CheckConnection = 'backup:checkConnection', Backup_CreateDirectory = 'backup:createDirectory', Backup_DeleteWebdavFile = 'backup:deleteWebdavFile', + Backup_BackupToLocalDir = 'backup:backupToLocalDir', + Backup_RestoreFromLocalBackup = 'backup:restoreFromLocalBackup', + Backup_ListLocalBackupFiles = 'backup:listLocalBackupFiles', + Backup_DeleteLocalBackupFile = 'backup:deleteLocalBackupFile', + Backup_SetLocalBackupDir = 'backup:setLocalBackupDir', + Backup_BackupToS3 = 'backup:backupToS3', + Backup_RestoreFromS3 = 'backup:restoreFromS3', + Backup_ListS3Files = 'backup:listS3Files', + Backup_DeleteS3File = 'backup:deleteS3File', + Backup_CheckS3Connection = 'backup:checkS3Connection', // zip Zip_Compress = 'zip:compress', diff --git a/src/main/index.ts b/src/main/index.ts index 46ebd7c6e6..e022bb71a8 100644 --- a/src/main/index.ts +++ b/src/main/index.ts @@ -28,6 +28,14 @@ import { windowService } from './services/WindowService' Logger.initialize() +/** + * Disable hardware acceleration if setting is enabled + */ +const disableHardwareAcceleration = configManager.getDisableHardwareAcceleration() +if (disableHardwareAcceleration) { + app.disableHardwareAcceleration() +} + /** * Disable chromium's window animations * main purpose for this is to avoid the transparent window flashing when it is shown diff --git a/src/main/ipc.ts b/src/main/ipc.ts index a9c5169096..be160c8d2a 100644 --- a/src/main/ipc.ts +++ b/src/main/ipc.ts @@ -12,6 +12,7 @@ import { BrowserWindow, dialog, ipcMain, session, shell, systemPreferences, webC import log from 'electron-log' import { Notification } from 'src/renderer/src/types/notification' +import appService from './services/AppService' import AppUpdater from './services/AppUpdater' import BackupManager from './services/BackupManager' import { configManager } from './services/ConfigManager' @@ -114,12 +115,8 @@ export function registerIpc(mainWindow: BrowserWindow, app: Electron.App) { }) // launch on boot - ipcMain.handle(IpcChannel.App_SetLaunchOnBoot, (_, openAtLogin: boolean) => { - // Set login item settings for windows and mac - // linux is not supported because it requires more file operations - if (isWin || isMac) { - app.setLoginItemSettings({ openAtLogin }) - } + ipcMain.handle(IpcChannel.App_SetLaunchOnBoot, (_, isLaunchOnBoot: boolean) => { + appService.setAppLaunchOnBoot(isLaunchOnBoot) }) // launch to tray @@ -368,6 +365,16 @@ export function registerIpc(mainWindow: BrowserWindow, app: Electron.App) { ipcMain.handle(IpcChannel.Backup_CheckConnection, backupManager.checkConnection) ipcMain.handle(IpcChannel.Backup_CreateDirectory, backupManager.createDirectory) ipcMain.handle(IpcChannel.Backup_DeleteWebdavFile, backupManager.deleteWebdavFile) + ipcMain.handle(IpcChannel.Backup_BackupToLocalDir, backupManager.backupToLocalDir) + ipcMain.handle(IpcChannel.Backup_RestoreFromLocalBackup, backupManager.restoreFromLocalBackup) + ipcMain.handle(IpcChannel.Backup_ListLocalBackupFiles, backupManager.listLocalBackupFiles) + ipcMain.handle(IpcChannel.Backup_DeleteLocalBackupFile, backupManager.deleteLocalBackupFile) + ipcMain.handle(IpcChannel.Backup_SetLocalBackupDir, backupManager.setLocalBackupDir) + ipcMain.handle(IpcChannel.Backup_BackupToS3, backupManager.backupToS3) + ipcMain.handle(IpcChannel.Backup_RestoreFromS3, backupManager.restoreFromS3) + ipcMain.handle(IpcChannel.Backup_ListS3Files, backupManager.listS3Files) + ipcMain.handle(IpcChannel.Backup_DeleteS3File, backupManager.deleteS3File) + ipcMain.handle(IpcChannel.Backup_CheckS3Connection, backupManager.checkS3Connection) // file ipcMain.handle(IpcChannel.File_Open, fileManager.open) @@ -494,6 +501,10 @@ export function registerIpc(mainWindow: BrowserWindow, app: Electron.App) { ipcMain.handle(IpcChannel.Mcp_GetResource, mcpService.getResource) ipcMain.handle(IpcChannel.Mcp_GetInstallInfo, mcpService.getInstallInfo) ipcMain.handle(IpcChannel.Mcp_CheckConnectivity, mcpService.checkMcpConnectivity) + ipcMain.handle(IpcChannel.Mcp_AbortTool, mcpService.abortTool) + ipcMain.handle(IpcChannel.Mcp_SetProgress, (_, progress: number) => { + mainWindow.webContents.send('mcp-progress', progress) + }) // Register Python execution handler ipcMain.handle( @@ -561,4 +572,8 @@ export function registerIpc(mainWindow: BrowserWindow, app: Electron.App) { SelectionService.registerIpcHandler() ipcMain.handle(IpcChannel.App_QuoteToMain, (_, text: string) => windowService.quoteToMainWindow(text)) + + ipcMain.handle(IpcChannel.App_SetDisableHardwareAcceleration, (_, isDisable: boolean) => { + configManager.setDisableHardwareAcceleration(isDisable) + }) } diff --git a/src/main/knowledage/loader/index.ts b/src/main/knowledage/loader/index.ts index 783e62881a..f86df65dba 100644 --- a/src/main/knowledage/loader/index.ts +++ b/src/main/knowledage/loader/index.ts @@ -1,8 +1,7 @@ -import * as fs from 'node:fs' - import { JsonLoader, LocalPathLoader, RAGApplication, TextLoader } from '@cherrystudio/embedjs' import type { AddLoaderReturn } from '@cherrystudio/embedjs-interfaces' import { WebLoader } from '@cherrystudio/embedjs-loader-web' +import { readTextFileWithAutoEncoding } from '@main/utils/file' import { LoaderReturn } from '@shared/config/types' import { FileMetadata, KnowledgeBaseParams } from '@types' import Logger from 'electron-log' @@ -115,7 +114,7 @@ export async function addFileLoader( // HTML类型处理 loaderReturn = await ragApplication.addLoader( new WebLoader({ - urlOrContent: fs.readFileSync(file.path, 'utf-8'), + urlOrContent: readTextFileWithAutoEncoding(file.path), chunkSize: base.chunkSize, chunkOverlap: base.chunkOverlap }) as any, @@ -125,7 +124,7 @@ export async function addFileLoader( case 'json': try { - jsonObject = JSON.parse(fs.readFileSync(file.path, 'utf-8')) + jsonObject = JSON.parse(readTextFileWithAutoEncoding(file.path)) } catch (error) { jsonParsed = false Logger.warn('[KnowledgeBase] failed parsing json file, falling back to text processing:', file.path, error) @@ -141,7 +140,7 @@ export async function addFileLoader( // 如果是其他文本类型且尚未读取文件，则读取文件 loaderReturn = await ragApplication.addLoader( new TextLoader({ - text: fs.readFileSync(file.path, 'utf-8'), + text: readTextFileWithAutoEncoding(file.path), chunkSize: base.chunkSize, chunkOverlap: base.chunkOverlap }) as any, diff --git a/src/main/ocr/BaseOcrProvider.ts b/src/main/knowledage/ocr/BaseOcrProvider.ts similarity index 100% rename from src/main/ocr/BaseOcrProvider.ts rename to src/main/knowledage/ocr/BaseOcrProvider.ts diff --git a/src/main/ocr/DefaultOcrProvider.ts b/src/main/knowledage/ocr/DefaultOcrProvider.ts similarity index 100% rename from src/main/ocr/DefaultOcrProvider.ts rename to src/main/knowledage/ocr/DefaultOcrProvider.ts diff --git a/src/main/ocr/MacSysOcrProvider.ts b/src/main/knowledage/ocr/MacSysOcrProvider.ts similarity index 100% rename from src/main/ocr/MacSysOcrProvider.ts rename to src/main/knowledage/ocr/MacSysOcrProvider.ts diff --git a/src/main/ocr/OcrProvider.ts b/src/main/knowledage/ocr/OcrProvider.ts similarity index 100% rename from src/main/ocr/OcrProvider.ts rename to src/main/knowledage/ocr/OcrProvider.ts diff --git a/src/main/ocr/OcrProviderFactory.ts b/src/main/knowledage/ocr/OcrProviderFactory.ts similarity index 100% rename from src/main/ocr/OcrProviderFactory.ts rename to src/main/knowledage/ocr/OcrProviderFactory.ts diff --git a/src/main/preprocess/BasePreprocessProvider.ts b/src/main/knowledage/preprocess/BasePreprocessProvider.ts similarity index 100% rename from src/main/preprocess/BasePreprocessProvider.ts rename to src/main/knowledage/preprocess/BasePreprocessProvider.ts diff --git a/src/main/preprocess/DefaultPreprocessProvider.ts b/src/main/knowledage/preprocess/DefaultPreprocessProvider.ts similarity index 100% rename from src/main/preprocess/DefaultPreprocessProvider.ts rename to src/main/knowledage/preprocess/DefaultPreprocessProvider.ts diff --git a/src/main/preprocess/Doc2xPreprocessProvider.ts b/src/main/knowledage/preprocess/Doc2xPreprocessProvider.ts similarity index 100% rename from src/main/preprocess/Doc2xPreprocessProvider.ts rename to src/main/knowledage/preprocess/Doc2xPreprocessProvider.ts diff --git a/src/main/preprocess/MineruPreprocessProvider.ts b/src/main/knowledage/preprocess/MineruPreprocessProvider.ts similarity index 98% rename from src/main/preprocess/MineruPreprocessProvider.ts rename to src/main/knowledage/preprocess/MineruPreprocessProvider.ts index a0a9c65417..58c0c00c23 100644 --- a/src/main/preprocess/MineruPreprocessProvider.ts +++ b/src/main/knowledage/preprocess/MineruPreprocessProvider.ts @@ -111,7 +111,6 @@ export default class MineruPreprocessProvider extends BasePreprocessProvider { } private async validateFile(filePath: string): Promise { - const quota = await this.checkQuota() const pdfBuffer = await fs.promises.readFile(filePath) const doc = await this.readPdf(new Uint8Array(pdfBuffer)) @@ -125,10 +124,6 @@ export default class MineruPreprocessProvider extends BasePreprocessProvider { const fileSizeMB = Math.round(pdfBuffer.length / (1024 * 1024)) throw new Error(`PDF file size (${fileSizeMB}MB) exceeds the limit of 200MB`) } - // 检查配额 - if (quota <= 0 || quota - doc.numPages <= 0) { - throw new Error('MinerU解析配额不足，请申请企业账户或自行部署，剩余额度：' + quota) - } } private createProcessedFileInfo(file: FileMetadata, outputPath: string): FileMetadata { diff --git a/src/main/preprocess/MistralPreprocessProvider.ts b/src/main/knowledage/preprocess/MistralPreprocessProvider.ts similarity index 100% rename from src/main/preprocess/MistralPreprocessProvider.ts rename to src/main/knowledage/preprocess/MistralPreprocessProvider.ts diff --git a/src/main/preprocess/PreprocessProvider.ts b/src/main/knowledage/preprocess/PreprocessProvider.ts similarity index 100% rename from src/main/preprocess/PreprocessProvider.ts rename to src/main/knowledage/preprocess/PreprocessProvider.ts diff --git a/src/main/preprocess/PreprocessProviderFactory.ts b/src/main/knowledage/preprocess/PreprocessProviderFactory.ts similarity index 100% rename from src/main/preprocess/PreprocessProviderFactory.ts rename to src/main/knowledage/preprocess/PreprocessProviderFactory.ts diff --git a/src/main/services/AppService.ts b/src/main/services/AppService.ts new file mode 100644 index 0000000000..f7dc5a7657 --- /dev/null +++ b/src/main/services/AppService.ts @@ -0,0 +1,81 @@ +import { isDev, isLinux, isMac, isWin } from '@main/constant' +import { app } from 'electron' +import log from 'electron-log' +import fs from 'fs' +import os from 'os' +import path from 'path' + +export class AppService { + private static instance: AppService + + private constructor() { + // Private constructor to prevent direct instantiation + } + + public static getInstance(): AppService { + if (!AppService.instance) { + AppService.instance = new AppService() + } + return AppService.instance + } + + public async setAppLaunchOnBoot(isLaunchOnBoot: boolean): Promise { + // Set login item settings for windows and mac + // linux is not supported because it requires more file operations + if (isWin || isMac) { + app.setLoginItemSettings({ openAtLogin: isLaunchOnBoot }) + } else if (isLinux) { + try { + const autostartDir = path.join(os.homedir(), '.config', 'autostart') + const desktopFile = path.join(autostartDir, isDev ? 'cherry-studio-dev.desktop' : 'cherry-studio.desktop') + + if (isLaunchOnBoot) { + // Ensure autostart directory exists + try { + await fs.promises.access(autostartDir) + } catch { + await fs.promises.mkdir(autostartDir, { recursive: true }) + } + + // Get executable path + let executablePath = app.getPath('exe') + if (process.env.APPIMAGE) { + // For AppImage packaged apps, use APPIMAGE environment variable + executablePath = process.env.APPIMAGE + } + + // Create desktop file content + const desktopContent = `[Desktop Entry] + Type=Application + Name=Cherry Studio + Comment=A powerful AI assistant for producer. + Exec=${executablePath} + Icon=cherrystudio + Terminal=false + StartupNotify=false + Categories=Development;Utility; + X-GNOME-Autostart-enabled=true + Hidden=false` + + // Write desktop file + await fs.promises.writeFile(desktopFile, desktopContent) + log.info('Created autostart desktop file for Linux') + } else { + // Remove desktop file + try { + await fs.promises.access(desktopFile) + await fs.promises.unlink(desktopFile) + log.info('Removed autostart desktop file for Linux') + } catch { + // File doesn't exist, no need to remove + } + } + } catch (error) { + log.error('Failed to set launch on boot for Linux:', error) + } + } + } +} + +// Default export as singleton instance +export default AppService.getInstance() diff --git a/src/main/services/BackupManager.ts b/src/main/services/BackupManager.ts index e994e90bed..6087cb6a2a 100644 --- a/src/main/services/BackupManager.ts +++ b/src/main/services/BackupManager.ts @@ -1,5 +1,6 @@ import { IpcChannel } from '@shared/IpcChannel' import { WebDavConfig } from '@types' +import { S3Config } from '@types' import archiver from 'archiver' import { exec } from 'child_process' import { app } from 'electron' @@ -10,6 +11,7 @@ import * as path from 'path' import { CreateDirectoryOptions, FileStat } from 'webdav' import { getDataPath } from '../utils' +import S3Storage from './S3Storage' import WebDav from './WebDav' import { windowService } from './WindowService' @@ -25,6 +27,16 @@ class BackupManager { this.restoreFromWebdav = this.restoreFromWebdav.bind(this) this.listWebdavFiles = this.listWebdavFiles.bind(this) this.deleteWebdavFile = this.deleteWebdavFile.bind(this) + this.listLocalBackupFiles = this.listLocalBackupFiles.bind(this) + this.deleteLocalBackupFile = this.deleteLocalBackupFile.bind(this) + this.backupToLocalDir = this.backupToLocalDir.bind(this) + this.restoreFromLocalBackup = this.restoreFromLocalBackup.bind(this) + this.setLocalBackupDir = this.setLocalBackupDir.bind(this) + this.backupToS3 = this.backupToS3.bind(this) + this.restoreFromS3 = this.restoreFromS3.bind(this) + this.listS3Files = this.listS3Files.bind(this) + this.deleteS3File = this.deleteS3File.bind(this) + this.checkS3Connection = this.checkS3Connection.bind(this) } private async setWritableRecursive(dirPath: string): Promise { @@ -85,7 +97,11 @@ class BackupManager { const onProgress = (processData: { stage: string; progress: number; total: number }) => { mainWindow?.webContents.send(IpcChannel.BackupProgress, processData) - Logger.log('[BackupManager] backup progress', processData) + // 只在关键阶段记录日志：开始、结束和主要阶段转换点 + const logStages = ['preparing', 'writing_data', 'preparing_compression', 'completed'] + if (logStages.includes(processData.stage) || processData.progress === 100) { + Logger.log('[BackupManager] backup progress', processData) + } } try { @@ -147,18 +163,23 @@ class BackupManager { let totalBytes = 0 let processedBytes = 0 - // 首先计算总文件数和总大小 + // 首先计算总文件数和总大小，但不记录详细日志 const calculateTotals = async (dirPath: string) => { - const items = await fs.readdir(dirPath, { withFileTypes: true }) - for (const item of items) { - const fullPath = path.join(dirPath, item.name) - if (item.isDirectory()) { - await calculateTotals(fullPath) - } else { - totalEntries++ - const stats = await fs.stat(fullPath) - totalBytes += stats.size + try { + const items = await fs.readdir(dirPath, { withFileTypes: true }) + for (const item of items) { + const fullPath = path.join(dirPath, item.name) + if (item.isDirectory()) { + await calculateTotals(fullPath) + } else { + totalEntries++ + const stats = await fs.stat(fullPath) + totalBytes += stats.size + } } + } catch (error) { + // 仅在出错时记录日志 + Logger.error('[BackupManager] Error calculating totals:', error) } } @@ -230,7 +251,11 @@ class BackupManager { const onProgress = (processData: { stage: string; progress: number; total: number }) => { mainWindow?.webContents.send(IpcChannel.RestoreProgress, processData) - Logger.log('[BackupManager] restore progress', processData) + // 只在关键阶段记录日志 + const logStages = ['preparing', 'extracting', 'extracted', 'reading_data', 'completed'] + if (logStages.includes(processData.stage) || processData.progress === 100) { + Logger.log('[BackupManager] restore progress', processData) + } } try { @@ -382,21 +407,54 @@ class BackupManager { destination: string, onProgress: (size: number) => void ): Promise { - const items = await fs.readdir(source, { withFileTypes: true }) + // 先统计总文件数 + let totalFiles = 0 + let processedFiles = 0 + let lastProgressReported = 0 - for (const item of items) { - const sourcePath = path.join(source, item.name) - const destPath = path.join(destination, item.name) + // 计算总文件数 + const countFiles = async (dir: string): Promise => { + let count = 0 + const items = await fs.readdir(dir, { withFileTypes: true }) + for (const item of items) { + if (item.isDirectory()) { + count += await countFiles(path.join(dir, item.name)) + } else { + count++ + } + } + return count + } - if (item.isDirectory()) { - await fs.ensureDir(destPath) - await this.copyDirWithProgress(sourcePath, destPath, onProgress) - } else { - const stats = await fs.stat(sourcePath) - await fs.copy(sourcePath, destPath) - onProgress(stats.size) + totalFiles = await countFiles(source) + + // 复制文件并更新进度 + const copyDir = async (src: string, dest: string): Promise => { + const items = await fs.readdir(src, { withFileTypes: true }) + + for (const item of items) { + const sourcePath = path.join(src, item.name) + const destPath = path.join(dest, item.name) + + if (item.isDirectory()) { + await fs.ensureDir(destPath) + await copyDir(sourcePath, destPath) + } else { + const stats = await fs.stat(sourcePath) + await fs.copy(sourcePath, destPath) + processedFiles++ + + // 只在进度变化超过5%时报告进度 + const currentProgress = Math.floor((processedFiles / totalFiles) * 100) + if (currentProgress - lastProgressReported >= 5 || processedFiles === totalFiles) { + lastProgressReported = currentProgress + onProgress(stats.size) + } + } } } + + await copyDir(source, destination) } async checkConnection(_: Electron.IpcMainInvokeEvent, webdavConfig: WebDavConfig) { @@ -423,6 +481,191 @@ class BackupManager { throw new Error(error.message || 'Failed to delete backup file') } } + + async backupToLocalDir( + _: Electron.IpcMainInvokeEvent, + data: string, + fileName: string, + localConfig: { + localBackupDir: string + skipBackupFile: boolean + } + ) { + try { + const backupDir = localConfig.localBackupDir + // Create backup directory if it doesn't exist + await fs.ensureDir(backupDir) + + const backupedFilePath = await this.backup(_, fileName, data, backupDir, localConfig.skipBackupFile) + return backupedFilePath + } catch (error) { + Logger.error('[BackupManager] Local backup failed:', error) + throw error + } + } + + async backupToS3(_: Electron.IpcMainInvokeEvent, data: string, s3Config: S3Config) { + const os = require('os') + const deviceName = os.hostname ? os.hostname() : 'device' + const timestamp = new Date() + .toISOString() + .replace(/[-:T.Z]/g, '') + .slice(0, 14) + const filename = s3Config.fileName || `cherry-studio.backup.${deviceName}.${timestamp}.zip` + + Logger.log(`[BackupManager] Starting S3 backup to ${filename}`) + + const backupedFilePath = await this.backup(_, filename, data, undefined, s3Config.skipBackupFile) + const s3Client = new S3Storage(s3Config) + try { + const fileBuffer = await fs.promises.readFile(backupedFilePath) + const result = await s3Client.putFileContents(filename, fileBuffer) + await fs.remove(backupedFilePath) + + Logger.log(`[BackupManager] S3 backup completed successfully: ${filename}`) + return result + } catch (error) { + Logger.error(`[BackupManager] S3 backup failed:`, error) + await fs.remove(backupedFilePath) + throw error + } + } + + async restoreFromLocalBackup(_: Electron.IpcMainInvokeEvent, fileName: string, localBackupDir: string) { + try { + const backupDir = localBackupDir + const backupPath = path.join(backupDir, fileName) + + if (!fs.existsSync(backupPath)) { + throw new Error(`Backup file not found: ${backupPath}`) + } + + return await this.restore(_, backupPath) + } catch (error) { + Logger.error('[BackupManager] Local restore failed:', error) + throw error + } + } + + async listLocalBackupFiles(_: Electron.IpcMainInvokeEvent, localBackupDir: string) { + try { + const files = await fs.readdir(localBackupDir) + const result: Array<{ fileName: string; modifiedTime: string; size: number }> = [] + + for (const file of files) { + const filePath = path.join(localBackupDir, file) + const stat = await fs.stat(filePath) + + if (stat.isFile() && file.endsWith('.zip')) { + result.push({ + fileName: file, + modifiedTime: stat.mtime.toISOString(), + size: stat.size + }) + } + } + + // Sort by modified time, newest first + return result.sort((a, b) => new Date(b.modifiedTime).getTime() - new Date(a.modifiedTime).getTime()) + } catch (error) { + Logger.error('[BackupManager] List local backup files failed:', error) + throw error + } + } + + async deleteLocalBackupFile(_: Electron.IpcMainInvokeEvent, fileName: string, localBackupDir: string) { + try { + const filePath = path.join(localBackupDir, fileName) + + if (!fs.existsSync(filePath)) { + throw new Error(`Backup file not found: ${filePath}`) + } + + await fs.remove(filePath) + return true + } catch (error) { + Logger.error('[BackupManager] Delete local backup file failed:', error) + throw error + } + } + + async setLocalBackupDir(_: Electron.IpcMainInvokeEvent, dirPath: string) { + try { + // Check if directory exists + await fs.ensureDir(dirPath) + return true + } catch (error) { + Logger.error('[BackupManager] Set local backup directory failed:', error) + throw error + } + } + + async restoreFromS3(_: Electron.IpcMainInvokeEvent, s3Config: S3Config) { + const filename = s3Config.fileName || 'cherry-studio.backup.zip' + + Logger.log(`[BackupManager] Starting restore from S3: ${filename}`) + + const s3Client = new S3Storage(s3Config) + try { + const retrievedFile = await s3Client.getFileContents(filename) + const backupedFilePath = path.join(this.backupDir, filename) + if (!fs.existsSync(this.backupDir)) { + fs.mkdirSync(this.backupDir, { recursive: true }) + } + await new Promise((resolve, reject) => { + const writeStream = fs.createWriteStream(backupedFilePath) + writeStream.write(retrievedFile as Buffer) + writeStream.end() + writeStream.on('finish', () => resolve()) + writeStream.on('error', (error) => reject(error)) + }) + + Logger.log(`[BackupManager] S3 restore file downloaded successfully: ${filename}`) + return await this.restore(_, backupedFilePath) + } catch (error: any) { + Logger.error('[BackupManager] Failed to restore from S3:', error) + throw new Error(error.message || 'Failed to restore backup file') + } + } + + listS3Files = async (_: Electron.IpcMainInvokeEvent, s3Config: S3Config) => { + try { + const s3Client = new S3Storage(s3Config) + + const objects = await s3Client.listFiles() + const files = objects + .filter((obj) => obj.key.endsWith('.zip')) + .map((obj) => { + const segments = obj.key.split('/') + const fileName = segments[segments.length - 1] + return { + fileName, + modifiedTime: obj.lastModified || '', + size: obj.size + } + }) + + return files.sort((a, b) => new Date(b.modifiedTime).getTime() - new Date(a.modifiedTime).getTime()) + } catch (error: any) { + Logger.error('Failed to list S3 files:', error) + throw new Error(error.message || 'Failed to list backup files') + } + } + + async deleteS3File(_: Electron.IpcMainInvokeEvent, fileName: string, s3Config: S3Config) { + try { + const s3Client = new S3Storage(s3Config) + return await s3Client.deleteFile(fileName) + } catch (error: any) { + Logger.error('Failed to delete S3 file:', error) + throw new Error(error.message || 'Failed to delete backup file') + } + } + + async checkS3Connection(_: Electron.IpcMainInvokeEvent, s3Config: S3Config) { + const s3Client = new S3Storage(s3Config) + return await s3Client.checkConnection() + } } export default BackupManager diff --git a/src/main/services/ConfigManager.ts b/src/main/services/ConfigManager.ts index 8e4b5d2bf1..a10e7521eb 100644 --- a/src/main/services/ConfigManager.ts +++ b/src/main/services/ConfigManager.ts @@ -24,7 +24,8 @@ export enum ConfigKeys { SelectionAssistantFollowToolbar = 'selectionAssistantFollowToolbar', SelectionAssistantRemeberWinSize = 'selectionAssistantRemeberWinSize', SelectionAssistantFilterMode = 'selectionAssistantFilterMode', - SelectionAssistantFilterList = 'selectionAssistantFilterList' + SelectionAssistantFilterList = 'selectionAssistantFilterList', + DisableHardwareAcceleration = 'disableHardwareAcceleration' } export class ConfigManager { @@ -218,6 +219,14 @@ export class ConfigManager { this.setAndNotify(ConfigKeys.SelectionAssistantFilterList, value) } + getDisableHardwareAcceleration(): boolean { + return this.get(ConfigKeys.DisableHardwareAcceleration, false) + } + + setDisableHardwareAcceleration(value: boolean) { + this.set(ConfigKeys.DisableHardwareAcceleration, value) + } + setAndNotify(key: string, value: unknown) { this.set(key, value, true) } diff --git a/src/main/services/FileStorage.ts b/src/main/services/FileStorage.ts index 0bdcdf56f5..baa94f535a 100644 --- a/src/main/services/FileStorage.ts +++ b/src/main/services/FileStorage.ts @@ -1,4 +1,4 @@ -import { getFilesDir, getFileType, getTempDir } from '@main/utils/file' +import { getFilesDir, getFileType, getTempDir, readTextFileWithAutoEncoding } from '@main/utils/file' import { documentExts, imageExts, MB } from '@shared/config/constant' import { FileMetadata } from '@types' import * as crypto from 'crypto' @@ -188,6 +188,8 @@ class FileStorage { count: 1 } + logger.info('[FileStorage] File uploaded:', fileMetadata) + return fileMetadata } @@ -256,7 +258,13 @@ class FileStorage { } } - return fs.readFileSync(filePath, 'utf8') + try { + const result = readTextFileWithAutoEncoding(filePath) + return result + } catch (error) { + logger.error(error) + return 'failed to read file' + } } public createTempFile = async (_: Electron.IpcMainInvokeEvent, fileName: string): Promise => { diff --git a/src/main/services/KnowledgeService.ts b/src/main/services/KnowledgeService.ts index c57c0eb104..2e5f3a44d0 100644 --- a/src/main/services/KnowledgeService.ts +++ b/src/main/services/KnowledgeService.ts @@ -24,9 +24,9 @@ import { WebLoader } from '@cherrystudio/embedjs-loader-web' import Embeddings from '@main/knowledage/embeddings/Embeddings' import { addFileLoader } from '@main/knowledage/loader' import { NoteLoader } from '@main/knowledage/loader/noteLoader' +import OcrProvider from '@main/knowledage/ocr/OcrProvider' +import PreprocessProvider from '@main/knowledage/preprocess/PreprocessProvider' import Reranker from '@main/knowledage/reranker/Reranker' -import OcrProvider from '@main/ocr/OcrProvider' -import PreprocessProvider from '@main/preprocess/PreprocessProvider' import { windowService } from '@main/services/WindowService' import { getDataPath } from '@main/utils' import { getAllFiles } from '@main/utils/file' diff --git a/src/main/services/MCPService.ts b/src/main/services/MCPService.ts index 2515c91416..9893c81474 100644 --- a/src/main/services/MCPService.ts +++ b/src/main/services/MCPService.ts @@ -28,6 +28,7 @@ import { app } from 'electron' import Logger from 'electron-log' import { EventEmitter } from 'events' import { memoize } from 'lodash' +import { v4 as uuidv4 } from 'uuid' import { CacheService } from './CacheService' import { CallBackServer } from './mcp/oauth/callback' @@ -71,6 +72,7 @@ function withCache( class McpService { private clients: Map = new Map() private pendingClients: Map> = new Map() + private activeToolCalls: Map = new Map() constructor() { this.initClient = this.initClient.bind(this) @@ -84,6 +86,7 @@ class McpService { this.removeServer = this.removeServer.bind(this) this.restartServer = this.restartServer.bind(this) this.stopServer = this.stopServer.bind(this) + this.abortTool = this.abortTool.bind(this) this.cleanup = this.cleanup.bind(this) } @@ -455,10 +458,14 @@ class McpService { */ public async callTool( _: Electron.IpcMainInvokeEvent, - { server, name, args }: { server: MCPServer; name: string; args: any } + { server, name, args, callId }: { server: MCPServer; name: string; args: any; callId?: string } ): Promise { + const toolCallId = callId || uuidv4() + const abortController = new AbortController() + this.activeToolCalls.set(toolCallId, abortController) + try { - Logger.info('[MCP] Calling:', server.name, name, args) + Logger.info('[MCP] Calling:', server.name, name, args, 'callId:', toolCallId) if (typeof args === 'string') { try { args = JSON.parse(args) @@ -468,12 +475,19 @@ class McpService { } const client = await this.initClient(server) const result = await client.callTool({ name, arguments: args }, undefined, { - timeout: server.timeout ? server.timeout * 1000 : 60000 // Default timeout of 1 minute + onprogress: (process) => { + console.log('[MCP] Progress:', process.progress / (process.total || 1)) + window.api.mcp.setProgress(process.progress / (process.total || 1)) + }, + timeout: server.timeout ? server.timeout * 1000 : 60000, // Default timeout of 1 minute + signal: this.activeToolCalls.get(toolCallId)?.signal }) return result as MCPCallToolResponse } catch (error) { Logger.error(`[MCP] Error calling tool ${name} on ${server.name}:`, error) throw error + } finally { + this.activeToolCalls.delete(toolCallId) } } @@ -664,6 +678,20 @@ class McpService { delete env.http_proxy delete env.https_proxy } + + // 实现 abortTool 方法 + public async abortTool(_: Electron.IpcMainInvokeEvent, callId: string) { + const activeToolCall = this.activeToolCalls.get(callId) + if (activeToolCall) { + activeToolCall.abort() + this.activeToolCalls.delete(callId) + Logger.info(`[MCP] Aborted tool call: ${callId}`) + return true + } else { + Logger.warn(`[MCP] No active tool call found for callId: ${callId}`) + return false + } + } } export default new McpService() diff --git a/src/main/services/RemoteStorage.ts b/src/main/services/RemoteStorage.ts deleted file mode 100644 index b62489bbbe..0000000000 --- a/src/main/services/RemoteStorage.ts +++ /dev/null @@ -1,57 +0,0 @@ -// import Logger from 'electron-log' -// import { Operator } from 'opendal' - -// export default class RemoteStorage { -// public instance: Operator | undefined - -// /** -// * -// * @param scheme is the scheme for opendal services. Available value includes "azblob", "azdls", "cos", "gcs", "obs", "oss", "s3", "webdav", "webhdfs", "aliyun-drive", "alluxio", "azfile", "dropbox", "gdrive", "onedrive", "postgresql", "mysql", "redis", "swift", "mongodb", "alluxio", "b2", "seafile", "upyun", "koofr", "yandex-disk" -// * @param options is the options for given opendal services. Valid options depend on the scheme. Checkout https://docs.rs/opendal/latest/opendal/services/index.html for all valid options. -// * -// * For example, use minio as remote storage: -// * -// * ```typescript -// * const storage = new RemoteStorage('s3', { -// * endpoint: 'http://localhost:9000', -// * region: 'us-east-1', -// * bucket: 'testbucket', -// * access_key_id: 'user', -// * secret_access_key: 'password', -// * root: '/path/to/basepath', -// * }) -// * ``` -// */ -// constructor(scheme: string, options?: Record | undefined | null) { -// this.instance = new Operator(scheme, options) - -// this.putFileContents = this.putFileContents.bind(this) -// this.getFileContents = this.getFileContents.bind(this) -// } - -// public putFileContents = async (filename: string, data: string | Buffer) => { -// if (!this.instance) { -// return new Error('RemoteStorage client not initialized') -// } - -// try { -// return await this.instance.write(filename, data) -// } catch (error) { -// Logger.error('[RemoteStorage] Error putting file contents:', error) -// throw error -// } -// } - -// public getFileContents = async (filename: string) => { -// if (!this.instance) { -// throw new Error('RemoteStorage client not initialized') -// } - -// try { -// return await this.instance.read(filename) -// } catch (error) { -// Logger.error('[RemoteStorage] Error getting file contents:', error) -// throw error -// } -// } -// } diff --git a/src/main/services/S3Storage.ts b/src/main/services/S3Storage.ts new file mode 100644 index 0000000000..0b45bb0387 --- /dev/null +++ b/src/main/services/S3Storage.ts @@ -0,0 +1,183 @@ +import { + DeleteObjectCommand, + GetObjectCommand, + HeadBucketCommand, + ListObjectsV2Command, + PutObjectCommand, + S3Client +} from '@aws-sdk/client-s3' +import type { S3Config } from '@types' +import Logger from 'electron-log' +import * as net from 'net' +import { Readable } from 'stream' + +/** + * 将可读流转换为 Buffer + */ +function streamToBuffer(stream: Readable): Promise { + return new Promise((resolve, reject) => { + const chunks: Buffer[] = [] + stream.on('data', (chunk) => chunks.push(Buffer.isBuffer(chunk) ? chunk : Buffer.from(chunk))) + stream.on('error', reject) + stream.on('end', () => resolve(Buffer.concat(chunks))) + }) +} + +// 需要使用 Virtual Host-Style 的服务商域名后缀白名单 +const VIRTUAL_HOST_SUFFIXES = ['aliyuncs.com', 'myqcloud.com'] + +/** + * 使用 AWS SDK v3 的简单 S3 封装，兼容之前 RemoteStorage 的最常用接口。 + */ +export default class S3Storage { + private client: S3Client + private bucket: string + private root: string + + constructor(config: S3Config) { + const { endpoint, region, accessKeyId, secretAccessKey, bucket, root } = config + + const usePathStyle = (() => { + if (!endpoint) return false + + try { + const { hostname } = new URL(endpoint) + + if (hostname === 'localhost' || net.isIP(hostname) !== 0) { + return true + } + + const isInWhiteList = VIRTUAL_HOST_SUFFIXES.some((suffix) => hostname.endsWith(suffix)) + return !isInWhiteList + } catch (e) { + Logger.warn('[S3Storage] Failed to parse endpoint, fallback to Path-Style:', endpoint, e) + return true + } + })() + + this.client = new S3Client({ + region, + endpoint: endpoint || undefined, + credentials: { + accessKeyId: accessKeyId, + secretAccessKey: secretAccessKey + }, + forcePathStyle: usePathStyle + }) + + this.bucket = bucket + this.root = root?.replace(/^\/+/g, '').replace(/\/+$/g, '') || '' + + this.putFileContents = this.putFileContents.bind(this) + this.getFileContents = this.getFileContents.bind(this) + this.deleteFile = this.deleteFile.bind(this) + this.listFiles = this.listFiles.bind(this) + this.checkConnection = this.checkConnection.bind(this) + } + + /** + * 内部辅助方法，用来拼接带 root 的对象 key + */ + private buildKey(key: string): string { + if (!this.root) return key + return key.startsWith(`${this.root}/`) ? key : `${this.root}/${key}` + } + + async putFileContents(key: string, data: Buffer | string) { + try { + const contentType = key.endsWith('.zip') ? 'application/zip' : 'application/octet-stream' + + return await this.client.send( + new PutObjectCommand({ + Bucket: this.bucket, + Key: this.buildKey(key), + Body: data, + ContentType: contentType + }) + ) + } catch (error) { + Logger.error('[S3Storage] Error putting object:', error) + throw error + } + } + + async getFileContents(key: string): Promise { + try { + const res = await this.client.send(new GetObjectCommand({ Bucket: this.bucket, Key: this.buildKey(key) })) + if (!res.Body || !(res.Body instanceof Readable)) { + throw new Error('Empty body received from S3') + } + return await streamToBuffer(res.Body as Readable) + } catch (error) { + Logger.error('[S3Storage] Error getting object:', error) + throw error + } + } + + async deleteFile(key: string) { + try { + const keyWithRoot = this.buildKey(key) + const variations = new Set([keyWithRoot, key.replace(/^\//, '')]) + for (const k of variations) { + try { + await this.client.send(new DeleteObjectCommand({ Bucket: this.bucket, Key: k })) + } catch { + // 忽略删除失败 + } + } + } catch (error) { + Logger.error('[S3Storage] Error deleting object:', error) + throw error + } + } + + /** + * 列举指定前缀下的对象，默认列举全部。 + */ + async listFiles(prefix = ''): Promise> { + const files: Array<{ key: string; lastModified?: string; size: number }> = [] + let continuationToken: string | undefined + const fullPrefix = this.buildKey(prefix) + + try { + do { + const res = await this.client.send( + new ListObjectsV2Command({ + Bucket: this.bucket, + Prefix: fullPrefix === '' ? undefined : fullPrefix, + ContinuationToken: continuationToken + }) + ) + + res.Contents?.forEach((obj) => { + if (!obj.Key) return + files.push({ + key: obj.Key, + lastModified: obj.LastModified?.toISOString(), + size: obj.Size ?? 0 + }) + }) + + continuationToken = res.IsTruncated ? res.NextContinuationToken : undefined + } while (continuationToken) + + return files + } catch (error) { + Logger.error('[S3Storage] Error listing objects:', error) + throw error + } + } + + /** + * 尝试调用 HeadBucket 判断凭证/网络是否可用 + */ + async checkConnection() { + try { + await this.client.send(new HeadBucketCommand({ Bucket: this.bucket })) + return true + } catch (error) { + Logger.error('[S3Storage] Error checking connection:', error) + throw error + } + } +} diff --git a/src/main/services/SelectionService.ts b/src/main/services/SelectionService.ts index 23578b75e0..3be2d5a95a 100644 --- a/src/main/services/SelectionService.ts +++ b/src/main/services/SelectionService.ts @@ -141,7 +141,7 @@ export class SelectionService { * Initialize zoom factor from config and subscribe to changes * Ensures UI elements scale properly with system DPI settings */ - private initZoomFactor() { + private initZoomFactor(): void { const zoomFactor = configManager.getZoomFactor() if (zoomFactor) { this.setZoomFactor(zoomFactor) @@ -154,7 +154,7 @@ export class SelectionService { this.zoomFactor = zoomFactor } - private initConfig() { + private initConfig(): void { this.triggerMode = configManager.getSelectionAssistantTriggerMode() as TriggerMode this.isFollowToolbar = configManager.getSelectionAssistantFollowToolbar() this.isRemeberWinSize = configManager.getSelectionAssistantRemeberWinSize() @@ -207,7 +207,7 @@ export class SelectionService { * @param mode - The mode to set, either 'default', 'whitelist', or 'blacklist' * @param list - An array of strings representing the list of items to include or exclude */ - private setHookGlobalFilterMode(mode: string, list: string[]) { + private setHookGlobalFilterMode(mode: string, list: string[]): void { if (!this.selectionHook) return const modeMap = { @@ -245,7 +245,7 @@ export class SelectionService { } } - private setHookFineTunedList() { + private setHookFineTunedList(): void { if (!this.selectionHook) return const excludeClipboardCursorDetectList = isWin @@ -271,6 +271,11 @@ export class SelectionService { * @returns {boolean} Success status of service start */ public start(): boolean { + if (!isSupportedOS) { + this.logError(new Error('SelectionService start(): not supported on this OS')) + return false + } + if (!this.selectionHook) { this.logError(new Error('SelectionService start(): instance is null')) return false @@ -373,7 +378,7 @@ export class SelectionService { * Toggle the enabled state of the selection service * Will sync the new enabled store to all renderer windows */ - public toggleEnabled(enabled: boolean | undefined = undefined) { + public toggleEnabled(enabled: boolean | undefined = undefined): void { if (!this.selectionHook) return const newEnabled = enabled === undefined ? !configManager.getSelectionAssistantEnabled() : enabled @@ -389,7 +394,7 @@ export class SelectionService { * Sets up window properties, event handlers, and loads the toolbar UI * @param readyCallback Optional callback when window is ready to show */ - private createToolbarWindow(readyCallback?: () => void) { + private createToolbarWindow(readyCallback?: () => void): void { if (this.isToolbarAlive()) return const { toolbarWidth, toolbarHeight } = this.getToolbarRealSize() @@ -414,9 +419,11 @@ export class SelectionService { backgroundMaterial: 'none', // Platform specific settings - // [macOS] DO NOT set type to 'panel', it will not work because it conflicts with other settings // [macOS] DO NOT set focusable to false, it will make other windows bring to front together - ...(isWin ? { type: 'toolbar', focusable: false } : {}), + // [macOS] `panel` conflicts with other settings , + // and log will show `NSWindow does not support nonactivating panel styleMask 0x80` + // but it seems still work on fullscreen apps, so we set this anyway + ...(isWin ? { type: 'toolbar', focusable: false } : { type: 'panel' }), hiddenInMissionControl: true, // [macOS only] acceptFirstMouse: true, // [macOS only] @@ -447,13 +454,6 @@ export class SelectionService { // Add show/hide event listeners this.toolbarWindow.on('show', () => { this.toolbarWindow?.webContents.send(IpcChannel.Selection_ToolbarVisibilityChange, true) - - // [macOS] force the toolbar window to be visible on current desktop - // but it will make docker icon flash. And we found that it's not necessary now. - // will remove after testing - // if (isMac) { - // this.toolbarWindow!.setVisibleOnAllWorkspaces(false) - // } }) this.toolbarWindow.on('hide', () => { @@ -485,10 +485,10 @@ export class SelectionService { * @param point Reference point for positioning, logical coordinates * @param orientation Preferred position relative to reference point */ - private showToolbarAtPosition(point: Point, orientation: RelativeOrientation) { + private showToolbarAtPosition(point: Point, orientation: RelativeOrientation, programName: string): void { if (!this.isToolbarAlive()) { this.createToolbarWindow(() => { - this.showToolbarAtPosition(point, orientation) + this.showToolbarAtPosition(point, orientation, programName) }) return } @@ -509,16 +509,45 @@ export class SelectionService { //should set every time the window is shown this.toolbarWindow!.setAlwaysOnTop(true, 'screen-saver') - // [macOS] force the toolbar window to be visible on current desktop - // but it will make docker icon flash. And we found that it's not necessary now. - // will remove after testing - // if (isMac) { - // this.toolbarWindow!.setVisibleOnAllWorkspaces(true, { visibleOnFullScreen: true }) - // } + // [macOS] a series of hacky ways only for macOS + if (isMac) { + // [macOS] a hacky way + // when set `skipTransformProcessType: true`, if the selection is in self app, it will make the selection canceled after toolbar showing + // so we just don't set `skipTransformProcessType: true` when in self app + const isSelf = ['com.github.Electron', 'com.kangfenmao.CherryStudio'].includes(programName) - // [macOS] MUST use `showInactive()` to prevent other windows bring to front together - // [Windows] is OK for both `show()` and `showInactive()` because of `focusable: false` - this.toolbarWindow!.showInactive() + if (!isSelf) { + // [macOS] an ugly hacky way + // `focusable: true` will make mainWindow disappeared when `setVisibleOnAllWorkspaces` + // so we set `focusable: true` before showing, and then set false after showing + this.toolbarWindow!.setFocusable(false) + + // [macOS] + // force `setVisibleOnAllWorkspaces: true` to let toolbar show in all workspaces. And we MUST not set it to false again + // set `skipTransformProcessType: true` to avoid dock icon spinning when `setVisibleOnAllWorkspaces` + this.toolbarWindow!.setVisibleOnAllWorkspaces(true, { + visibleOnFullScreen: true, + skipTransformProcessType: true + }) + } + + // [macOS] MUST use `showInactive()` to prevent other windows bring to front together + // [Windows] is OK for both `show()` and `showInactive()` because of `focusable: false` + this.toolbarWindow!.showInactive() + + // [macOS] restore the focusable status + this.toolbarWindow!.setFocusable(true) + + this.startHideByMouseKeyListener() + + return + } + + /** + * The following is for Windows + */ + + this.toolbarWindow!.show() /** * [Windows] @@ -588,8 +617,8 @@ export class SelectionService { * Check if toolbar window exists and is not destroyed * @returns {boolean} Toolbar window status */ - private isToolbarAlive() { - return this.toolbarWindow && !this.toolbarWindow.isDestroyed() + private isToolbarAlive(): boolean { + return !!(this.toolbarWindow && !this.toolbarWindow.isDestroyed()) } /** @@ -598,7 +627,7 @@ export class SelectionService { * @param width New toolbar width * @param height New toolbar height */ - public determineToolbarSize(width: number, height: number) { + public determineToolbarSize(width: number, height: number): void { const toolbarWidth = Math.ceil(width) // only update toolbar width if it's changed @@ -611,7 +640,7 @@ export class SelectionService { * Get actual toolbar dimensions accounting for zoom factor * @returns Object containing toolbar width and height */ - private getToolbarRealSize() { + private getToolbarRealSize(): { toolbarWidth: number; toolbarHeight: number } { return { toolbarWidth: this.TOOLBAR_WIDTH * this.zoomFactor, toolbarHeight: this.TOOLBAR_HEIGHT * this.zoomFactor @@ -882,8 +911,8 @@ export class SelectionService { refPoint = { x: Math.round(refPoint.x), y: Math.round(refPoint.y) } } - this.showToolbarAtPosition(refPoint, refOrientation) - this.toolbarWindow?.webContents.send(IpcChannel.Selection_TextSelected, selectionData) + this.showToolbarAtPosition(refPoint, refOrientation, selectionData.programName) + this.toolbarWindow!.webContents.send(IpcChannel.Selection_TextSelected, selectionData) } /** @@ -891,7 +920,7 @@ export class SelectionService { */ // Start monitoring global mouse clicks - private startHideByMouseKeyListener() { + private startHideByMouseKeyListener(): void { try { // Register event handlers this.selectionHook!.on('mouse-down', this.handleMouseDownHide) @@ -904,7 +933,7 @@ export class SelectionService { } // Stop monitoring global mouse clicks - private stopHideByMouseKeyListener() { + private stopHideByMouseKeyListener(): void { if (!this.isHideByMouseKeyListenerActive) return try { @@ -1098,7 +1127,7 @@ export class SelectionService { * Initialize preloaded action windows * Creates a pool of windows at startup for faster response */ - private async initPreloadedActionWindows() { + private async initPreloadedActionWindows(): Promise { try { // Create initial pool of preloaded windows for (let i = 0; i < this.PRELOAD_ACTION_WINDOW_COUNT; i++) { @@ -1112,7 +1141,7 @@ export class SelectionService { /** * Close all preloaded action windows */ - private closePreloadedActionWindows() { + private closePreloadedActionWindows(): void { for (const actionWindow of this.preloadedActionWindows) { if (!actionWindow.isDestroyed()) { actionWindow.destroy() @@ -1124,7 +1153,7 @@ export class SelectionService { * Preload a new action window asynchronously * This method is called after popping a window to ensure we always have windows ready */ - private async pushNewActionWindow() { + private async pushNewActionWindow(): Promise { try { const actionWindow = this.createPreloadedActionWindow() this.preloadedActionWindows.push(actionWindow) @@ -1138,7 +1167,7 @@ export class SelectionService { * Immediately returns a window and asynchronously creates a new one * @returns {BrowserWindow} The action window */ - private popActionWindow() { + private popActionWindow(): BrowserWindow { // Get a window from the preloaded queue or create a new one if empty const actionWindow = this.preloadedActionWindows.pop() || this.createPreloadedActionWindow() @@ -1202,7 +1231,7 @@ export class SelectionService { * Ensures window stays within screen boundaries * @param actionWindow Window to position and show */ - private showActionWindow(actionWindow: BrowserWindow) { + private showActionWindow(actionWindow: BrowserWindow): void { let actionWindowWidth = this.ACTION_WINDOW_WIDTH let actionWindowHeight = this.ACTION_WINDOW_HEIGHT @@ -1228,6 +1257,7 @@ export class SelectionService { }) actionWindow.show() + return } @@ -1292,38 +1322,40 @@ export class SelectionService { * Switches between selection-based and alt-key based triggering * Manages appropriate event listeners for each mode */ - private processTriggerMode() { + private processTriggerMode(): void { + if (!this.selectionHook) return + switch (this.triggerMode) { case TriggerMode.Selected: if (this.isCtrlkeyListenerActive) { - this.selectionHook!.off('key-down', this.handleKeyDownCtrlkeyMode) - this.selectionHook!.off('key-up', this.handleKeyUpCtrlkeyMode) + this.selectionHook.off('key-down', this.handleKeyDownCtrlkeyMode) + this.selectionHook.off('key-up', this.handleKeyUpCtrlkeyMode) this.isCtrlkeyListenerActive = false } - this.selectionHook!.setSelectionPassiveMode(false) + this.selectionHook.setSelectionPassiveMode(false) break case TriggerMode.Ctrlkey: if (!this.isCtrlkeyListenerActive) { - this.selectionHook!.on('key-down', this.handleKeyDownCtrlkeyMode) - this.selectionHook!.on('key-up', this.handleKeyUpCtrlkeyMode) + this.selectionHook.on('key-down', this.handleKeyDownCtrlkeyMode) + this.selectionHook.on('key-up', this.handleKeyUpCtrlkeyMode) this.isCtrlkeyListenerActive = true } - this.selectionHook!.setSelectionPassiveMode(true) + this.selectionHook.setSelectionPassiveMode(true) break case TriggerMode.Shortcut: //remove the ctrlkey listener, don't need any key listener for shortcut mode if (this.isCtrlkeyListenerActive) { - this.selectionHook!.off('key-down', this.handleKeyDownCtrlkeyMode) - this.selectionHook!.off('key-up', this.handleKeyUpCtrlkeyMode) + this.selectionHook.off('key-down', this.handleKeyDownCtrlkeyMode) + this.selectionHook.off('key-up', this.handleKeyUpCtrlkeyMode) this.isCtrlkeyListenerActive = false } - this.selectionHook!.setSelectionPassiveMode(true) + this.selectionHook.setSelectionPassiveMode(true) break } } @@ -1404,13 +1436,13 @@ export class SelectionService { this.isIpcHandlerRegistered = true } - private logInfo(message: string, forceShow: boolean = false) { + private logInfo(message: string, forceShow: boolean = false): void { if (isDev || forceShow) { Logger.info('[SelectionService] Info: ', message) } } - private logError(...args: [...string[], Error]) { + private logError(...args: [...string[], Error]): void { Logger.error('[SelectionService] Error: ', ...args) } } @@ -1423,7 +1455,7 @@ export class SelectionService { export function initSelectionService(): boolean { if (!isSupportedOS) return false - configManager.subscribe(ConfigKeys.SelectionAssistantEnabled, (enabled: boolean) => { + configManager.subscribe(ConfigKeys.SelectionAssistantEnabled, (enabled: boolean): void => { //avoid closure const ss = SelectionService.getInstance() if (!ss) { diff --git a/src/main/services/ThemeService.ts b/src/main/services/ThemeService.ts index 7ccaf3bf9a..a56b559357 100644 --- a/src/main/services/ThemeService.ts +++ b/src/main/services/ThemeService.ts @@ -1,48 +1,48 @@ -import { IpcChannel } from '@shared/IpcChannel' -import { ThemeMode } from '@types' -import { BrowserWindow, nativeTheme } from 'electron' - -import { titleBarOverlayDark, titleBarOverlayLight } from '../config' -import { configManager } from './ConfigManager' - -class ThemeService { - private theme: ThemeMode = ThemeMode.system - constructor() { - this.theme = configManager.getTheme() - - if (this.theme === ThemeMode.dark || this.theme === ThemeMode.light || this.theme === ThemeMode.system) { - nativeTheme.themeSource = this.theme - } else { - // 兼容旧版本 - configManager.setTheme(ThemeMode.system) - nativeTheme.themeSource = ThemeMode.system - } - nativeTheme.on('updated', this.themeUpdatadHandler.bind(this)) - } - - themeUpdatadHandler() { - BrowserWindow.getAllWindows().forEach((win) => { - if (win && !win.isDestroyed() && win.setTitleBarOverlay) { - try { - win.setTitleBarOverlay(nativeTheme.shouldUseDarkColors ? titleBarOverlayDark : titleBarOverlayLight) - } catch (error) { - // don't throw error if setTitleBarOverlay failed - // Because it may be called with some windows have some title bar - } - } - win.webContents.send(IpcChannel.ThemeUpdated, nativeTheme.shouldUseDarkColors ? ThemeMode.dark : ThemeMode.light) - }) - } - - setTheme(theme: ThemeMode) { - if (theme === this.theme) { - return - } - - this.theme = theme - nativeTheme.themeSource = theme - configManager.setTheme(theme) - } -} - -export const themeService = new ThemeService() +import { IpcChannel } from '@shared/IpcChannel' +import { ThemeMode } from '@types' +import { BrowserWindow, nativeTheme } from 'electron' + +import { titleBarOverlayDark, titleBarOverlayLight } from '../config' +import { configManager } from './ConfigManager' + +class ThemeService { + private theme: ThemeMode = ThemeMode.system + constructor() { + this.theme = configManager.getTheme() + + if (this.theme === ThemeMode.dark || this.theme === ThemeMode.light || this.theme === ThemeMode.system) { + nativeTheme.themeSource = this.theme + } else { + // 兼容旧版本 + configManager.setTheme(ThemeMode.system) + nativeTheme.themeSource = ThemeMode.system + } + nativeTheme.on('updated', this.themeUpdatadHandler.bind(this)) + } + + themeUpdatadHandler() { + BrowserWindow.getAllWindows().forEach((win) => { + if (win && !win.isDestroyed() && win.setTitleBarOverlay) { + try { + win.setTitleBarOverlay(nativeTheme.shouldUseDarkColors ? titleBarOverlayDark : titleBarOverlayLight) + } catch (error) { + // don't throw error if setTitleBarOverlay failed + // Because it may be called with some windows have some title bar + } + } + win.webContents.send(IpcChannel.ThemeUpdated, nativeTheme.shouldUseDarkColors ? ThemeMode.dark : ThemeMode.light) + }) + } + + setTheme(theme: ThemeMode) { + if (theme === this.theme) { + return + } + + this.theme = theme + nativeTheme.themeSource = theme + configManager.setTheme(theme) + } +} + +export const themeService = new ThemeService() diff --git a/src/main/services/WindowService.ts b/src/main/services/WindowService.ts index 65132eb54f..7c4f13a254 100644 --- a/src/main/services/WindowService.ts +++ b/src/main/services/WindowService.ts @@ -41,8 +41,8 @@ export class WindowService { } const mainWindowState = windowStateKeeper({ - defaultWidth: 1080, - defaultHeight: 670, + defaultWidth: 960, + defaultHeight: 600, fullScreen: false, maximize: false }) @@ -52,7 +52,7 @@ export class WindowService { y: mainWindowState.y, width: mainWindowState.width, height: mainWindowState.height, - minWidth: 1080, + minWidth: 960, minHeight: 600, show: false, autoHideMenuBar: true, diff --git a/src/main/services/urlschema/handle-providers.ts b/src/main/services/urlschema/handle-providers.ts index d23f3749db..9a598fc459 100644 --- a/src/main/services/urlschema/handle-providers.ts +++ b/src/main/services/urlschema/handle-providers.ts @@ -1,3 +1,4 @@ +import { isMac } from '@main/constant' import Logger from 'electron-log' import { windowService } from '../WindowService' @@ -33,8 +34,13 @@ export async function handleProvidersProtocolUrl(url: URL) { (await mainWindow.webContents.executeJavaScript(`typeof window.navigate === 'function'`)) ) { mainWindow.webContents.executeJavaScript(`window.navigate('/settings/provider?addProviderData=${data}')`) + + if (isMac) { + windowService.showMainWindow() + } } else { setTimeout(() => { + Logger.info('handleProvidersProtocolUrl timeout', { data, version }) handleProvidersProtocolUrl(url) }, 1000) } diff --git a/src/main/utils/__tests__/file.test.ts b/src/main/utils/__tests__/file.test.ts index 14f4801524..6066729dc7 100644 --- a/src/main/utils/__tests__/file.test.ts +++ b/src/main/utils/__tests__/file.test.ts @@ -3,8 +3,10 @@ import os from 'node:os' import path from 'node:path' import { FileTypes } from '@types' +import iconv from 'iconv-lite' import { afterEach, beforeEach, describe, expect, it, vi } from 'vitest' +import { detectEncoding, readTextFileWithAutoEncoding } from '../file' import { getAllFiles, getAppConfigDir, getConfigDir, getFilesDir, getFileType, getTempDir } from '../file' // Mock dependencies @@ -241,4 +243,104 @@ describe('file', () => { expect(appConfigDir).toBe('/mock/home/.cherrystudio/config/') }) }) + + // 在 describe('file') 块内部添加新的 describe 块 + describe('detectEncoding', () => { + const mockFilePath = '/path/to/mock/file.txt' + + beforeEach(() => { + vi.mocked(fs.openSync).mockReturnValue(123) + vi.mocked(fs.closeSync).mockImplementation(() => {}) + }) + + it('should correctly detect UTF-8 encoding', () => { + // 准备UTF-8编码的Buffer + const content = '这是UTF-8测试内容' + const buffer = Buffer.from(content, 'utf-8') + + // 模拟文件读取 + vi.mocked(fs.readSync).mockImplementation((_, buf) => { + const targetBuffer = new Uint8Array(buf.buffer) + const sourceBuffer = new Uint8Array(buffer) + targetBuffer.set(sourceBuffer) + return 1024 + }) + + const encoding = detectEncoding(mockFilePath) + expect(encoding).toBe('UTF-8') + }) + + it('should correctly detect GB2312 encoding', () => { + // 使用iconv创建GB2312编码内容 + const content = '这是一段GB2312编码的测试内容' + const gb2312Buffer = iconv.encode(content, 'GB2312') + + // 模拟文件读取 + vi.mocked(fs.readSync).mockImplementation((_, buf) => { + const targetBuffer = new Uint8Array(buf.buffer) + const sourceBuffer = new Uint8Array(gb2312Buffer) + targetBuffer.set(sourceBuffer) + return gb2312Buffer.length + }) + + const encoding = detectEncoding(mockFilePath) + expect(encoding).toMatch(/GB2312|GB18030/i) + }) + + it('should correctly detect ASCII encoding', () => { + // 准备ASCII编码内容 + const content = 'ASCII content' + const buffer = Buffer.from(content, 'ascii') + + // 模拟文件读取 + vi.mocked(fs.readSync).mockImplementation((_, buf) => { + const targetBuffer = new Uint8Array(buf.buffer) + const sourceBuffer = new Uint8Array(buffer) + targetBuffer.set(sourceBuffer) + return buffer.length + }) + + const encoding = detectEncoding(mockFilePath) + expect(encoding.toLowerCase()).toBe('ascii') + }) + }) + + describe('readTextFileWithAutoEncoding', () => { + const mockFilePath = '/path/to/mock/file.txt' + + beforeEach(() => { + vi.mocked(fs.openSync).mockReturnValue(123) + vi.mocked(fs.closeSync).mockImplementation(() => {}) + }) + + it('should read file with auto encoding', () => { + const content = '这是一段GB2312编码的测试内容' + const buffer = iconv.encode(content, 'GB2312') + vi.mocked(fs.readSync).mockImplementation((_, buf) => { + const targetBuffer = new Uint8Array(buf.buffer) + const sourceBuffer = new Uint8Array(buffer) + targetBuffer.set(sourceBuffer) + return buffer.length + }) + vi.mocked(fs.readFileSync).mockReturnValue(buffer) + + const result = readTextFileWithAutoEncoding(mockFilePath) + expect(result).toBe(content) + }) + + it('should try to fix bad detected encoding', () => { + const content = '这是一段GB2312编码的测试内容' + const buffer = iconv.encode(content, 'GB2312') + vi.mocked(fs.readSync).mockImplementation((_, buf) => { + const targetBuffer = new Uint8Array(buf.buffer) + const sourceBuffer = new Uint8Array(buffer) + targetBuffer.set(sourceBuffer) + return buffer.length + }) + vi.mocked(fs.readFileSync).mockReturnValue(buffer) + vi.mocked(vi.fn(detectEncoding)).mockReturnValue('UTF-8') + const result = readTextFileWithAutoEncoding(mockFilePath) + expect(result).toBe(content) + }) + }) }) diff --git a/src/main/utils/file.ts b/src/main/utils/file.ts index 2c52e82a71..baba7ec8ba 100644 --- a/src/main/utils/file.ts +++ b/src/main/utils/file.ts @@ -6,6 +6,9 @@ import { isLinux, isPortable } from '@main/constant' import { audioExts, documentExts, imageExts, textExts, videoExts } from '@shared/config/constant' import { FileMetadata, FileTypes } from '@types' import { app } from 'electron' +import Logger from 'electron-log' +import iconv from 'iconv-lite' +import { detect as detectEncoding_, detectAll as detectEncodingAll } from 'jschardet' import { v4 as uuidv4 } from 'uuid' export function initAppDataDir() { @@ -202,3 +205,57 @@ export function getCacheDir() { export function getAppConfigDir(name: string) { return path.join(getConfigDir(), name) } + +/** + * 使用 jschardet 库检测文件编码格式 + * @param filePath - 文件路径 + * @returns 返回文件的编码格式，如 UTF-8, ascii, GB2312 等 + */ +export function detectEncoding(filePath: string): string { + // 读取文件前1KB来检测编码 + const buffer = Buffer.alloc(1024) + const fd = fs.openSync(filePath, 'r') + fs.readSync(fd, buffer, 0, 1024, 0) + fs.closeSync(fd) + const { encoding } = detectEncoding_(buffer) + return encoding +} + +/** + * 读取文件内容并自动检测编码格式进行解码 + * @param filePath - 文件路径 + * @returns 解码后的文件内容 + */ +export function readTextFileWithAutoEncoding(filePath: string) { + const encoding = detectEncoding(filePath) + const data = fs.readFileSync(filePath) + const content = iconv.decode(data, encoding) + + if (content.includes('\uFFFD') && encoding !== 'UTF-8') { + Logger.error(`文件 ${filePath} 自动识别编码为 ${encoding}，但包含错误字符。尝试其他编码`) + const buffer = Buffer.alloc(1024) + const fd = fs.openSync(filePath, 'r') + fs.readSync(fd, buffer, 0, 1024, 0) + fs.closeSync(fd) + const encodings = detectEncodingAll(buffer) + if (encodings.length > 0) { + for (const item of encodings) { + if (item.encoding === encoding) { + continue + } + Logger.log(`尝试使用 ${item.encoding} 解码文件 ${filePath}`) + const content = iconv.decode(buffer, item.encoding) + if (!content.includes('\uFFFD')) { + Logger.log(`文件 ${filePath} 解码成功，编码为 ${item.encoding}`) + return content + } else { + Logger.error(`文件 ${filePath} 使用 ${item.encoding} 解码失败，尝试下一个编码`) + } + } + } + Logger.error(`文件 ${filePath} 所有可能的编码均解码失败，尝试使用 UTF-8 解码`) + return iconv.decode(buffer, 'UTF-8') + } + + return content +} diff --git a/src/main/utils/zoom.ts b/src/main/utils/zoom.ts index 8e7145313c..d91d411591 100644 --- a/src/main/utils/zoom.ts +++ b/src/main/utils/zoom.ts @@ -1,26 +1,26 @@ -import { BrowserWindow } from 'electron' - -import { configManager } from '../services/ConfigManager' - -export function handleZoomFactor(wins: BrowserWindow[], delta: number, reset: boolean = false) { - if (reset) { - wins.forEach((win) => { - win.webContents.setZoomFactor(1) - }) - configManager.setZoomFactor(1) - return - } - - if (delta === 0) { - return - } - - const currentZoom = configManager.getZoomFactor() - const newZoom = Number((currentZoom + delta).toFixed(1)) - if (newZoom >= 0.5 && newZoom <= 2.0) { - wins.forEach((win) => { - win.webContents.setZoomFactor(newZoom) - }) - configManager.setZoomFactor(newZoom) - } -} +import { BrowserWindow } from 'electron' + +import { configManager } from '../services/ConfigManager' + +export function handleZoomFactor(wins: BrowserWindow[], delta: number, reset: boolean = false) { + if (reset) { + wins.forEach((win) => { + win.webContents.setZoomFactor(1) + }) + configManager.setZoomFactor(1) + return + } + + if (delta === 0) { + return + } + + const currentZoom = configManager.getZoomFactor() + const newZoom = Number((currentZoom + delta).toFixed(1)) + if (newZoom >= 0.5 && newZoom <= 2.0) { + wins.forEach((win) => { + win.webContents.setZoomFactor(newZoom) + }) + configManager.setZoomFactor(newZoom) + } +} diff --git a/src/preload/index.ts b/src/preload/index.ts index 3120492dde..ea1a2897f9 100644 --- a/src/preload/index.ts +++ b/src/preload/index.ts @@ -10,6 +10,7 @@ import { KnowledgeItem, MCPServer, Provider, + S3Config, Shortcut, ThemeMode, WebDavConfig @@ -72,9 +73,9 @@ const api = { decompress: (text: Buffer) => ipcRenderer.invoke(IpcChannel.Zip_Decompress, text) }, backup: { - backup: (fileName: string, data: string, destinationPath?: string, skipBackupFile?: boolean) => - ipcRenderer.invoke(IpcChannel.Backup_Backup, fileName, data, destinationPath, skipBackupFile), - restore: (backupPath: string) => ipcRenderer.invoke(IpcChannel.Backup_Restore, backupPath), + backup: (filename: string, content: string, path: string, skipBackupFile: boolean) => + ipcRenderer.invoke(IpcChannel.Backup_Backup, filename, content, path, skipBackupFile), + restore: (path: string) => ipcRenderer.invoke(IpcChannel.Backup_Restore, path), backupToWebdav: (data: string, webdavConfig: WebDavConfig) => ipcRenderer.invoke(IpcChannel.Backup_BackupToWebdav, data, webdavConfig), restoreFromWebdav: (webdavConfig: WebDavConfig) => @@ -86,7 +87,28 @@ const api = { createDirectory: (webdavConfig: WebDavConfig, path: string, options?: CreateDirectoryOptions) => ipcRenderer.invoke(IpcChannel.Backup_CreateDirectory, webdavConfig, path, options), deleteWebdavFile: (fileName: string, webdavConfig: WebDavConfig) => - ipcRenderer.invoke(IpcChannel.Backup_DeleteWebdavFile, fileName, webdavConfig) + ipcRenderer.invoke(IpcChannel.Backup_DeleteWebdavFile, fileName, webdavConfig), + backupToLocalDir: ( + data: string, + fileName: string, + localConfig: { localBackupDir?: string; skipBackupFile?: boolean } + ) => ipcRenderer.invoke(IpcChannel.Backup_BackupToLocalDir, data, fileName, localConfig), + restoreFromLocalBackup: (fileName: string, localBackupDir?: string) => + ipcRenderer.invoke(IpcChannel.Backup_RestoreFromLocalBackup, fileName, localBackupDir), + listLocalBackupFiles: (localBackupDir?: string) => + ipcRenderer.invoke(IpcChannel.Backup_ListLocalBackupFiles, localBackupDir), + deleteLocalBackupFile: (fileName: string, localBackupDir?: string) => + ipcRenderer.invoke(IpcChannel.Backup_DeleteLocalBackupFile, fileName, localBackupDir), + setLocalBackupDir: (dirPath: string) => ipcRenderer.invoke(IpcChannel.Backup_SetLocalBackupDir, dirPath), + checkWebdavConnection: (webdavConfig: WebDavConfig) => + ipcRenderer.invoke(IpcChannel.Backup_CheckConnection, webdavConfig), + + backupToS3: (data: string, s3Config: S3Config) => ipcRenderer.invoke(IpcChannel.Backup_BackupToS3, data, s3Config), + restoreFromS3: (s3Config: S3Config) => ipcRenderer.invoke(IpcChannel.Backup_RestoreFromS3, s3Config), + listS3Files: (s3Config: S3Config) => ipcRenderer.invoke(IpcChannel.Backup_ListS3Files, s3Config), + deleteS3File: (fileName: string, s3Config: S3Config) => + ipcRenderer.invoke(IpcChannel.Backup_DeleteS3File, fileName, s3Config), + checkS3Connection: (s3Config: S3Config) => ipcRenderer.invoke(IpcChannel.Backup_CheckS3Connection, s3Config) }, file: { select: (options?: OpenDialogOptions) => ipcRenderer.invoke(IpcChannel.File_Select, options), @@ -206,8 +228,8 @@ const api = { restartServer: (server: MCPServer) => ipcRenderer.invoke(IpcChannel.Mcp_RestartServer, server), stopServer: (server: MCPServer) => ipcRenderer.invoke(IpcChannel.Mcp_StopServer, server), listTools: (server: MCPServer) => ipcRenderer.invoke(IpcChannel.Mcp_ListTools, server), - callTool: ({ server, name, args }: { server: MCPServer; name: string; args: any }) => - ipcRenderer.invoke(IpcChannel.Mcp_CallTool, { server, name, args }), + callTool: ({ server, name, args, callId }: { server: MCPServer; name: string; args: any; callId?: string }) => + ipcRenderer.invoke(IpcChannel.Mcp_CallTool, { server, name, args, callId }), listPrompts: (server: MCPServer) => ipcRenderer.invoke(IpcChannel.Mcp_ListPrompts, server), getPrompt: ({ server, name, args }: { server: MCPServer; name: string; args?: Record }) => ipcRenderer.invoke(IpcChannel.Mcp_GetPrompt, { server, name, args }), @@ -215,7 +237,9 @@ const api = { getResource: ({ server, uri }: { server: MCPServer; uri: string }) => ipcRenderer.invoke(IpcChannel.Mcp_GetResource, { server, uri }), getInstallInfo: () => ipcRenderer.invoke(IpcChannel.Mcp_GetInstallInfo), - checkMcpConnectivity: (server: any) => ipcRenderer.invoke(IpcChannel.Mcp_CheckConnectivity, server) + checkMcpConnectivity: (server: any) => ipcRenderer.invoke(IpcChannel.Mcp_CheckConnectivity, server), + abortTool: (callId: string) => ipcRenderer.invoke(IpcChannel.Mcp_AbortTool, callId), + setProgress: (progress: number) => ipcRenderer.invoke(IpcChannel.Mcp_SetProgress, progress) }, python: { execute: (script: string, context?: Record, timeout?: number) => @@ -290,7 +314,9 @@ const api = { minimizeActionWindow: () => ipcRenderer.invoke(IpcChannel.Selection_ActionWindowMinimize), pinActionWindow: (isPinned: boolean) => ipcRenderer.invoke(IpcChannel.Selection_ActionWindowPin, isPinned) }, - quoteToMainWindow: (text: string) => ipcRenderer.invoke(IpcChannel.App_QuoteToMain, text) + quoteToMainWindow: (text: string) => ipcRenderer.invoke(IpcChannel.App_QuoteToMain, text), + setDisableHardwareAcceleration: (isDisable: boolean) => + ipcRenderer.invoke(IpcChannel.App_SetDisableHardwareAcceleration, isDisable) } // Use `contextBridge` APIs to expose Electron APIs to diff --git a/src/renderer/index.html b/src/renderer/index.html index c8832dc573..239d9c794c 100644 --- a/src/renderer/index.html +++ b/src/renderer/index.html @@ -1,46 +1,45 @@ + + + + + Cherry Studio - - - - - Cherry Studio + + - #spinner img { - width: 100px; - border-radius: 50px; - } - - - - -
-
- -
- - - - - - \ No newline at end of file + +
+
+ +
+ + + + + diff --git a/src/renderer/miniWindow.html b/src/renderer/miniWindow.html index c2748618f1..83b108b8a4 100644 --- a/src/renderer/miniWindow.html +++ b/src/renderer/miniWindow.html @@ -1,24 +1,23 @@ + + + + + Cherry Studio - - - - - Cherry Studio + + - - - - -
- - - - \ No newline at end of file + +
+ + + diff --git a/src/renderer/selectionAction.html b/src/renderer/selectionAction.html index 1dd3fa616c..1078b35264 100644 --- a/src/renderer/selectionAction.html +++ b/src/renderer/selectionAction.html @@ -1,41 +1,39 @@ - - + - + Cherry Studio Selection Assistant + - - - +
- - - \ No newline at end of file + + diff --git a/src/renderer/selectionToolbar.html b/src/renderer/selectionToolbar.html index 34efa7effc..f8f6777adb 100644 --- a/src/renderer/selectionToolbar.html +++ b/src/renderer/selectionToolbar.html @@ -1,46 +1,43 @@ + + + + + Cherry Studio Selection Toolbar + - - - - - Cherry Studio Selection Toolbar + +
+ + - - - \ No newline at end of file + #root { + margin: 0 !important; + padding: 0 !important; + width: max-content !important; + height: fit-content !important; + } + + + diff --git a/src/renderer/src/aiCore/clients/ApiClientFactory.ts b/src/renderer/src/aiCore/clients/ApiClientFactory.ts index 14e342da83..d815dc923f 100644 --- a/src/renderer/src/aiCore/clients/ApiClientFactory.ts +++ b/src/renderer/src/aiCore/clients/ApiClientFactory.ts @@ -47,10 +47,9 @@ export class ApiClientFactory { // 然后检查标准的provider type switch (provider.type) { case 'openai': - case 'azure-openai': - console.log(`[ApiClientFactory] Creating OpenAIApiClient for provider: ${provider.id}`) instance = new OpenAIAPIClient(provider) as BaseApiClient break + case 'azure-openai': case 'openai-response': instance = new OpenAIResponseAPIClient(provider) as BaseApiClient break diff --git a/src/renderer/src/aiCore/clients/NewAPIClient.ts b/src/renderer/src/aiCore/clients/NewAPIClient.ts index 3162cad0fe..769ca90acf 100644 --- a/src/renderer/src/aiCore/clients/NewAPIClient.ts +++ b/src/renderer/src/aiCore/clients/NewAPIClient.ts @@ -106,7 +106,7 @@ export class NewAPIClient extends BaseApiClient { return client } - if (model.endpoint_type === 'openai') { + if (model.endpoint_type === 'openai' || model.endpoint_type === 'image-generation') { const client = this.clients.get('openai') if (!client || !this.isValidClient(client)) { throw new Error('Failed to get openai client') diff --git a/src/renderer/src/aiCore/clients/anthropic/AnthropicAPIClient.ts b/src/renderer/src/aiCore/clients/anthropic/AnthropicAPIClient.ts index ebe76d8152..c946f114fe 100644 --- a/src/renderer/src/aiCore/clients/anthropic/AnthropicAPIClient.ts +++ b/src/renderer/src/aiCore/clients/anthropic/AnthropicAPIClient.ts @@ -49,7 +49,9 @@ import { LLMWebSearchCompleteChunk, LLMWebSearchInProgressChunk, MCPToolCreatedChunk, + TextCompleteChunk, TextDeltaChunk, + ThinkingCompleteChunk, ThinkingDeltaChunk } from '@renderer/types/chunk' import { type Message } from '@renderer/types/newMessage' @@ -517,7 +519,7 @@ export class AnthropicAPIClient extends BaseApiClient< return () => { let accumulatedJson = '' const toolCalls: Record = {} - + const ChunkIdTypeMap: Record = {} return { async transform(rawChunk: AnthropicSdkRawChunk, controller: TransformStreamDefaultController) { switch (rawChunk.type) { @@ -612,6 +614,19 @@ export class AnthropicAPIClient extends BaseApiClient< toolCalls[rawChunk.index] = contentBlock break } + case 'text': { + if (!ChunkIdTypeMap[rawChunk.index]) { + ChunkIdTypeMap[rawChunk.index] = ChunkType.TEXT_DELTA // 用textdelta代表文本块 + } + break + } + case 'thinking': + case 'redacted_thinking': { + if (!ChunkIdTypeMap[rawChunk.index]) { + ChunkIdTypeMap[rawChunk.index] = ChunkType.THINKING_DELTA // 用thinkingdelta代表思考块 + } + break + } } break } @@ -646,6 +661,15 @@ export class AnthropicAPIClient extends BaseApiClient< break } case 'content_block_stop': { + if (ChunkIdTypeMap[rawChunk.index] === ChunkType.TEXT_DELTA) { + controller.enqueue({ + type: ChunkType.TEXT_COMPLETE + } as TextCompleteChunk) + } else if (ChunkIdTypeMap[rawChunk.index] === ChunkType.THINKING_DELTA) { + controller.enqueue({ + type: ChunkType.THINKING_COMPLETE + } as ThinkingCompleteChunk) + } const toolCall = toolCalls[rawChunk.index] if (toolCall) { try { diff --git a/src/renderer/src/aiCore/clients/openai/OpenAIApiClient.ts b/src/renderer/src/aiCore/clients/openai/OpenAIApiClient.ts index 65e9cc67c4..c1994dcb95 100644 --- a/src/renderer/src/aiCore/clients/openai/OpenAIApiClient.ts +++ b/src/renderer/src/aiCore/clients/openai/OpenAIApiClient.ts @@ -564,11 +564,11 @@ export class OpenAIAPIClient extends OpenAIBaseClient< // Perplexity citations // @ts-ignore - citations may not be in standard type definitions - if (context.provider?.id === 'perplexity' && chunk.citations && chunk.citations.length > 0) { + if (context.provider?.id === 'perplexity' && chunk.search_results && chunk.search_results.length > 0) { hasBeenCollectedWebSearch = true return { // @ts-ignore - citations may not be in standard type definitions - results: chunk.citations, + results: chunk.search_results, source: WebSearchSource.PERPLEXITY } } @@ -672,74 +672,21 @@ export class OpenAIAPIClient extends OpenAIBaseClient< // 处理chunk if ('choices' in chunk && chunk.choices && chunk.choices.length > 0) { - const choice = chunk.choices[0] + for (const choice of chunk.choices) { + if (!choice) continue - if (!choice) return - - // 对于流式响应，使用 delta；对于非流式响应，使用 message。 - // 然而某些 OpenAI 兼容平台在非流式请求时会错误地返回一个空对象的 delta 字段。 - // 如果 delta 为空对象，应当忽略它并回退到 message，避免造成内容缺失。 - let contentSource: OpenAISdkRawContentSource | null = null - if ('delta' in choice && choice.delta && Object.keys(choice.delta).length > 0) { - contentSource = choice.delta - } else if ('message' in choice) { - contentSource = choice.message - } - - if (!contentSource) return - - const webSearchData = collectWebSearchData(chunk, contentSource, context) - if (webSearchData) { - controller.enqueue({ - type: ChunkType.LLM_WEB_SEARCH_COMPLETE, - llm_web_search: webSearchData - }) - } - - // 处理推理内容 (e.g. from OpenRouter DeepSeek-R1) - // @ts-ignore - reasoning_content is not in standard OpenAI types but some providers use it - const reasoningText = contentSource.reasoning_content || contentSource.reasoning - if (reasoningText) { - controller.enqueue({ - type: ChunkType.THINKING_DELTA, - text: reasoningText - }) - } - - // 处理文本内容 - if (contentSource.content) { - controller.enqueue({ - type: ChunkType.TEXT_DELTA, - text: contentSource.content - }) - } - - // 处理工具调用 - if (contentSource.tool_calls) { - for (const toolCall of contentSource.tool_calls) { - if ('index' in toolCall) { - const { id, index, function: fun } = toolCall - if (fun?.name) { - toolCalls[index] = { - id: id || '', - function: { - name: fun.name, - arguments: fun.arguments || '' - }, - type: 'function' - } - } else if (fun?.arguments) { - toolCalls[index].function.arguments += fun.arguments - } - } else { - toolCalls.push(toolCall) - } + // 对于流式响应，使用 delta；对于非流式响应，使用 message。 + // 然而某些 OpenAI 兼容平台在非流式请求时会错误地返回一个空对象的 delta 字段。 + // 如果 delta 为空对象，应当忽略它并回退到 message，避免造成内容缺失。 + let contentSource: OpenAISdkRawContentSource | null = null + if ('delta' in choice && choice.delta && Object.keys(choice.delta).length > 0) { + contentSource = choice.delta + } else if ('message' in choice) { + contentSource = choice.message } - } - // 处理finish_reason，发送流结束信号 - if ('finish_reason' in choice && choice.finish_reason) { - Logger.debug(`[OpenAIApiClient] Stream finished with reason: ${choice.finish_reason}`) + if (!contentSource) continue + const webSearchData = collectWebSearchData(chunk, contentSource, context) if (webSearchData) { controller.enqueue({ @@ -747,7 +694,60 @@ export class OpenAIAPIClient extends OpenAIBaseClient< llm_web_search: webSearchData }) } - emitCompletionSignals(controller) + + // 处理推理内容 (e.g. from OpenRouter DeepSeek-R1) + // @ts-ignore - reasoning_content is not in standard OpenAI types but some providers use it + const reasoningText = contentSource.reasoning_content || contentSource.reasoning + if (reasoningText) { + controller.enqueue({ + type: ChunkType.THINKING_DELTA, + text: reasoningText + }) + } + + // 处理文本内容 + if (contentSource.content) { + controller.enqueue({ + type: ChunkType.TEXT_DELTA, + text: contentSource.content + }) + } + + // 处理工具调用 + if (contentSource.tool_calls) { + for (const toolCall of contentSource.tool_calls) { + if ('index' in toolCall) { + const { id, index, function: fun } = toolCall + if (fun?.name) { + toolCalls[index] = { + id: id || '', + function: { + name: fun.name, + arguments: fun.arguments || '' + }, + type: 'function' + } + } else if (fun?.arguments) { + toolCalls[index].function.arguments += fun.arguments + } + } else { + toolCalls.push(toolCall) + } + } + } + + // 处理finish_reason，发送流结束信号 + if ('finish_reason' in choice && choice.finish_reason) { + Logger.debug(`[OpenAIApiClient] Stream finished with reason: ${choice.finish_reason}`) + const webSearchData = collectWebSearchData(chunk, contentSource, context) + if (webSearchData) { + controller.enqueue({ + type: ChunkType.LLM_WEB_SEARCH_COMPLETE, + llm_web_search: webSearchData + }) + } + emitCompletionSignals(controller) + } } } }, diff --git a/src/renderer/src/aiCore/clients/openai/OpenAIResponseAPIClient.ts b/src/renderer/src/aiCore/clients/openai/OpenAIResponseAPIClient.ts index 99e40ed818..2af0b8376f 100644 --- a/src/renderer/src/aiCore/clients/openai/OpenAIResponseAPIClient.ts +++ b/src/renderer/src/aiCore/clients/openai/OpenAIResponseAPIClient.ts @@ -2,6 +2,7 @@ import { GenericChunk } from '@renderer/aiCore/middleware/schemas' import { CompletionsContext } from '@renderer/aiCore/middleware/types' import { isOpenAIChatCompletionOnlyModel, + isOpenAILLMModel, isSupportedReasoningEffortOpenAIModel, isVisionModel } from '@renderer/config/models' @@ -64,10 +65,10 @@ export class OpenAIResponseAPIClient extends OpenAIBaseClient< * 根据模型特征选择合适的客户端 */ public getClient(model: Model) { - if (isOpenAIChatCompletionOnlyModel(model)) { - return this.client - } else { + if (isOpenAILLMModel(model) && !isOpenAIChatCompletionOnlyModel(model)) { return this + } else { + return this.client } } @@ -492,6 +493,10 @@ export class OpenAIResponseAPIClient extends OpenAIBaseClient< case 'response.output_item.added': if (chunk.item.type === 'function_call') { outputItems.push(chunk.item) + } else if (chunk.item.type === 'web_search_call') { + controller.enqueue({ + type: ChunkType.LLM_WEB_SEARCH_IN_PROGRESS + }) } break case 'response.reasoning_summary_part.added': diff --git a/src/renderer/src/aiCore/middleware/common/AbortHandlerMiddleware.ts b/src/renderer/src/aiCore/middleware/common/AbortHandlerMiddleware.ts index 7186cec12f..2acf553533 100644 --- a/src/renderer/src/aiCore/middleware/common/AbortHandlerMiddleware.ts +++ b/src/renderer/src/aiCore/middleware/common/AbortHandlerMiddleware.ts @@ -67,7 +67,12 @@ export const AbortHandlerMiddleware: CompletionsMiddleware = const streamWithAbortHandler = (result.stream as ReadableStream).pipeThrough( new TransformStream({ transform(chunk, controller) { - // 检查 abort 状态 + // 如果已经收到错误块，不再检查 abort 状态 + if (chunk.type === ChunkType.ERROR) { + controller.enqueue(chunk) + return + } + if (abortSignal?.aborted) { // 转换为 ErrorChunk const errorChunk: ErrorChunk = { diff --git a/src/renderer/src/aiCore/middleware/common/FinalChunkConsumerMiddleware.ts b/src/renderer/src/aiCore/middleware/common/FinalChunkConsumerMiddleware.ts index b0b9bd7ce6..80e0cdc5e6 100644 --- a/src/renderer/src/aiCore/middleware/common/FinalChunkConsumerMiddleware.ts +++ b/src/renderer/src/aiCore/middleware/common/FinalChunkConsumerMiddleware.ts @@ -136,7 +136,6 @@ function extractAndAccumulateUsageMetrics(ctx: CompletionsContext, chunk: Generi Logger.debug(`[${MIDDLEWARE_NAME}] First token timestamp: ${ctx._internal.customState.firstTokenTimestamp}`) } if (chunk.type === ChunkType.LLM_RESPONSE_COMPLETE) { - Logger.debug(`[${MIDDLEWARE_NAME}] LLM_RESPONSE_COMPLETE chunk received:`, ctx._internal) // 从LLM_RESPONSE_COMPLETE chunk中提取usage数据 if (chunk.response?.usage) { accumulateUsage(ctx._internal.observer.usage, chunk.response.usage) diff --git a/src/renderer/src/aiCore/middleware/core/McpToolChunkMiddleware.ts b/src/renderer/src/aiCore/middleware/core/McpToolChunkMiddleware.ts index 893018d4c5..c5156dbd53 100644 --- a/src/renderer/src/aiCore/middleware/core/McpToolChunkMiddleware.ts +++ b/src/renderer/src/aiCore/middleware/core/McpToolChunkMiddleware.ts @@ -89,6 +89,11 @@ function createToolHandlingTransform( let hasToolUseResponses = false let streamEnded = false + // 存储已执行的工具结果 + const executedToolResults: SdkMessageParam[] = [] + const executedToolCalls: SdkToolCall[] = [] + const executionPromises: Promise[] = [] + return new TransformStream({ async transform(chunk: GenericChunk, controller) { try { @@ -98,22 +103,64 @@ function createToolHandlingTransform( // 1. 处理Function Call方式的工具调用 if (createdChunk.tool_calls && createdChunk.tool_calls.length > 0) { - toolCalls.push(...createdChunk.tool_calls) hasToolCalls = true + + for (const toolCall of createdChunk.tool_calls) { + toolCalls.push(toolCall) + + const executionPromise = (async () => { + try { + const result = await executeToolCalls( + ctx, + [toolCall], + mcpTools, + allToolResponses, + currentParams.onChunk, + currentParams.assistant.model! + ) + + // 缓存执行结果 + executedToolResults.push(...result.toolResults) + executedToolCalls.push(...result.confirmedToolCalls) + } catch (error) { + console.error(`🔧 [${MIDDLEWARE_NAME}] Error executing tool call asynchronously:`, error) + } + })() + + executionPromises.push(executionPromise) + } } // 2. 处理Tool Use方式的工具调用 if (createdChunk.tool_use_responses && createdChunk.tool_use_responses.length > 0) { - toolUseResponses.push(...createdChunk.tool_use_responses) hasToolUseResponses = true + for (const toolUseResponse of createdChunk.tool_use_responses) { + toolUseResponses.push(toolUseResponse) + const executionPromise = (async () => { + try { + const result = await executeToolUseResponses( + ctx, + [toolUseResponse], // 单个执行 + mcpTools, + allToolResponses, + currentParams.onChunk, + currentParams.assistant.model! + ) + + // 缓存执行结果 + executedToolResults.push(...result.toolResults) + } catch (error) { + console.error(`🔧 [${MIDDLEWARE_NAME}] Error executing tool use response asynchronously:`, error) + // 错误时不影响其他工具的执行 + } + })() + + executionPromises.push(executionPromise) + } } - - // 不转发MCP工具进展chunks，避免重复处理 - return + } else { + controller.enqueue(chunk) } - - // 转发其他所有chunk - controller.enqueue(chunk) } catch (error) { console.error(`🔧 [${MIDDLEWARE_NAME}] Error processing chunk:`, error) controller.error(error) @@ -121,43 +168,33 @@ function createToolHandlingTransform( }, async flush(controller) { - const shouldExecuteToolCalls = hasToolCalls && toolCalls.length > 0 - const shouldExecuteToolUseResponses = hasToolUseResponses && toolUseResponses.length > 0 - - if (!streamEnded && (shouldExecuteToolCalls || shouldExecuteToolUseResponses)) { + // 在流结束时等待所有异步工具执行完成，然后进行递归调用 + if (!streamEnded && (hasToolCalls || hasToolUseResponses)) { streamEnded = true try { - let toolResult: SdkMessageParam[] = [] - - if (shouldExecuteToolCalls) { - toolResult = await executeToolCalls( - ctx, - toolCalls, - mcpTools, - allToolResponses, - currentParams.onChunk, - currentParams.assistant.model! - ) - } else if (shouldExecuteToolUseResponses) { - toolResult = await executeToolUseResponses( - ctx, - toolUseResponses, - mcpTools, - allToolResponses, - currentParams.onChunk, - currentParams.assistant.model! - ) - } - - if (toolResult.length > 0) { + await Promise.all(executionPromises) + if (executedToolResults.length > 0) { const output = ctx._internal.toolProcessingState?.output + const newParams = buildParamsWithToolResults( + ctx, + currentParams, + output, + executedToolResults, + executedToolCalls + ) + + // 在递归调用前通知UI开始新的LLM响应处理 + if (currentParams.onChunk) { + currentParams.onChunk({ + type: ChunkType.LLM_RESPONSE_CREATED + }) + } - const newParams = buildParamsWithToolResults(ctx, currentParams, output, toolResult, toolCalls) await executeWithToolHandling(newParams, depth + 1) } } catch (error) { - console.error(`🔧 [${MIDDLEWARE_NAME}] Error in tool processing:`, error) + Logger.error(`🔧 [${MIDDLEWARE_NAME}] Error in tool processing:`, error) controller.error(error) } finally { hasToolCalls = false @@ -178,8 +215,7 @@ async function executeToolCalls( allToolResponses: MCPToolResponse[], onChunk: CompletionsParams['onChunk'], model: Model -): Promise { - // 转换为MCPToolResponse格式 +): Promise<{ toolResults: SdkMessageParam[]; confirmedToolCalls: SdkToolCall[] }> { const mcpToolResponses: ToolCallResponse[] = toolCalls .map((toolCall) => { const mcpTool = ctx.apiClientInstance.convertSdkToolCallToMcp(toolCall, mcpTools) @@ -192,11 +228,11 @@ async function executeToolCalls( if (mcpToolResponses.length === 0) { console.warn(`🔧 [${MIDDLEWARE_NAME}] No valid MCP tool responses to execute`) - return [] + return { toolResults: [], confirmedToolCalls: [] } } // 使用现有的parseAndCallTools函数执行工具 - const toolResults = await parseAndCallTools( + const { toolResults, confirmedToolResponses } = await parseAndCallTools( mcpToolResponses, allToolResponses, onChunk, @@ -204,10 +240,24 @@ async function executeToolCalls( return ctx.apiClientInstance.convertMcpToolResponseToSdkMessageParam(mcpToolResponse, resp, model) }, model, - mcpTools + mcpTools, + ctx._internal?.flowControl?.abortSignal ) - return toolResults + // 找出已确认工具对应的原始toolCalls + const confirmedToolCalls = toolCalls.filter((toolCall) => { + return confirmedToolResponses.find((confirmed) => { + // 根据不同的ID字段匹配原始toolCall + return ( + ('name' in toolCall && + (toolCall.name?.includes(confirmed.tool.name) || toolCall.name?.includes(confirmed.tool.id))) || + confirmed.tool.name === toolCall.id || + confirmed.tool.id === toolCall.id + ) + }) + }) + + return { toolResults, confirmedToolCalls } } /** @@ -221,9 +271,9 @@ async function executeToolUseResponses( allToolResponses: MCPToolResponse[], onChunk: CompletionsParams['onChunk'], model: Model -): Promise { +): Promise<{ toolResults: SdkMessageParam[] }> { // 直接使用parseAndCallTools函数处理已经解析好的ToolUseResponse - const toolResults = await parseAndCallTools( + const { toolResults } = await parseAndCallTools( toolUseResponses, allToolResponses, onChunk, @@ -231,10 +281,11 @@ async function executeToolUseResponses( return ctx.apiClientInstance.convertMcpToolResponseToSdkMessageParam(mcpToolResponse, resp, model) }, model, - mcpTools + mcpTools, + ctx._internal?.flowControl?.abortSignal ) - return toolResults + return { toolResults } } /** @@ -245,7 +296,7 @@ function buildParamsWithToolResults( currentParams: CompletionsParams, output: SdkRawOutput | string | undefined, toolResults: SdkMessageParam[], - toolCalls: SdkToolCall[] + confirmedToolCalls: SdkToolCall[] ): CompletionsParams { // 获取当前已经转换好的reqMessages，如果没有则使用原始messages const currentReqMessages = getCurrentReqMessages(ctx) @@ -253,7 +304,7 @@ function buildParamsWithToolResults( const apiClient = ctx.apiClientInstance // 从回复中构建助手消息 - const newReqMessages = apiClient.buildSdkMessages(currentReqMessages, output, toolResults, toolCalls) + const newReqMessages = apiClient.buildSdkMessages(currentReqMessages, output, toolResults, confirmedToolCalls) if (output && ctx._internal.toolProcessingState) { ctx._internal.toolProcessingState.output = undefined diff --git a/src/renderer/src/aiCore/middleware/core/TextChunkMiddleware.ts b/src/renderer/src/aiCore/middleware/core/TextChunkMiddleware.ts index 2a3255356f..3905d52058 100644 --- a/src/renderer/src/aiCore/middleware/core/TextChunkMiddleware.ts +++ b/src/renderer/src/aiCore/middleware/core/TextChunkMiddleware.ts @@ -1,5 +1,5 @@ import Logger from '@renderer/config/logger' -import { ChunkType, TextDeltaChunk } from '@renderer/types/chunk' +import { ChunkType, TextCompleteChunk, TextDeltaChunk } from '@renderer/types/chunk' import { CompletionsParams, CompletionsResult, GenericChunk } from '../schemas' import { CompletionsContext, CompletionsMiddleware } from '../types' @@ -38,7 +38,7 @@ export const TextChunkMiddleware: CompletionsMiddleware = // 用于跨chunk的状态管理 let accumulatedTextContent = '' - let hasEnqueue = false + let hasTextCompleteEventEnqueue = false const enhancedTextStream = resultFromUpstream.pipeThrough( new TransformStream({ transform(chunk: GenericChunk, controller) { @@ -53,30 +53,44 @@ export const TextChunkMiddleware: CompletionsMiddleware = // 创建新的chunk，包含处理后的文本 controller.enqueue(chunk) - } else if (accumulatedTextContent) { - if (chunk.type !== ChunkType.LLM_RESPONSE_COMPLETE) { - controller.enqueue(chunk) - hasEnqueue = true - } - const finalText = accumulatedTextContent - ctx._internal.customState!.accumulatedText = finalText - if (ctx._internal.toolProcessingState && !ctx._internal.toolProcessingState?.output) { - ctx._internal.toolProcessingState.output = finalText - } - - // 处理 onResponse 回调 - 发送最终完整文本 - if (params.onResponse) { - params.onResponse(finalText, true) - } - + } else if (chunk.type === ChunkType.TEXT_COMPLETE) { + const textChunk = chunk as TextCompleteChunk controller.enqueue({ - type: ChunkType.TEXT_COMPLETE, - text: finalText + ...textChunk, + text: accumulatedTextContent }) + if (params.onResponse) { + params.onResponse(accumulatedTextContent, true) + } + hasTextCompleteEventEnqueue = true accumulatedTextContent = '' - if (!hasEnqueue) { + } else if (accumulatedTextContent && !hasTextCompleteEventEnqueue) { + if (chunk.type === ChunkType.LLM_RESPONSE_COMPLETE) { + const finalText = accumulatedTextContent + ctx._internal.customState!.accumulatedText = finalText + if (ctx._internal.toolProcessingState && !ctx._internal.toolProcessingState?.output) { + ctx._internal.toolProcessingState.output = finalText + } + + // 处理 onResponse 回调 - 发送最终完整文本 + if (params.onResponse) { + params.onResponse(finalText, true) + } + + controller.enqueue({ + type: ChunkType.TEXT_COMPLETE, + text: finalText + }) + controller.enqueue(chunk) + } else { + controller.enqueue({ + type: ChunkType.TEXT_COMPLETE, + text: accumulatedTextContent + }) controller.enqueue(chunk) } + hasTextCompleteEventEnqueue = true + accumulatedTextContent = '' } else { // 其他类型的chunk直接传递 controller.enqueue(chunk) diff --git a/src/renderer/src/aiCore/middleware/core/ThinkChunkMiddleware.ts b/src/renderer/src/aiCore/middleware/core/ThinkChunkMiddleware.ts index b0df8313a5..dccdde7f10 100644 --- a/src/renderer/src/aiCore/middleware/core/ThinkChunkMiddleware.ts +++ b/src/renderer/src/aiCore/middleware/core/ThinkChunkMiddleware.ts @@ -65,6 +65,16 @@ export const ThinkChunkMiddleware: CompletionsMiddleware = thinking_millsec: thinkingStartTime > 0 ? Date.now() - thinkingStartTime : 0 } controller.enqueue(enhancedChunk) + } else if (chunk.type === ChunkType.THINKING_COMPLETE) { + const thinkingCompleteChunk = chunk as ThinkingCompleteChunk + controller.enqueue({ + ...thinkingCompleteChunk, + text: accumulatedThinkingContent, + thinking_millsec: thinkingStartTime > 0 ? Date.now() - thinkingStartTime : 0 + }) + hasThinkingContent = false + accumulatedThinkingContent = '' + thinkingStartTime = 0 } else if (hasThinkingContent && thinkingStartTime > 0) { // 收到任何非THINKING_DELTA的chunk时，如果有累积的思考内容，生成THINKING_COMPLETE const thinkingCompleteChunk: ThinkingCompleteChunk = { diff --git a/src/renderer/src/aiCore/middleware/core/WebSearchMiddleware.ts b/src/renderer/src/aiCore/middleware/core/WebSearchMiddleware.ts index 70915abffa..d4c8f71eff 100644 --- a/src/renderer/src/aiCore/middleware/core/WebSearchMiddleware.ts +++ b/src/renderer/src/aiCore/middleware/core/WebSearchMiddleware.ts @@ -42,7 +42,12 @@ export const WebSearchMiddleware: CompletionsMiddleware = const providerType = model.provider || 'openai' // 使用当前可用的Web搜索结果进行链接转换 const text = chunk.text - const result = smartLinkConverter(text, providerType, isFirstChunk) + const result = smartLinkConverter( + text, + providerType, + isFirstChunk, + ctx._internal.webSearchState!.results + ) if (isFirstChunk) { isFirstChunk = false } diff --git a/src/renderer/src/aiCore/middleware/feat/ThinkingTagExtractionMiddleware.ts b/src/renderer/src/aiCore/middleware/feat/ThinkingTagExtractionMiddleware.ts index fe2d51d8de..2fcefc1b34 100644 --- a/src/renderer/src/aiCore/middleware/feat/ThinkingTagExtractionMiddleware.ts +++ b/src/renderer/src/aiCore/middleware/feat/ThinkingTagExtractionMiddleware.ts @@ -69,7 +69,7 @@ export const ThinkingTagExtractionMiddleware: CompletionsMiddleware = const extractionResults = tagExtractor.processText(textChunk.text) for (const extractionResult of extractionResults) { - if (extractionResult.complete && extractionResult.tagContentExtracted) { + if (extractionResult.complete && extractionResult.tagContentExtracted?.trim()) { // 生成 THINKING_COMPLETE 事件 const thinkingCompleteChunk: ThinkingCompleteChunk = { type: ChunkType.THINKING_COMPLETE, @@ -89,12 +89,14 @@ export const ThinkingTagExtractionMiddleware: CompletionsMiddleware = thinkingStartTime = Date.now() } - const thinkingDeltaChunk: ThinkingDeltaChunk = { - type: ChunkType.THINKING_DELTA, - text: extractionResult.content, - thinking_millsec: thinkingStartTime > 0 ? Date.now() - thinkingStartTime : 0 + if (extractionResult.content?.trim()) { + const thinkingDeltaChunk: ThinkingDeltaChunk = { + type: ChunkType.THINKING_DELTA, + text: extractionResult.content, + thinking_millsec: thinkingStartTime > 0 ? Date.now() - thinkingStartTime : 0 + } + controller.enqueue(thinkingDeltaChunk) } - controller.enqueue(thinkingDeltaChunk) } else { // 发送清理后的文本内容 const cleanTextChunk: TextDeltaChunk = { diff --git a/src/renderer/src/aiCore/middleware/feat/ToolUseExtractionMiddleware.ts b/src/renderer/src/aiCore/middleware/feat/ToolUseExtractionMiddleware.ts index 5f444953a9..b53d7348f1 100644 --- a/src/renderer/src/aiCore/middleware/feat/ToolUseExtractionMiddleware.ts +++ b/src/renderer/src/aiCore/middleware/feat/ToolUseExtractionMiddleware.ts @@ -22,7 +22,8 @@ const TOOL_USE_TAG_CONFIG: TagConfig = { * 1. 从文本流中检测并提取 标签 * 2. 解析工具调用信息并转换为 ToolUseResponse 格式 * 3. 生成 MCP_TOOL_CREATED chunk 供 McpToolChunkMiddleware 处理 - * 4. 清理文本流，移除工具使用标签但保留正常文本 + * 4. 丢弃 tool_use 之后的所有内容（助手幻觉） + * 5. 清理文本流，移除工具使用标签但保留正常文本 * * 注意：此中间件只负责提取和转换，实际工具调用由 McpToolChunkMiddleware 处理 */ @@ -32,13 +33,10 @@ export const ToolUseExtractionMiddleware: CompletionsMiddleware = async (ctx: CompletionsContext, params: CompletionsParams): Promise => { const mcpTools = params.mcpTools || [] - // 如果没有工具，直接调用下一个中间件 if (!mcpTools || mcpTools.length === 0) return next(ctx, params) - // 调用下游中间件 const result = await next(ctx, params) - // 响应后处理：处理工具使用标签提取 if (result.stream) { const resultFromUpstream = result.stream as ReadableStream @@ -60,7 +58,9 @@ function createToolUseExtractionTransform( _ctx: CompletionsContext, mcpTools: MCPTool[] ): TransformStream { - const tagExtractor = new TagExtractor(TOOL_USE_TAG_CONFIG) + const toolUseExtractor = new TagExtractor(TOOL_USE_TAG_CONFIG) + let hasAnyToolUse = false + let toolCounter = 0 return new TransformStream({ async transform(chunk: GenericChunk, controller) { @@ -68,30 +68,37 @@ function createToolUseExtractionTransform( // 处理文本内容，检测工具使用标签 if (chunk.type === ChunkType.TEXT_DELTA) { const textChunk = chunk as TextDeltaChunk - const extractionResults = tagExtractor.processText(textChunk.text) - for (const result of extractionResults) { + // 处理 tool_use 标签 + const toolUseResults = toolUseExtractor.processText(textChunk.text) + + for (const result of toolUseResults) { if (result.complete && result.tagContentExtracted) { // 提取到完整的工具使用内容，解析并转换为 SDK ToolCall 格式 - const toolUseResponses = parseToolUse(result.tagContentExtracted, mcpTools) + const toolUseResponses = parseToolUse(result.tagContentExtracted, mcpTools, toolCounter) + toolCounter += toolUseResponses.length if (toolUseResponses.length > 0) { - // 生成 MCP_TOOL_CREATED chunk，复用现有的处理流程 + // 生成 MCP_TOOL_CREATED chunk const mcpToolCreatedChunk: MCPToolCreatedChunk = { type: ChunkType.MCP_TOOL_CREATED, tool_use_responses: toolUseResponses } controller.enqueue(mcpToolCreatedChunk) + + // 标记已有工具调用 + hasAnyToolUse = true } } else if (!result.isTagContent && result.content) { - // 发送标签外的正常文本内容 - const cleanTextChunk: TextDeltaChunk = { - ...textChunk, - text: result.content + if (!hasAnyToolUse) { + const cleanTextChunk: TextDeltaChunk = { + ...textChunk, + text: result.content + } + controller.enqueue(cleanTextChunk) } - controller.enqueue(cleanTextChunk) } - // 注意：标签内的内容不会作为TEXT_DELTA转发，避免重复显示 + // tool_use 标签内的内容不转发，避免重复显示 } return } @@ -105,16 +112,17 @@ function createToolUseExtractionTransform( }, async flush(controller) { - // 检查是否有未完成的标签内容 - const finalResult = tagExtractor.finalize() - if (finalResult && finalResult.tagContentExtracted) { - const toolUseResponses = parseToolUse(finalResult.tagContentExtracted, mcpTools) + // 检查是否有未完成的 tool_use 标签内容 + const finalToolUseResult = toolUseExtractor.finalize() + if (finalToolUseResult && finalToolUseResult.tagContentExtracted) { + const toolUseResponses = parseToolUse(finalToolUseResult.tagContentExtracted, mcpTools, toolCounter) if (toolUseResponses.length > 0) { const mcpToolCreatedChunk: MCPToolCreatedChunk = { type: ChunkType.MCP_TOOL_CREATED, tool_use_responses: toolUseResponses } controller.enqueue(mcpToolCreatedChunk) + hasAnyToolUse = true } } } diff --git a/src/renderer/src/assets/fonts/country-flag-fonts/flag.css b/src/renderer/src/assets/fonts/country-flag-fonts/flag.css index b3daed9da0..b73d4ae76f 100644 --- a/src/renderer/src/assets/fonts/country-flag-fonts/flag.css +++ b/src/renderer/src/assets/fonts/country-flag-fonts/flag.css @@ -1,13 +1,13 @@ -@font-face { - font-family: 'Twemoji Country Flags'; - unicode-range: - U+1F1E6-1F1FF, U+1F3F4, U+E0062-E0063, U+E0065, U+E0067, U+E006C, U+E006E, U+E0073-E0074, U+E0077, U+E007F; - /*https://github.com/beyondkmp/country-flag-emoji-polyfill/blob/master/font/TwemojiCountryFlags.woff2 */ - src: url('TwemojiCountryFlags.woff2') format('woff2'); - font-display: swap; -} - -/* 国旗字体样式类 */ -.country-flag-font { - font-family: 'Twemoji Country Flags', 'Apple Color Emoji', 'Segoe UI Emoji', sans-serif; -} +@font-face { + font-family: 'Twemoji Country Flags'; + unicode-range: + U+1F1E6-1F1FF, U+1F3F4, U+E0062-E0063, U+E0065, U+E0067, U+E006C, U+E006E, U+E0073-E0074, U+E0077, U+E007F; + /*https://github.com/beyondkmp/country-flag-emoji-polyfill/blob/master/font/TwemojiCountryFlags.woff2 */ + src: url('TwemojiCountryFlags.woff2') format('woff2'); + font-display: swap; +} + +/* 国旗字体样式类 */ +.country-flag-font { + font-family: 'Twemoji Country Flags', 'Apple Color Emoji', 'Segoe UI Emoji', sans-serif; +} diff --git a/src/renderer/src/assets/styles/color.scss b/src/renderer/src/assets/styles/color.scss index ce7e9cefe9..3f23425afc 100644 --- a/src/renderer/src/assets/styles/color.scss +++ b/src/renderer/src/assets/styles/color.scss @@ -72,6 +72,10 @@ --chat-text-user: var(--color-black); --list-item-border-radius: 20px; + + --color-status-success: #52c41a; + --color-status-error: #ff4d4f; + --color-status-warning: #faad14; } [theme-mode='light'] { diff --git a/src/renderer/src/assets/styles/font.scss b/src/renderer/src/assets/styles/font.scss index 5db6290bfc..75a0e6fc8b 100644 --- a/src/renderer/src/assets/styles/font.scss +++ b/src/renderer/src/assets/styles/font.scss @@ -1,20 +1,20 @@ -:root { - --font-family: - Ubuntu, -apple-system, BlinkMacSystemFont, 'Segoe UI', system-ui, Roboto, Oxygen, Cantarell, 'Open Sans', - 'Helvetica Neue', Arial, 'Noto Sans', sans-serif, 'Apple Color Emoji', 'Segoe UI Emoji', 'Segoe UI Symbol', - 'Noto Color Emoji'; - - --font-family-serif: - serif, -apple-system, BlinkMacSystemFont, 'Segoe UI', system-ui, Ubuntu, Roboto, Oxygen, Cantarell, 'Open Sans', - 'Helvetica Neue', Arial, 'Noto Sans', 'Apple Color Emoji', 'Segoe UI Emoji', 'Segoe UI Symbol', 'Noto Color Emoji'; - - --code-font-family: 'Cascadia Code', 'Fira Code', 'Consolas', Menlo, Courier, monospace; -} - -// Windows系统专用字体配置 -body[os='windows'] { - --font-family: - 'Twemoji Country Flags', Ubuntu, -apple-system, BlinkMacSystemFont, 'Segoe UI', system-ui, Roboto, Oxygen, - Cantarell, 'Open Sans', 'Helvetica Neue', Arial, 'Noto Sans', sans-serif, 'Apple Color Emoji', 'Segoe UI Emoji', - 'Segoe UI Symbol', 'Noto Color Emoji'; -} +:root { + --font-family: + Ubuntu, -apple-system, BlinkMacSystemFont, 'Segoe UI', system-ui, Roboto, Oxygen, Cantarell, 'Open Sans', + 'Helvetica Neue', Arial, 'Noto Sans', sans-serif, 'Apple Color Emoji', 'Segoe UI Emoji', 'Segoe UI Symbol', + 'Noto Color Emoji'; + + --font-family-serif: + serif, -apple-system, BlinkMacSystemFont, 'Segoe UI', system-ui, Ubuntu, Roboto, Oxygen, Cantarell, 'Open Sans', + 'Helvetica Neue', Arial, 'Noto Sans', 'Apple Color Emoji', 'Segoe UI Emoji', 'Segoe UI Symbol', 'Noto Color Emoji'; + + --code-font-family: 'Cascadia Code', 'Fira Code', 'Consolas', Menlo, Courier, monospace; +} + +// Windows系统专用字体配置 +body[os='windows'] { + --font-family: + 'Twemoji Country Flags', Ubuntu, -apple-system, BlinkMacSystemFont, 'Segoe UI', system-ui, Roboto, Oxygen, + Cantarell, 'Open Sans', 'Helvetica Neue', Arial, 'Noto Sans', sans-serif, 'Apple Color Emoji', 'Segoe UI Emoji', + 'Segoe UI Symbol', 'Noto Color Emoji'; +} diff --git a/src/renderer/src/assets/styles/markdown.scss b/src/renderer/src/assets/styles/markdown.scss index 2cb60e6bb4..d497c6e793 100644 --- a/src/renderer/src/assets/styles/markdown.scss +++ b/src/renderer/src/assets/styles/markdown.scss @@ -326,6 +326,8 @@ mjx-container { /* Shiki 相关样式 */ .shiki { font-family: var(--code-font-family); + // 保持行高为初始值，在 shiki 代码块中处理 + line-height: initial; } /* CodeMirror 相关样式 */ diff --git a/src/renderer/src/assets/styles/scrollbar.scss b/src/renderer/src/assets/styles/scrollbar.scss index c5df842f78..21039de9c2 100644 --- a/src/renderer/src/assets/styles/scrollbar.scss +++ b/src/renderer/src/assets/styles/scrollbar.scss @@ -49,3 +49,11 @@ pre:not(.shiki)::-webkit-scrollbar-thumb { --color-scrollbar-thumb: var(--color-scrollbar-thumb-light); --color-scrollbar-thumb-hover: var(--color-scrollbar-thumb-light-hover); } + +/* 用于截图时隐藏滚动条 + * FIXME: 临时方案，因为 html-to-image 没有正确处理伪元素。 + */ +.hide-scrollbar, +.hide-scrollbar * { + scrollbar-width: none !important; +} diff --git a/src/renderer/src/components/CodeBlockView/CodePreview.tsx b/src/renderer/src/components/CodeBlockView/CodePreview.tsx index 106d03d21f..3df9491e13 100644 --- a/src/renderer/src/components/CodeBlockView/CodePreview.tsx +++ b/src/renderer/src/components/CodeBlockView/CodePreview.tsx @@ -9,6 +9,7 @@ import { debounce } from 'lodash' import { ChevronsDownUp, ChevronsUpDown, Text as UnWrapIcon, WrapText as WrapIcon } from 'lucide-react' import React, { memo, useCallback, useEffect, useLayoutEffect, useMemo, useRef, useState } from 'react' import { useTranslation } from 'react-i18next' +import { ThemedToken } from 'shiki/core' import styled from 'styled-components' interface CodePreviewProps { @@ -107,7 +108,8 @@ const CodePreview = ({ children, language, setTools }: CodePreviewProps) => { // Virtualizer 配置 const getScrollElement = useCallback(() => scrollerRef.current, []) const getItemKey = useCallback((index: number) => `${callerId}-${index}`, [callerId]) - const estimateSize = useCallback(() => (fontSize - 1) * 1.6, [fontSize]) // 同步全局样式 + // `line-height: 1.6` 为全局样式，但是为了避免测量误差在这里取整 + const estimateSize = useCallback(() => Math.round((fontSize - 1) * 1.6), [fontSize]) // 创建 virtualizer 实例 const virtualizer = useVirtualizer({ @@ -144,11 +146,13 @@ const CodePreview = ({ children, language, setTools }: CodePreviewProps) => { ref={scrollerRef} className="shiki-scroller" $wrap={shouldWrap} + $lineHeight={estimateSize()} style={ { '--gutter-width': `${gutterDigits}ch`, fontSize: `${fontSize - 1}px`, - maxHeight: shouldCollapse ? MAX_COLLAPSE_HEIGHT : undefined + maxHeight: shouldCollapse ? MAX_COLLAPSE_HEIGHT : undefined, + overflowY: shouldCollapse ? 'auto' : 'hidden' } as React.CSSProperties }>
{ CodePreview.displayName = 'CodePreview' +/** + * 补全代码行 tokens，把原始内容拼接到高亮内容之后，确保渲染出整行来。 + */ +function completeLineTokens(themedTokens: ThemedToken[], rawLine: string): ThemedToken[] { + // 如果出现空行，补一个空格保证行高 + if (rawLine.length === 0) { + return [ + { + content: ' ', + offset: 0, + color: 'inherit', + bgColor: 'inherit', + htmlStyle: { + opacity: '0.35' + } + } + ] + } + + const themedContent = themedTokens.map((token) => token.content).join('') + const extraContent = rawLine.slice(themedContent.length) + + // 已有内容已经全部高亮，直接返回 + if (!extraContent) return themedTokens + + // 补全剩余内容 + return [ + ...themedTokens, + { + content: extraContent, + offset: themedContent.length, + color: 'inherit', + bgColor: 'inherit', + htmlStyle: { + opacity: '0.35' + } + } + ] +} + interface VirtualizedRowData { rawLine: string - tokenLine?: any[] + tokenLine?: ThemedToken[] showLineNumbers: boolean } @@ -208,17 +252,11 @@ const VirtualizedRow = memo(
{showLineNumbers && {index + 1}} - {tokenLine ? ( - // 渲染高亮后的内容 - tokenLine.map((token, tokenIndex) => ( - - {token.content} - - )) - ) : ( - // 渲染原始内容 - {rawLine || ' '} - )} + {completeLineTokens(tokenLine ?? [], rawLine).map((token, tokenIndex) => ( + + {token.content} + + ))}
) @@ -229,18 +267,19 @@ VirtualizedRow.displayName = 'VirtualizedRow' const ScrollContainer = styled.div<{ $wrap?: boolean + $lineHeight?: number }>` display: block; - overflow: auto; + overflow-x: auto; position: relative; border-radius: inherit; - height: auto; padding: 0.5em 1em; .line { display: flex; align-items: flex-start; width: 100%; + line-height: ${(props) => props.$lineHeight}px; .line-number { width: var(--gutter-width, 1.2ch); @@ -250,23 +289,17 @@ const ScrollContainer = styled.div<{ user-select: none; flex-shrink: 0; overflow: hidden; - line-height: inherit; font-family: inherit; font-variant-numeric: tabular-nums; } .line-content { flex: 1; - line-height: inherit; * { white-space: ${(props) => (props.$wrap ? 'pre-wrap' : 'pre')}; overflow-wrap: ${(props) => (props.$wrap ? 'break-word' : 'normal')}; } } - - .line-content-raw { - opacity: 0.35; - } } ` diff --git a/src/renderer/src/components/CodeBlockView/index.tsx b/src/renderer/src/components/CodeBlockView/index.tsx index c25ab3079d..98297428ec 100644 --- a/src/renderer/src/components/CodeBlockView/index.tsx +++ b/src/renderer/src/components/CodeBlockView/index.tsx @@ -292,6 +292,7 @@ const SplitViewWrapper = styled.div` &:not(:has(+ [class*='Container'])) { border-radius: 0 0 8px 8px; + overflow: hidden; } ` diff --git a/src/renderer/src/components/CodeEditor/hook.ts b/src/renderer/src/components/CodeEditor/hook.ts deleted file mode 100644 index c5bbab2d0d..0000000000 --- a/src/renderer/src/components/CodeEditor/hook.ts +++ /dev/null @@ -1,65 +0,0 @@ -import { useCodeStyle } from '@renderer/context/CodeStyleProvider' -import { Extension } from '@uiw/react-codemirror' -import { useEffect, useState } from 'react' - -let linterPromise: Promise | null = null -function importLintPackage() { - if (!linterPromise) { - linterPromise = import('@codemirror/lint').then((mod) => mod.linter) - } - return linterPromise -} - -// 语言对应的 linter 加载器 -const linterLoaders: Record Promise> = { - json: async () => { - const [linter, jsonParseLinter] = await Promise.all([ - importLintPackage(), - import('@codemirror/lang-json').then((mod) => mod.jsonParseLinter) - ]) - return linter(jsonParseLinter()) - } -} - -export const useLanguageExtensions = (language: string, lint?: boolean) => { - const { languageMap } = useCodeStyle() - const [extensions, setExtensions] = useState([]) - - // 加载语言 - useEffect(() => { - let normalizedLang = languageMap[language as keyof typeof languageMap] || language.toLowerCase() - - // 如果语言名包含 `-`，转换为驼峰命名法 - if (normalizedLang.includes('-')) { - normalizedLang = normalizedLang.replace(/-([a-z])/g, (_, char) => char.toUpperCase()) - } - - import('@uiw/codemirror-extensions-langs') - .then(({ loadLanguage }) => { - const extension = loadLanguage(normalizedLang as any) - if (extension) { - setExtensions((prev) => [...prev, extension]) - } - }) - .catch((error) => { - console.debug(`Failed to load language: ${normalizedLang}`, error) - }) - }, [language, languageMap]) - - useEffect(() => { - if (!lint) return - - const loader = linterLoaders[language] - if (loader) { - loader() - .then((extension) => { - setExtensions((prev) => [...prev, extension]) - }) - .catch((error) => { - console.error(`Failed to load linter for ${language}`, error) - }) - } - }, [language, lint]) - - return extensions -} diff --git a/src/renderer/src/components/CodeEditor/hooks.ts b/src/renderer/src/components/CodeEditor/hooks.ts new file mode 100644 index 0000000000..71d74ca3a5 --- /dev/null +++ b/src/renderer/src/components/CodeEditor/hooks.ts @@ -0,0 +1,108 @@ +import { linter } from '@codemirror/lint' // statically imported by @uiw/codemirror-extensions-basic-setup +import { EditorView } from '@codemirror/view' +import { useCodeStyle } from '@renderer/context/CodeStyleProvider' +import { Extension, keymap } from '@uiw/react-codemirror' +import { useEffect, useMemo, useState } from 'react' + +// 语言对应的 linter 加载器 +const linterLoaders: Record Promise> = { + json: async () => { + const jsonParseLinter = await import('@codemirror/lang-json').then((mod) => mod.jsonParseLinter) + return linter(jsonParseLinter()) + } +} + +export const useLanguageExtensions = (language: string, lint?: boolean) => { + const { languageMap } = useCodeStyle() + const [extensions, setExtensions] = useState([]) + + // 加载语言 + useEffect(() => { + let normalizedLang = languageMap[language as keyof typeof languageMap] || language.toLowerCase() + + // 如果语言名包含 `-`，转换为驼峰命名法 + if (normalizedLang.includes('-')) { + normalizedLang = normalizedLang.replace(/-([a-z])/g, (_, char) => char.toUpperCase()) + } + + import('@uiw/codemirror-extensions-langs') + .then(({ loadLanguage }) => { + const extension = loadLanguage(normalizedLang as any) + if (extension) { + setExtensions((prev) => [...prev, extension]) + } + }) + .catch((error) => { + console.debug(`Failed to load language: ${normalizedLang}`, error) + }) + }, [language, languageMap]) + + useEffect(() => { + if (!lint) return + + const loader = linterLoaders[language] + if (loader) { + loader() + .then((extension) => { + setExtensions((prev) => [...prev, extension]) + }) + .catch((error) => { + console.error(`Failed to load linter for ${language}`, error) + }) + } + }, [language, lint]) + + return extensions +} + +interface UseSaveKeymapProps { + onSave?: (content: string) => void + enabled?: boolean +} + +/** + * CodeMirror 扩展，用于处理保存快捷键 (Cmd/Ctrl + S) + * @param onSave 保存时触发的回调函数 + * @param enabled 是否启用此快捷键 + * @returns 扩展或空数组 + */ +export function useSaveKeymap({ onSave, enabled = true }: UseSaveKeymapProps) { + return useMemo(() => { + if (!enabled || !onSave) { + return [] + } + + return keymap.of([ + { + key: 'Mod-s', + run: (view: EditorView) => { + onSave(view.state.doc.toString()) + return true + }, + preventDefault: true + } + ]) + }, [onSave, enabled]) +} + +interface UseBlurHandlerProps { + onBlur?: (content: string) => void +} + +/** + * CodeMirror 扩展，用于处理编辑器的 blur 事件 + * @param onBlur blur 事件触发时的回调函数 + * @returns 扩展或空数组 + */ +export function useBlurHandler({ onBlur }: UseBlurHandlerProps) { + return useMemo(() => { + if (!onBlur) { + return [] + } + return EditorView.domEventHandlers({ + blur: (_event, view) => { + onBlur(view.state.doc.toString()) + } + }) + }, [onBlur]) +} diff --git a/src/renderer/src/components/CodeEditor/index.tsx b/src/renderer/src/components/CodeEditor/index.tsx index db699fa030..db7dd5f1ba 100644 --- a/src/renderer/src/components/CodeEditor/index.tsx +++ b/src/renderer/src/components/CodeEditor/index.tsx @@ -1,7 +1,7 @@ import { CodeTool, TOOL_SPECS, useCodeTool } from '@renderer/components/CodeToolbar' import { useCodeStyle } from '@renderer/context/CodeStyleProvider' import { useSettings } from '@renderer/hooks/useSettings' -import CodeMirror, { Annotation, BasicSetupOptions, EditorView, Extension, keymap } from '@uiw/react-codemirror' +import CodeMirror, { Annotation, BasicSetupOptions, EditorView, Extension } from '@uiw/react-codemirror' import diff from 'fast-diff' import { ChevronsDownUp, @@ -14,7 +14,7 @@ import { useCallback, useEffect, useMemo, useRef, useState } from 'react' import { memo } from 'react' import { useTranslation } from 'react-i18next' -import { useLanguageExtensions } from './hook' +import { useBlurHandler, useLanguageExtensions, useSaveKeymap } from './hooks' // 标记非用户编辑的变更 const External = Annotation.define() @@ -25,6 +25,7 @@ interface Props { language: string onSave?: (newContent: string) => void onChange?: (newContent: string) => void + onBlur?: (newContent: string) => void setTools?: (value: React.SetStateAction) => void height?: string minHeight?: string @@ -54,6 +55,7 @@ const CodeEditor = ({ language, onSave, onChange, + onBlur, setTools, height, minHeight, @@ -166,28 +168,18 @@ const CodeEditor = ({ setIsUnwrapped(!wrappable) }, [wrappable]) - // 保存功能的快捷键 - const saveKeymap = useMemo(() => { - return keymap.of([ - { - key: 'Mod-s', - run: () => { - handleSave() - return true - }, - preventDefault: true - } - ]) - }, [handleSave]) + const saveKeymapExtension = useSaveKeymap({ onSave, enabled: enableKeymap }) + const blurExtension = useBlurHandler({ onBlur }) const customExtensions = useMemo(() => { return [ ...(extensions ?? []), ...langExtensions, ...(isUnwrapped ? [] : [EditorView.lineWrapping]), - ...(enableKeymap ? [saveKeymap] : []) - ] - }, [extensions, langExtensions, isUnwrapped, enableKeymap, saveKeymap]) + saveKeymapExtension, + blurExtension + ].flat() + }, [extensions, langExtensions, isUnwrapped, saveKeymapExtension, blurExtension]) return ( ( inputEl.focus() inputEl.select() search() - CSS.highlights.clear() - setSearchCompleted(SearchCompletedState.NotSearched) }) } else { requestAnimationFrame(() => { diff --git a/src/renderer/src/components/DraggableList/index.tsx b/src/renderer/src/components/DraggableList/index.tsx new file mode 100644 index 0000000000..de98dd00d5 --- /dev/null +++ b/src/renderer/src/components/DraggableList/index.tsx @@ -0,0 +1,2 @@ +export { default as DraggableList } from './list' +export { default as DraggableVirtualList } from './virtual-list' diff --git a/src/renderer/src/components/DragableList/index.tsx b/src/renderer/src/components/DraggableList/list.tsx similarity index 97% rename from src/renderer/src/components/DragableList/index.tsx rename to src/renderer/src/components/DraggableList/list.tsx index cc281e7b01..0f23a69978 100644 --- a/src/renderer/src/components/DragableList/index.tsx +++ b/src/renderer/src/components/DraggableList/list.tsx @@ -23,7 +23,7 @@ interface Props { droppableProps?: Partial } -const DragableList: FC> = ({ +const DraggableList: FC> = ({ children, list, style, @@ -78,4 +78,4 @@ const DragableList: FC> = ({ ) } -export default DragableList +export default DraggableList diff --git a/src/renderer/src/components/DraggableList/virtual-list.tsx b/src/renderer/src/components/DraggableList/virtual-list.tsx new file mode 100644 index 0000000000..b8e51642e9 --- /dev/null +++ b/src/renderer/src/components/DraggableList/virtual-list.tsx @@ -0,0 +1,212 @@ +import { + DragDropContext, + Draggable, + Droppable, + DroppableProps, + DropResult, + OnDragEndResponder, + OnDragStartResponder, + ResponderProvided +} from '@hello-pangea/dnd' +import Scrollbar from '@renderer/components/Scrollbar' +import { droppableReorder } from '@renderer/utils' +import { useVirtualizer } from '@tanstack/react-virtual' +import { type Key, memo, useCallback, useRef } from 'react' + +/** + * 泛型 Props，用于配置 DraggableVirtualList。 + * + * @template T 列表元素的类型 + * @property {string} [className] 根节点附加 class + * @property {React.CSSProperties} [style] 根节点附加样式 + * @property {React.CSSProperties} [itemStyle] 元素内容区域的附加样式 + * @property {React.CSSProperties} [itemContainerStyle] 元素拖拽容器的附加样式 + * @property {Partial} [droppableProps] 透传给 Droppable 的额外配置 + * @property {(list: T[]) => void} onUpdate 拖拽排序完成后的回调，返回新的列表顺序 + * @property {OnDragStartResponder} [onDragStart] 开始拖拽时的回调 + * @property {OnDragEndResponder} [onDragEnd] 结束拖拽时的回调 + * @property {T[]} list 渲染的数据源 + * @property {(index: number) => Key} [itemKey] 提供给虚拟列表的行 key，若不提供默认使用 index + * @property {number} [overscan=5] 前后额外渲染的行数，提升快速滚动时的体验 + * @property {(item: T, index: number) => React.ReactNode} children 列表项渲染函数 + */ +interface DraggableVirtualListProps { + ref?: React.Ref + className?: string + style?: React.CSSProperties + itemStyle?: React.CSSProperties + itemContainerStyle?: React.CSSProperties + droppableProps?: Partial + onUpdate: (list: T[]) => void + onDragStart?: OnDragStartResponder + onDragEnd?: OnDragEndResponder + list: T[] + itemKey?: (index: number) => Key + overscan?: number + children: (item: T, index: number) => React.ReactNode +} + +/** + * 带虚拟滚动与拖拽排序能力的（垂直）列表组件。 + * - 滚动容器由该组件内部管理。 + * @template T 列表元素的类型 + * @param {DraggableVirtualListProps} props 组件参数 + * @returns {React.ReactElement} + */ +function DraggableVirtualList({ + ref, + className, + style, + itemStyle, + itemContainerStyle, + droppableProps, + onDragStart, + onUpdate, + onDragEnd, + list, + itemKey, + overscan = 5, + children +}: DraggableVirtualListProps): React.ReactElement { + const _onDragEnd = (result: DropResult, provided: ResponderProvided) => { + onDragEnd?.(result, provided) + if (result.destination) { + const sourceIndex = result.source.index + const destIndex = result.destination.index + const reorderAgents = droppableReorder(list, sourceIndex, destIndex) + onUpdate(reorderAgents) + } + } + + // 虚拟列表滚动容器的 ref + const parentRef = useRef(null) + + const virtualizer = useVirtualizer({ + count: list.length, + getScrollElement: useCallback(() => parentRef.current, []), + getItemKey: itemKey, + estimateSize: useCallback(() => 50, []), + overscan + }) + + return ( +
+ + { + const item = list[rubric.source.index] + return ( +
+ {item && children(item, rubric.source.index)} +
+ ) + }} + {...droppableProps}> + {(provided) => { + // 让 dnd 和虚拟列表共享同一个滚动容器 + const setRefs = (el: HTMLDivElement | null) => { + provided.innerRef(el) + parentRef.current = el + } + + return ( + +
+ {virtualizer.getVirtualItems().map((virtualItem) => ( + + ))} +
+ + ) + }} + + +
+ ) +} + +/** + * 渲染单个可拖拽的虚拟列表项，高度为动态测量 + */ +const VirtualRow = memo(({ virtualItem, list, children, itemStyle, itemContainerStyle, virtualizer }: any) => { + const item = list[virtualItem.index] + const draggableId = String(virtualItem.key) + return ( + + {(provided) => { + const setDragRefs = (el: HTMLElement | null) => { + provided.innerRef(el) + virtualizer.measureElement(el) + } + + const dndStyle = provided.draggableProps.style + const virtualizerTransform = `translateY(${virtualItem.start}px)` + + // dnd 的 transform 负责拖拽时的位移和让位动画， + // virtualizer 的 translateY 负责将项定位到虚拟列表的正确位置， + // 它们拼接起来可以同时实现拖拽视觉效果和虚拟化定位。 + const combinedTransform = dndStyle?.transform + ? `${dndStyle.transform} ${virtualizerTransform}` + : virtualizerTransform + + return ( +
+
+ {item && children(item, virtualItem.index)} +
+
+ ) + }} + + ) +}) + +export default DraggableVirtualList diff --git a/src/renderer/src/components/Icons/SVGIcon.tsx b/src/renderer/src/components/Icons/SVGIcon.tsx index d58eab7ee5..988d9657c1 100644 --- a/src/renderer/src/components/Icons/SVGIcon.tsx +++ b/src/renderer/src/components/Icons/SVGIcon.tsx @@ -1,14 +1,25 @@ +import { lightbulbVariants } from '@renderer/utils/motionVariants' +import { motion } from 'framer-motion' import { SVGProps } from 'react' -export const StreamlineGoodHealthAndWellBeing = (props: SVGProps) => { +export const StreamlineGoodHealthAndWellBeing = ( + props: SVGProps & { + size?: number | string + isActive?: boolean + } +) => { + const { size = '1em', isActive, ...svgProps } = props + return ( - - {/* Icon from Streamline by Streamline - https://creativecommons.org/licenses/by/4.0/ */} - - - - - + + + {/* Icon from Streamline by Streamline - https://creativecommons.org/licenses/by/4.0/ */} + + + + + + ) } diff --git a/src/renderer/src/components/Icons/__tests__/CopyIcon.test.tsx b/src/renderer/src/components/Icons/__tests__/CopyIcon.test.tsx new file mode 100644 index 0000000000..e94967a12a --- /dev/null +++ b/src/renderer/src/components/Icons/__tests__/CopyIcon.test.tsx @@ -0,0 +1,15 @@ +import { render } from '@testing-library/react' +import { describe, expect, it, vi } from 'vitest' + +import CopyIcon from '../CopyIcon' + +describe('CopyIcon', () => { + it('should match snapshot with props and className', () => { + const onClick = vi.fn() + const { container } = render( + + ) + + expect(container.firstChild).toMatchSnapshot() + }) +}) diff --git a/src/renderer/src/components/Icons/__tests__/MinAppIcon.test.tsx b/src/renderer/src/components/Icons/__tests__/MinAppIcon.test.tsx new file mode 100644 index 0000000000..a06c19a840 --- /dev/null +++ b/src/renderer/src/components/Icons/__tests__/MinAppIcon.test.tsx @@ -0,0 +1,65 @@ +import { render } from '@testing-library/react' +import { describe, expect, it, vi } from 'vitest' + +import MinAppIcon from '../MinAppIcon' + +vi.mock('@renderer/config/minapps', () => ({ + DEFAULT_MIN_APPS: [ + { + id: 'test-app-1', + name: 'Test App 1', + logo: '/test-logo-1.png', + url: 'https://test1.com', + bodered: true, + background: '#f0f0f0' + }, + { + id: 'test-app-2', + name: 'Test App 2', + logo: '/test-logo-2.png', + url: 'https://test2.com', + bodered: false, + background: undefined + } + ] +})) + +describe('MinAppIcon', () => { + const mockApp = { + id: 'test-app-1', + name: 'Test App', + url: 'https://test.com', + style: { + opacity: 0.8, + transform: 'scale(1.1)' + } + } + + it('should render correctly with various props', () => { + const customStyle = { marginTop: '10px' } + const { container } = render() + + expect(container.firstChild).toMatchSnapshot() + }) + + it('should not apply app.style when sidebar is true', () => { + const { container } = render() + const img = container.querySelector('img') + + expect(img).not.toHaveStyle({ + opacity: '0.8', + transform: 'scale(1.1)' + }) + }) + + it('should return null when app is not found in DEFAULT_MIN_APPS', () => { + const unknownApp = { + id: 'unknown-app', + name: 'Unknown App', + url: 'https://unknown.com' + } + const { container } = render() + + expect(container.firstChild).toBeNull() + }) +}) diff --git a/src/renderer/src/components/Icons/__tests__/__snapshots__/CopyIcon.test.tsx.snap b/src/renderer/src/components/Icons/__tests__/__snapshots__/CopyIcon.test.tsx.snap new file mode 100644 index 0000000000..d333ca4578 --- /dev/null +++ b/src/renderer/src/components/Icons/__tests__/__snapshots__/CopyIcon.test.tsx.snap @@ -0,0 +1,9 @@ +// Vitest Snapshot v1, https://vitest.dev/guide/snapshot.html + +exports[`CopyIcon > should match snapshot with props and className 1`] = ` + +`; diff --git a/src/renderer/src/components/Icons/__tests__/__snapshots__/MinAppIcon.test.tsx.snap b/src/renderer/src/components/Icons/__tests__/__snapshots__/MinAppIcon.test.tsx.snap new file mode 100644 index 0000000000..e41515fed6 --- /dev/null +++ b/src/renderer/src/components/Icons/__tests__/__snapshots__/MinAppIcon.test.tsx.snap @@ -0,0 +1,15 @@ +// Vitest Snapshot v1, https://vitest.dev/guide/snapshot.html + +exports[`MinAppIcon > should render correctly with various props 1`] = ` +.c0 { + border-radius: 16px; + user-select: none; + -webkit-user-drag: none; +} + + +`; diff --git a/src/renderer/src/components/LocalBackupManager.tsx b/src/renderer/src/components/LocalBackupManager.tsx new file mode 100644 index 0000000000..2b4f949b38 --- /dev/null +++ b/src/renderer/src/components/LocalBackupManager.tsx @@ -0,0 +1,255 @@ +import { DeleteOutlined, ExclamationCircleOutlined, ReloadOutlined } from '@ant-design/icons' +import { restoreFromLocalBackup } from '@renderer/services/BackupService' +import { formatFileSize } from '@renderer/utils' +import { Button, message, Modal, Table, Tooltip } from 'antd' +import dayjs from 'dayjs' +import { useCallback, useEffect, useState } from 'react' +import { useTranslation } from 'react-i18next' + +interface BackupFile { + fileName: string + modifiedTime: string + size: number +} + +interface LocalBackupManagerProps { + visible: boolean + onClose: () => void + localBackupDir?: string + restoreMethod?: (fileName: string) => Promise +} + +export function LocalBackupManager({ visible, onClose, localBackupDir, restoreMethod }: LocalBackupManagerProps) { + const { t } = useTranslation() + const [backupFiles, setBackupFiles] = useState([]) + const [loading, setLoading] = useState(false) + const [selectedRowKeys, setSelectedRowKeys] = useState([]) + const [deleting, setDeleting] = useState(false) + const [restoring, setRestoring] = useState(false) + const [pagination, setPagination] = useState({ + current: 1, + pageSize: 5, + total: 0 + }) + + const fetchBackupFiles = useCallback(async () => { + if (!localBackupDir) { + return + } + + setLoading(true) + try { + const files = await window.api.backup.listLocalBackupFiles(localBackupDir) + setBackupFiles(files) + setPagination((prev) => ({ + ...prev, + total: files.length + })) + } catch (error: any) { + message.error(`${t('settings.data.local.backup.manager.fetch.error')}: ${error.message}`) + } finally { + setLoading(false) + } + }, [localBackupDir, t]) + + useEffect(() => { + if (visible) { + fetchBackupFiles() + setSelectedRowKeys([]) + setPagination((prev) => ({ + ...prev, + current: 1 + })) + } + }, [visible, fetchBackupFiles]) + + const handleTableChange = (pagination: any) => { + setPagination(pagination) + } + + const handleDeleteSelected = async () => { + if (selectedRowKeys.length === 0) { + message.warning(t('settings.data.local.backup.manager.select.files.delete')) + return + } + + if (!localBackupDir) { + return + } + + window.modal.confirm({ + title: t('settings.data.local.backup.manager.delete.confirm.title'), + icon: , + content: t('settings.data.local.backup.manager.delete.confirm.multiple', { count: selectedRowKeys.length }), + okText: t('common.confirm'), + cancelText: t('common.cancel'), + centered: true, + onOk: async () => { + setDeleting(true) + try { + // Delete selected files one by one + for (const key of selectedRowKeys) { + await window.api.backup.deleteLocalBackupFile(key.toString(), localBackupDir) + } + message.success( + t('settings.data.local.backup.manager.delete.success.multiple', { count: selectedRowKeys.length }) + ) + setSelectedRowKeys([]) + await fetchBackupFiles() + } catch (error: any) { + message.error(`${t('settings.data.local.backup.manager.delete.error')}: ${error.message}`) + } finally { + setDeleting(false) + } + } + }) + } + + const handleDeleteSingle = async (fileName: string) => { + if (!localBackupDir) { + return + } + + window.modal.confirm({ + title: t('settings.data.local.backup.manager.delete.confirm.title'), + icon: , + content: t('settings.data.local.backup.manager.delete.confirm.single', { fileName }), + okText: t('common.confirm'), + cancelText: t('common.cancel'), + centered: true, + onOk: async () => { + setDeleting(true) + try { + await window.api.backup.deleteLocalBackupFile(fileName, localBackupDir) + message.success(t('settings.data.local.backup.manager.delete.success.single')) + await fetchBackupFiles() + } catch (error: any) { + message.error(`${t('settings.data.local.backup.manager.delete.error')}: ${error.message}`) + } finally { + setDeleting(false) + } + } + }) + } + + const handleRestore = async (fileName: string) => { + if (!localBackupDir) { + return + } + + window.modal.confirm({ + title: t('settings.data.local.restore.confirm.title'), + icon: , + content: t('settings.data.local.restore.confirm.content'), + okText: t('common.confirm'), + cancelText: t('common.cancel'), + centered: true, + onOk: async () => { + setRestoring(true) + try { + await (restoreMethod || restoreFromLocalBackup)(fileName) + message.success(t('settings.data.local.backup.manager.restore.success')) + onClose() // Close the modal + } catch (error: any) { + message.error(`${t('settings.data.local.backup.manager.restore.error')}: ${error.message}`) + } finally { + setRestoring(false) + } + } + }) + } + + const columns = [ + { + title: t('settings.data.local.backup.manager.columns.fileName'), + dataIndex: 'fileName', + key: 'fileName', + ellipsis: { + showTitle: false + }, + render: (fileName: string) => ( + + {fileName} + + ) + }, + { + title: t('settings.data.local.backup.manager.columns.modifiedTime'), + dataIndex: 'modifiedTime', + key: 'modifiedTime', + width: 180, + render: (time: string) => dayjs(time).format('YYYY-MM-DD HH:mm:ss') + }, + { + title: t('settings.data.local.backup.manager.columns.size'), + dataIndex: 'size', + key: 'size', + width: 120, + render: (size: number) => formatFileSize(size) + }, + { + title: t('settings.data.local.backup.manager.columns.actions'), + key: 'action', + width: 160, + render: (_: any, record: BackupFile) => ( + <> + + + + ) + } + ] + + const rowSelection = { + selectedRowKeys, + onChange: (selectedRowKeys: React.Key[]) => { + setSelectedRowKeys(selectedRowKeys) + } + } + + return ( + } onClick={fetchBackupFiles} disabled={loading}> + {t('settings.data.local.backup.manager.refresh')} + , + , + + ]}> + + + ) +} diff --git a/src/renderer/src/components/LocalBackupModals.tsx b/src/renderer/src/components/LocalBackupModals.tsx new file mode 100644 index 0000000000..9f2a700dd2 --- /dev/null +++ b/src/renderer/src/components/LocalBackupModals.tsx @@ -0,0 +1,98 @@ +import { backupToLocalDir } from '@renderer/services/BackupService' +import { Button, Input, Modal } from 'antd' +import dayjs from 'dayjs' +import { useCallback, useState } from 'react' +import { useTranslation } from 'react-i18next' + +interface LocalBackupModalProps { + isModalVisible: boolean + handleBackup: () => void + handleCancel: () => void + backuping: boolean + customFileName: string + setCustomFileName: (value: string) => void +} + +export function LocalBackupModal({ + isModalVisible, + handleBackup, + handleCancel, + backuping, + customFileName, + setCustomFileName +}: LocalBackupModalProps) { + const { t } = useTranslation() + + return ( + + {t('common.cancel')} + , + + ]}> + setCustomFileName(e.target.value)} + placeholder={t('settings.data.local.backup.modal.filename.placeholder')} + /> + + ) +} + +// Hook for backup modal +export function useLocalBackupModal(localBackupDir: string | undefined) { + const [isModalVisible, setIsModalVisible] = useState(false) + const [backuping, setBackuping] = useState(false) + const [customFileName, setCustomFileName] = useState('') + + const handleCancel = () => { + setIsModalVisible(false) + } + + const showBackupModal = useCallback(async () => { + // 获取默认文件名 + const deviceType = await window.api.system.getDeviceType() + const hostname = await window.api.system.getHostname() + const timestamp = dayjs().format('YYYYMMDDHHmmss') + const defaultFileName = `cherry-studio.${timestamp}.${hostname}.${deviceType}.zip` + setCustomFileName(defaultFileName) + setIsModalVisible(true) + }, []) + + const handleBackup = async () => { + if (!localBackupDir) { + setIsModalVisible(false) + return + } + + setBackuping(true) + try { + await backupToLocalDir({ + showMessage: true, + customFileName + }) + setIsModalVisible(false) + } catch (error) { + console.error('[LocalBackupModal] Backup failed:', error) + } finally { + setBackuping(false) + } + } + + return { + isModalVisible, + handleBackup, + handleCancel, + backuping, + customFileName, + setCustomFileName, + showBackupModal + } +} diff --git a/src/renderer/src/components/MarkdownEditor/README.md b/src/renderer/src/components/MarkdownEditor/README.md deleted file mode 100644 index 0519ecba6e..0000000000 --- a/src/renderer/src/components/MarkdownEditor/README.md +++ /dev/null @@ -1 +0,0 @@ - \ No newline at end of file diff --git a/src/renderer/src/components/Popups/ApiKeyListPopup/hook.ts b/src/renderer/src/components/Popups/ApiKeyListPopup/hook.ts new file mode 100644 index 0000000000..5bd9072d12 --- /dev/null +++ b/src/renderer/src/components/Popups/ApiKeyListPopup/hook.ts @@ -0,0 +1,307 @@ +import Logger from '@renderer/config/logger' +import { isEmbeddingModel, isRerankModel } from '@renderer/config/models' +import SelectProviderModelPopup from '@renderer/pages/settings/ProviderSettings/SelectProviderModelPopup' +import { checkApi } from '@renderer/services/ApiService' +import WebSearchService from '@renderer/services/WebSearchService' +import { Model, PreprocessProvider, Provider, WebSearchProvider } from '@renderer/types' +import { formatApiKeys, splitApiKeyString } from '@renderer/utils/api' +import { formatErrorMessage } from '@renderer/utils/error' +import { TFunction } from 'i18next' +import { isEmpty } from 'lodash' +import { useCallback, useMemo, useState } from 'react' +import { useTranslation } from 'react-i18next' + +import { ApiKeyConnectivity, ApiKeyValidity, ApiKeyWithStatus, ApiProviderKind, ApiProviderUnion } from './types' + +interface UseApiKeysProps { + provider: ApiProviderUnion + updateProvider: (provider: Partial) => void + providerKind: ApiProviderKind +} + +/** + * API Keys 管理 hook + */ +export function useApiKeys({ provider, updateProvider, providerKind }: UseApiKeysProps) { + const { t } = useTranslation() + + // 连通性检查的 UI 状态管理 + const [connectivityStates, setConnectivityStates] = useState>(new Map()) + + // 保存 apiKey 到 provider + const updateProviderWithKey = useCallback( + (newKeys: string[]) => { + const validKeys = newKeys.filter((k) => k.trim()) + const formattedKeyString = formatApiKeys(validKeys.join(',')) + updateProvider({ apiKey: formattedKeyString }) + }, + [updateProvider] + ) + + // 解析 keyString 为数组 + const keys = useMemo(() => { + if (!provider.apiKey) return [] + const formattedApiKeys = formatApiKeys(provider.apiKey) + const keys = splitApiKeyString(formattedApiKeys) + return Array.from(new Set(keys)) + }, [provider.apiKey]) + + // 合并基本数据和连通性状态 + const keysWithStatus = useMemo((): ApiKeyWithStatus[] => { + return keys.map((key) => { + const connectivityState = connectivityStates.get(key) || { + status: 'not_checked' as const, + checking: false, + error: undefined, + model: undefined, + latency: undefined + } + return { + key, + ...connectivityState + } + }) + }, [keys, connectivityStates]) + + // 更新单个 key 的连通性状态 + const updateConnectivityState = useCallback((key: string, state: Partial) => { + setConnectivityStates((prev) => { + const newMap = new Map(prev) + const currentState = prev.get(key) || { + status: 'not_checked' as const, + checking: false, + error: undefined, + model: undefined, + latency: undefined + } + newMap.set(key, { ...currentState, ...state }) + return newMap + }) + }, []) + + // 验证 API key 格式 + const validateApiKey = useCallback( + (key: string, existingKeys: string[] = []): ApiKeyValidity => { + const trimmedKey = key.trim() + + if (!trimmedKey) { + return { isValid: false, error: t('settings.provider.api.key.error.empty') } + } + + if (existingKeys.includes(trimmedKey)) { + return { isValid: false, error: t('settings.provider.api.key.error.duplicate') } + } + + return { isValid: true } + }, + [t] + ) + + // 添加新 key + const addKey = useCallback( + (key: string): ApiKeyValidity => { + const validation = validateApiKey(key, keys) + + if (!validation.isValid) { + return validation + } + + updateProviderWithKey([...keys, key.trim()]) + return { isValid: true } + }, + [validateApiKey, keys, updateProviderWithKey] + ) + + // 更新 key + const updateKey = useCallback( + (index: number, key: string): ApiKeyValidity => { + if (index < 0 || index >= keys.length) { + Logger.error('[ApiKeyList] invalid key index', { index }) + return { isValid: false, error: 'Invalid index' } + } + + const otherKeys = keys.filter((_, i) => i !== index) + const validation = validateApiKey(key, otherKeys) + + if (!validation.isValid) { + return validation + } + + // 清除旧 key 的连通性状态 + const oldKey = keys[index] + if (oldKey !== key.trim()) { + setConnectivityStates((prev) => { + const newMap = new Map(prev) + newMap.delete(oldKey) + return newMap + }) + } + + const newKeys = [...keys] + newKeys[index] = key.trim() + updateProviderWithKey(newKeys) + + return { isValid: true } + }, + [keys, validateApiKey, updateProviderWithKey] + ) + + // 移除 key + const removeKey = useCallback( + (index: number) => { + if (index < 0 || index >= keys.length) return + + const keyToRemove = keys[index] + const newKeys = keys.filter((_, i) => i !== index) + + // 清除对应的连通性状态 + setConnectivityStates((prev) => { + const newMap = new Map(prev) + newMap.delete(keyToRemove) + return newMap + }) + + updateProviderWithKey(newKeys) + }, + [keys, updateProviderWithKey] + ) + + // 移除连通性检查失败的 keys + const removeInvalidKeys = useCallback(() => { + const validKeys = keysWithStatus.filter((keyStatus) => keyStatus.status !== 'error').map((k) => k.key) + + // 清除被删除的 keys 的连通性状态 + const keysToRemove = keysWithStatus.filter((keyStatus) => keyStatus.status === 'error').map((k) => k.key) + + setConnectivityStates((prev) => { + const newMap = new Map(prev) + keysToRemove.forEach((key) => newMap.delete(key)) + return newMap + }) + + updateProviderWithKey(validKeys) + }, [keysWithStatus, updateProviderWithKey]) + + // 检查单个 key 的连通性，不负责选择和验证模型 + const runConnectivityCheck = useCallback( + async (index: number, model?: Model): Promise => { + const keyToCheck = keys[index] + const currentState = connectivityStates.get(keyToCheck) + if (currentState?.checking) return + + // 设置检查状态 + updateConnectivityState(keyToCheck, { checking: true }) + + try { + const startTime = Date.now() + if (isLlmProvider(provider, providerKind) && model) { + await checkApi({ ...provider, apiKey: keyToCheck }, model) + } else { + const result = await WebSearchService.checkSearch({ ...provider, apiKey: keyToCheck }) + if (!result.valid) throw new Error(result.error) + } + const latency = Date.now() - startTime + + // 连通性检查成功 + updateConnectivityState(keyToCheck, { + checking: false, + status: 'success', + model, + latency, + error: undefined + }) + } catch (error: any) { + // 连通性检查失败 + updateConnectivityState(keyToCheck, { + checking: false, + status: 'error', + error: formatErrorMessage(error), + model: undefined, + latency: undefined + }) + + Logger.error('[ApiKeyList] failed to validate the connectivity of the api key', error) + } + }, + [keys, connectivityStates, updateConnectivityState, provider, providerKind] + ) + + // 检查单个 key 的连通性 + const checkKeyConnectivity = useCallback( + async (index: number): Promise => { + if (!provider || index < 0 || index >= keys.length) return + + const keyToCheck = keys[index] + const currentState = connectivityStates.get(keyToCheck) + if (currentState?.checking) return + + const model = isLlmProvider(provider, providerKind) ? await getModelForCheck(provider, t) : undefined + if (model === null) return + + await runConnectivityCheck(index, model) + }, + [provider, keys, connectivityStates, providerKind, t, runConnectivityCheck] + ) + + // 检查所有 keys 的连通性 + const checkAllKeysConnectivity = useCallback(async () => { + if (!provider || keys.length === 0) return + + const model = isLlmProvider(provider, providerKind) ? await getModelForCheck(provider, t) : undefined + if (model === null) return + + await Promise.allSettled(keys.map((_, index) => runConnectivityCheck(index, model))) + }, [provider, keys, providerKind, t, runConnectivityCheck]) + + // 计算是否有 key 正在检查 + const isChecking = useMemo(() => { + return Array.from(connectivityStates.values()).some((state) => state.checking) + }, [connectivityStates]) + + return { + keys: keysWithStatus, + addKey, + updateKey, + removeKey, + removeInvalidKeys, + checkKeyConnectivity, + checkAllKeysConnectivity, + isChecking + } +} + +export function isLlmProvider(obj: any, kind: ApiProviderKind): obj is Provider { + return kind === 'llm' && 'type' in obj && 'models' in obj +} + +export function isWebSearchProvider(obj: any, kind: ApiProviderKind): obj is WebSearchProvider { + return kind === 'websearch' && ('url' in obj || 'engines' in obj) +} + +export function isPreprocessProvider(obj: any, kind: ApiProviderKind): obj is PreprocessProvider { + return kind === 'doc-preprocess' && ('quota' in obj || 'options' in obj) +} + +// 获取模型用于检查 +async function getModelForCheck(provider: Provider, t: TFunction): Promise { + const modelsToCheck = provider.models.filter((model) => !isEmbeddingModel(model) && !isRerankModel(model)) + + if (isEmpty(modelsToCheck)) { + window.message.error({ + key: 'no-models', + style: { marginTop: '3vh' }, + duration: 5, + content: t('settings.provider.no_models_for_check') + }) + return null + } + + try { + const selectedModel = await SelectProviderModelPopup.show({ provider }) + if (!selectedModel) return null + return selectedModel + } catch (error) { + Logger.error('[ApiKeyList] failed to select model', error) + return null + } +} diff --git a/src/renderer/src/components/Popups/ApiKeyListPopup/index.ts b/src/renderer/src/components/Popups/ApiKeyListPopup/index.ts new file mode 100644 index 0000000000..b599832456 --- /dev/null +++ b/src/renderer/src/components/Popups/ApiKeyListPopup/index.ts @@ -0,0 +1,2 @@ +export { default as ApiKeyListPopup } from './popup' +export * from './types' diff --git a/src/renderer/src/components/Popups/ApiKeyListPopup/item.tsx b/src/renderer/src/components/Popups/ApiKeyListPopup/item.tsx new file mode 100644 index 0000000000..cf830897cb --- /dev/null +++ b/src/renderer/src/components/Popups/ApiKeyListPopup/item.tsx @@ -0,0 +1,213 @@ +import { CheckCircleFilled, CloseCircleFilled, MinusOutlined } from '@ant-design/icons' +import { StreamlineGoodHealthAndWellBeing } from '@renderer/components/Icons/SVGIcon' +import { maskApiKey } from '@renderer/utils/api' +import { Button, Flex, Input, InputRef, List, Popconfirm, Tooltip, Typography } from 'antd' +import { Check, PenLine, X } from 'lucide-react' +import { FC, memo, useEffect, useRef, useState } from 'react' +import { useTranslation } from 'react-i18next' +import styled from 'styled-components' + +import { ApiKeyValidity, ApiKeyWithStatus } from './types' + +export interface ApiKeyItemProps { + keyStatus: ApiKeyWithStatus + onUpdate: (newKey: string) => ApiKeyValidity + onRemove: () => void + onCheck: () => Promise + disabled?: boolean + showHealthCheck?: boolean + isNew?: boolean +} + +/** + * API Key 项组件 + * 支持编辑、删除、连接检查等操作 + */ +const ApiKeyItem: FC = ({ + keyStatus, + onUpdate, + onRemove, + onCheck, + disabled: _disabled = false, + showHealthCheck = true, + isNew = false +}) => { + const { t } = useTranslation() + const [isEditing, setIsEditing] = useState(isNew || !keyStatus.key.trim()) + const [editValue, setEditValue] = useState(keyStatus.key) + const [hasUnsavedChanges, setHasUnsavedChanges] = useState(false) + const inputRef = useRef(null) + + const disabled = keyStatus.checking || _disabled + const isNotChecked = keyStatus.status === 'not_checked' + const isSuccess = keyStatus.status === 'success' + const statusColor = isSuccess ? 'var(--color-status-success)' : 'var(--color-status-error)' + + useEffect(() => { + if (isEditing && inputRef.current) { + inputRef.current.focus() + } + }, [isEditing]) + + useEffect(() => { + setHasUnsavedChanges(editValue.trim() !== keyStatus.key.trim()) + }, [editValue, keyStatus.key]) + + const handleEdit = () => { + if (disabled) return + setIsEditing(true) + setEditValue(keyStatus.key) + } + + const handleSave = () => { + const result = onUpdate(editValue) + if (!result.isValid) { + window.message.warning({ + key: 'api-key-error', + content: result.error + }) + return + } + + setIsEditing(false) + } + + const handleCancelEdit = () => { + if (isNew || !keyStatus.key.trim()) { + // 临时项取消时直接移除 + onRemove() + } else { + // 现有项取消时恢复原值 + setEditValue(keyStatus.key) + setIsEditing(false) + } + } + + const renderStatusIcon = () => { + if (keyStatus.checking || isNotChecked) return null + + const StatusIcon = isSuccess ? CheckCircleFilled : CloseCircleFilled + return + } + + const renderKeyCheckResultTooltip = () => { + if (keyStatus.checking) { + return t('settings.models.check.checking') + } + + if (isNotChecked) { + return '' + } + + const statusTitle = isSuccess ? t('settings.models.check.passed') : t('settings.models.check.failed') + + return ( +
+ {statusTitle} + {keyStatus.model && ( +
+ {t('common.model')}: {keyStatus.model.name} +
+ )} + {keyStatus.latency && isSuccess && ( +
+ {t('settings.provider.api.key.check.latency')}: {(keyStatus.latency / 1000).toFixed(2)}s +
+ )} + {keyStatus.error &&
{keyStatus.error}
} +
+ ) + } + + return ( + + {isEditing ? ( + + setEditValue(e.target.value)} + onPressEnter={handleSave} + placeholder={t('settings.provider.api.key.new_key.placeholder')} + style={{ flex: 1, fontSize: '14px', marginLeft: '-10px' }} + spellCheck={false} + disabled={disabled} + /> + + + + + + + ) +} + +interface SpecificApiKeyListProps { + providerId: string + providerKind: ApiProviderKind + showHealthCheck?: boolean +} + +export const LlmApiKeyList: FC = ({ providerId, providerKind, showHealthCheck = true }) => { + const { provider, updateProvider } = useProvider(providerId) + + return ( + + ) +} + +export const WebSearchApiKeyList: FC = ({ + providerId, + providerKind, + showHealthCheck = true +}) => { + const { provider, updateProvider } = useWebSearchProvider(providerId) + + return ( + + ) +} + +export const DocPreprocessApiKeyList: FC = ({ + providerId, + providerKind, + showHealthCheck = true +}) => { + const { provider, updateProvider } = usePreprocessProvider(providerId) + + return ( + + ) +} diff --git a/src/renderer/src/components/Popups/ApiKeyListPopup/popup.tsx b/src/renderer/src/components/Popups/ApiKeyListPopup/popup.tsx new file mode 100644 index 0000000000..096e00ca58 --- /dev/null +++ b/src/renderer/src/components/Popups/ApiKeyListPopup/popup.tsx @@ -0,0 +1,88 @@ +import { TopView } from '@renderer/components/TopView' +import { Modal } from 'antd' +import { useMemo, useState } from 'react' +import { useTranslation } from 'react-i18next' + +import { DocPreprocessApiKeyList, LlmApiKeyList, WebSearchApiKeyList } from './list' +import { ApiProviderKind } from './types' + +interface ShowParams { + providerId: string + providerKind: ApiProviderKind + title?: string + showHealthCheck?: boolean +} + +interface Props extends ShowParams { + resolve: (value: any) => void +} + +/** + * API Key 列表弹窗容器组件 + */ +const PopupContainer: React.FC = ({ providerId, providerKind, title, resolve, showHealthCheck = true }) => { + const [open, setOpen] = useState(true) + const { t } = useTranslation() + + const onCancel = () => { + setOpen(false) + } + + const onClose = () => { + resolve(null) + } + + const ListComponent = useMemo(() => { + switch (providerKind) { + case 'llm': + return LlmApiKeyList + case 'websearch': + return WebSearchApiKeyList + case 'doc-preprocess': + return DocPreprocessApiKeyList + default: + return null + } + }, [providerKind]) + + return ( + + {ListComponent && ( + + )} + + ) +} + +const TopViewKey = 'ApiKeyListPopup' + +export default class ApiKeyListPopup { + static topviewId = 0 + + static hide() { + TopView.hide(TopViewKey) + } + + static show(props: ShowParams) { + return new Promise((resolve) => { + TopView.show( + { + resolve(v) + TopView.hide(TopViewKey) + }} + />, + TopViewKey + ) + }) + } +} diff --git a/src/renderer/src/components/Popups/ApiKeyListPopup/types.ts b/src/renderer/src/components/Popups/ApiKeyListPopup/types.ts new file mode 100644 index 0000000000..f5ed1c62d1 --- /dev/null +++ b/src/renderer/src/components/Popups/ApiKeyListPopup/types.ts @@ -0,0 +1,31 @@ +import { Model, PreprocessProvider, Provider, WebSearchProvider } from '@renderer/types' + +/** + * API Key 连通性检查的状态 + */ +export type ApiKeyConnectivity = { + status: 'success' | 'error' | 'not_checked' + checking?: boolean + error?: string + model?: Model + latency?: number +} + +/** + * API key 及其连通性检查的状态 + */ +export type ApiKeyWithStatus = { + key: string +} & ApiKeyConnectivity + +/** + * API key 格式有效性 + */ +export type ApiKeyValidity = { + isValid: boolean + error?: string +} + +export type ApiProviderUnion = Provider | WebSearchProvider | PreprocessProvider + +export type ApiProviderKind = 'llm' | 'websearch' | 'doc-preprocess' diff --git a/src/renderer/src/components/Popups/TextEditPopup.tsx b/src/renderer/src/components/Popups/TextEditPopup.tsx index 46bca109fc..ab7bf40cb6 100644 --- a/src/renderer/src/components/Popups/TextEditPopup.tsx +++ b/src/renderer/src/components/Popups/TextEditPopup.tsx @@ -3,6 +3,7 @@ import { useDefaultModel } from '@renderer/hooks/useAssistant' import { useSettings } from '@renderer/hooks/useSettings' import { fetchTranslate } from '@renderer/services/ApiService' import { getDefaultTranslateAssistant } from '@renderer/services/AssistantService' +import { getLanguageByLangcode } from '@renderer/utils/translate' import { Modal, ModalProps } from 'antd' import TextArea from 'antd/es/input/TextArea' import { TextAreaProps } from 'antd/lib/input' @@ -111,7 +112,7 @@ const PopupContainer: React.FC = ({ } try { - const assistant = getDefaultTranslateAssistant(targetLanguage, textValue) + const assistant = getDefaultTranslateAssistant(getLanguageByLangcode(targetLanguage), textValue) const translatedText = await fetchTranslate({ content: textValue, assistant }) if (isMounted.current) { setTextValue(translatedText) diff --git a/src/renderer/src/components/S3BackupManager.tsx b/src/renderer/src/components/S3BackupManager.tsx new file mode 100644 index 0000000000..f644d2dce6 --- /dev/null +++ b/src/renderer/src/components/S3BackupManager.tsx @@ -0,0 +1,295 @@ +import { DeleteOutlined, ExclamationCircleOutlined, ReloadOutlined } from '@ant-design/icons' +import { restoreFromS3 } from '@renderer/services/BackupService' +import type { S3Config } from '@renderer/types' +import { formatFileSize } from '@renderer/utils' +import { Button, Modal, Table, Tooltip } from 'antd' +import dayjs from 'dayjs' +import { useCallback, useEffect, useState } from 'react' +import { useTranslation } from 'react-i18next' + +interface BackupFile { + fileName: string + modifiedTime: string + size: number +} + +interface S3BackupManagerProps { + visible: boolean + onClose: () => void + s3Config: Partial + restoreMethod?: (fileName: string) => Promise +} + +export function S3BackupManager({ visible, onClose, s3Config, restoreMethod }: S3BackupManagerProps) { + const [backupFiles, setBackupFiles] = useState([]) + const [loading, setLoading] = useState(false) + const [selectedRowKeys, setSelectedRowKeys] = useState([]) + const [deleting, setDeleting] = useState(false) + const [restoring, setRestoring] = useState(false) + const [pagination, setPagination] = useState({ + current: 1, + pageSize: 5, + total: 0 + }) + const { t } = useTranslation() + + const { endpoint, region, bucket, accessKeyId, secretAccessKey } = s3Config + + const fetchBackupFiles = useCallback(async () => { + if (!endpoint || !region || !bucket || !accessKeyId || !secretAccessKey) { + window.message.error(t('settings.data.s3.manager.config.incomplete')) + return + } + + setLoading(true) + try { + const files = await window.api.backup.listS3Files({ + ...s3Config, + endpoint, + region, + bucket, + accessKeyId, + secretAccessKey, + skipBackupFile: false, + autoSync: false, + syncInterval: 0, + maxBackups: 0 + }) + setBackupFiles(files) + setPagination((prev) => ({ + ...prev, + total: files.length + })) + } catch (error: any) { + window.message.error(t('settings.data.s3.manager.files.fetch.error', { message: error.message })) + } finally { + setLoading(false) + } + }, [endpoint, region, bucket, accessKeyId, secretAccessKey, t, s3Config]) + + useEffect(() => { + if (visible) { + fetchBackupFiles() + setSelectedRowKeys([]) + setPagination((prev) => ({ + ...prev, + current: 1 + })) + } + }, [visible, fetchBackupFiles]) + + const handleTableChange = (pagination: any) => { + setPagination(pagination) + } + + const handleDeleteSelected = async () => { + if (selectedRowKeys.length === 0) { + window.message.warning(t('settings.data.s3.manager.select.warning')) + return + } + + if (!endpoint || !region || !bucket || !accessKeyId || !secretAccessKey) { + window.message.error(t('settings.data.s3.manager.config.incomplete')) + return + } + + window.modal.confirm({ + title: t('settings.data.s3.manager.delete.confirm.title'), + icon: , + content: t('settings.data.s3.manager.delete.confirm.multiple', { count: selectedRowKeys.length }), + okText: t('settings.data.s3.manager.delete.confirm.title'), + cancelText: t('common.cancel'), + centered: true, + onOk: async () => { + setDeleting(true) + try { + // 依次删除选中的文件 + for (const key of selectedRowKeys) { + await window.api.backup.deleteS3File(key.toString(), { + ...s3Config, + endpoint, + region, + bucket, + accessKeyId, + secretAccessKey, + skipBackupFile: false, + autoSync: false, + syncInterval: 0, + maxBackups: 0 + }) + } + window.message.success( + t('settings.data.s3.manager.delete.success.multiple', { count: selectedRowKeys.length }) + ) + setSelectedRowKeys([]) + await fetchBackupFiles() + } catch (error: any) { + window.message.error(t('settings.data.s3.manager.delete.error', { message: error.message })) + } finally { + setDeleting(false) + } + } + }) + } + + const handleDeleteSingle = async (fileName: string) => { + if (!endpoint || !region || !bucket || !accessKeyId || !secretAccessKey) { + window.message.error(t('settings.data.s3.manager.config.incomplete')) + return + } + + window.modal.confirm({ + title: t('settings.data.s3.manager.delete.confirm.title'), + icon: , + content: t('settings.data.s3.manager.delete.confirm.single', { fileName }), + okText: t('settings.data.s3.manager.delete.confirm.title'), + cancelText: t('common.cancel'), + centered: true, + onOk: async () => { + setDeleting(true) + try { + await window.api.backup.deleteS3File(fileName, { + ...s3Config, + endpoint, + region, + bucket, + accessKeyId, + secretAccessKey, + skipBackupFile: false, + autoSync: false, + syncInterval: 0, + maxBackups: 0 + }) + window.message.success(t('settings.data.s3.manager.delete.success.single')) + await fetchBackupFiles() + } catch (error: any) { + window.message.error(t('settings.data.s3.manager.delete.error', { message: error.message })) + } finally { + setDeleting(false) + } + } + }) + } + + const handleRestore = async (fileName: string) => { + if (!endpoint || !region || !bucket || !accessKeyId || !secretAccessKey) { + window.message.error(t('settings.data.s3.manager.config.incomplete')) + return + } + + window.modal.confirm({ + title: t('settings.data.s3.restore.confirm.title'), + icon: , + content: t('settings.data.s3.restore.confirm.content'), + okText: t('settings.data.s3.restore.confirm.ok'), + cancelText: t('settings.data.s3.restore.confirm.cancel'), + centered: true, + onOk: async () => { + setRestoring(true) + try { + await (restoreMethod || restoreFromS3)(fileName) + window.message.success(t('settings.data.s3.restore.success')) + onClose() // 关闭模态框 + } catch (error: any) { + window.message.error(t('settings.data.s3.restore.error', { message: error.message })) + } finally { + setRestoring(false) + } + } + }) + } + + const columns = [ + { + title: t('settings.data.s3.manager.columns.fileName'), + dataIndex: 'fileName', + key: 'fileName', + ellipsis: { + showTitle: false + }, + render: (fileName: string) => ( + + {fileName} + + ) + }, + { + title: t('settings.data.s3.manager.columns.modifiedTime'), + dataIndex: 'modifiedTime', + key: 'modifiedTime', + width: 180, + render: (time: string) => dayjs(time).format('YYYY-MM-DD HH:mm:ss') + }, + { + title: t('settings.data.s3.manager.columns.size'), + dataIndex: 'size', + key: 'size', + width: 120, + render: (size: number) => formatFileSize(size) + }, + { + title: t('settings.data.s3.manager.columns.actions'), + key: 'action', + width: 160, + render: (_: any, record: BackupFile) => ( + <> + + + + ) + } + ] + + const rowSelection = { + selectedRowKeys, + onChange: (selectedRowKeys: React.Key[]) => { + setSelectedRowKeys(selectedRowKeys) + } + } + + return ( + } onClick={fetchBackupFiles} disabled={loading}> + {t('settings.data.s3.manager.refresh')} + , + , + + ]}> +
+ + ) +} diff --git a/src/renderer/src/components/S3Modals.tsx b/src/renderer/src/components/S3Modals.tsx new file mode 100644 index 0000000000..75c8b31b3a --- /dev/null +++ b/src/renderer/src/components/S3Modals.tsx @@ -0,0 +1,265 @@ +import { backupToS3 } from '@renderer/services/BackupService' +import { formatFileSize } from '@renderer/utils' +import { Input, Modal, Select, Spin } from 'antd' +import dayjs from 'dayjs' +import { useCallback, useState } from 'react' +import { useTranslation } from 'react-i18next' + +interface BackupFile { + fileName: string + modifiedTime: string + size: number +} + +export function useS3BackupModal() { + const [customFileName, setCustomFileName] = useState('') + const [isModalVisible, setIsModalVisible] = useState(false) + const [backuping, setBackuping] = useState(false) + + const handleBackup = async () => { + setBackuping(true) + try { + await backupToS3({ customFileName, showMessage: true }) + } finally { + setBackuping(false) + setIsModalVisible(false) + } + } + + const handleCancel = () => { + setIsModalVisible(false) + } + + const showBackupModal = useCallback(async () => { + // 获取默认文件名 + const deviceType = await window.api.system.getDeviceType() + const hostname = await window.api.system.getHostname() + const timestamp = dayjs().format('YYYYMMDDHHmmss') + const defaultFileName = `cherry-studio.${timestamp}.${hostname}.${deviceType}.zip` + setCustomFileName(defaultFileName) + setIsModalVisible(true) + }, []) + + return { + isModalVisible, + handleBackup, + handleCancel, + backuping, + customFileName, + setCustomFileName, + showBackupModal + } +} + +type S3BackupModalProps = { + isModalVisible: boolean + handleBackup: () => Promise + handleCancel: () => void + backuping: boolean + customFileName: string + setCustomFileName: (value: string) => void +} + +export function S3BackupModal({ + isModalVisible, + handleBackup, + handleCancel, + backuping, + customFileName, + setCustomFileName +}: S3BackupModalProps) { + const { t } = useTranslation() + + return ( + + setCustomFileName(e.target.value)} + placeholder={t('settings.data.s3.backup.modal.filename.placeholder')} + /> + + ) +} + +interface UseS3RestoreModalProps { + endpoint: string | undefined + region: string | undefined + bucket: string | undefined + accessKeyId: string | undefined + secretAccessKey: string | undefined + root?: string | undefined +} + +export function useS3RestoreModal({ + endpoint, + region, + bucket, + accessKeyId, + secretAccessKey, + root +}: UseS3RestoreModalProps) { + const [isRestoreModalVisible, setIsRestoreModalVisible] = useState(false) + const [restoring, setRestoring] = useState(false) + const [selectedFile, setSelectedFile] = useState(null) + const [loadingFiles, setLoadingFiles] = useState(false) + const [backupFiles, setBackupFiles] = useState([]) + const { t } = useTranslation() + + const showRestoreModal = useCallback(async () => { + if (!endpoint || !region || !bucket || !accessKeyId || !secretAccessKey) { + window.message.error({ content: t('settings.data.s3.manager.config.incomplete'), key: 's3-error' }) + return + } + + setIsRestoreModalVisible(true) + setLoadingFiles(true) + try { + const files = await window.api.backup.listS3Files({ + endpoint, + region, + bucket, + accessKeyId, + secretAccessKey, + root, + autoSync: false, + syncInterval: 0, + maxBackups: 0, + skipBackupFile: false + }) + setBackupFiles(files) + } catch (error: any) { + window.message.error({ + content: t('settings.data.s3.manager.files.fetch.error', { message: error.message }), + key: 'list-files-error' + }) + } finally { + setLoadingFiles(false) + } + }, [endpoint, region, bucket, accessKeyId, secretAccessKey, root, t]) + + const handleRestore = useCallback(async () => { + if (!selectedFile || !endpoint || !region || !bucket || !accessKeyId || !secretAccessKey) { + window.message.error({ + content: !selectedFile + ? t('settings.data.s3.restore.file.required') + : t('settings.data.s3.restore.config.incomplete'), + key: 'restore-error' + }) + return + } + + window.modal.confirm({ + title: t('settings.data.s3.restore.confirm.title'), + content: t('settings.data.s3.restore.confirm.content', { fileName: selectedFile }), + okText: t('settings.data.s3.restore.confirm.ok'), + cancelText: t('settings.data.s3.restore.confirm.cancel'), + centered: true, + onOk: async () => { + setRestoring(true) + try { + await window.api.backup.restoreFromS3({ + endpoint, + region, + bucket, + accessKeyId, + secretAccessKey, + root, + fileName: selectedFile, + autoSync: false, + syncInterval: 0, + maxBackups: 0, + skipBackupFile: false + }) + window.message.success({ content: t('message.restore.success'), key: 's3-restore' }) + setIsRestoreModalVisible(false) + } catch (error: any) { + window.message.error({ + content: t('settings.data.s3.restore.error', { message: error.message }), + key: 'restore-error' + }) + } finally { + setRestoring(false) + } + } + }) + }, [selectedFile, endpoint, region, bucket, accessKeyId, secretAccessKey, root, t]) + + const handleCancel = () => { + setIsRestoreModalVisible(false) + } + + return { + isRestoreModalVisible, + handleRestore, + handleCancel, + restoring, + selectedFile, + setSelectedFile, + loadingFiles, + backupFiles, + showRestoreModal + } +} + +type S3RestoreModalProps = ReturnType + +export function S3RestoreModal({ + isRestoreModalVisible, + handleRestore, + handleCancel, + restoring, + selectedFile, + setSelectedFile, + loadingFiles, + backupFiles +}: S3RestoreModalProps) { + const { t } = useTranslation() + + return ( + +
+ p.value === 'new-api')?.value} + onChange={handleProviderChange} + style={{ width: '100%' }}> + {providerOptions.map((provider) => ( + + + + {provider.label} + + + ))} + + + {/* 当没有可用的 Image Generation 模型时，提示用户先去新增 */} + {modelOptions.length === 0 && ( + + + + )} + + {modelOptions.length > 0 && ( + <> + {mode === 'openai_image_edit' && ( + <> + {t('paintings.input_image')} + + + + + + + )} + + {/* Model Selector */} + {t('paintings.model')} + + + {/* Image Size */} + {selectedModelConfig?.imageSizes && selectedModelConfig.imageSizes.length > 0 && ( + <> + {t('paintings.image.size')} + + + )} + + {/* Quality */} + {selectedModelConfig?.quality && selectedModelConfig.quality.length > 0 && ( + <> + {t('paintings.quality')} + + + )} + + {/* Moderation */} + {mode !== 'openai_image_edit' && + selectedModelConfig?.moderation && + selectedModelConfig.moderation.length > 0 && ( + <> + {t('paintings.moderation')} + + + )} + + {/* Background */} + {mode === 'openai_image_edit' && + selectedModelConfig?.background && + selectedModelConfig.background.length > 0 && ( + <> + {t('paintings.background')} + + + )} + + {/* Number of Images (n) */} + {selectedModelConfig?.max_images && ( + <> + {t('paintings.number_images')} + + + )} + + )} + + + {/* 添加功能切换分段控制器 */} + + + + + +