Northern_Lights/cherry-studio
1
0
Fork 0
mirror of https://github.com/CherryHQ/cherry-studio.git synced 2025-12-24 10:40:07 +08:00
Code Issues Actions 3 Packages Projects Releases Wiki Activity

refactor(ocr): 将Tesseract相关配置移至服务内部

Browse Source 
将语言列表和下载URL常量从共享配置移至Tesseract服务内部
使用常量定义图片大小阈值以提高可读性
This commit is contained in:
icarus 2025-08-23 13:26:09 +08:00
parent 843e230af6
commit bfb64522cd
2 changed files with 12 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
packages/shared/config/constant.ts
View File

@ -197,12 +197,6 @@ export enum FeedUrl {
GITHUB_LATEST = 'https://github.com/CherryHQ/cherry-studio/releases/latest/download'
}
export const tesseractLangs = ['chi_sim', 'chi_tra', 'eng']
export enum TesseractLangsDownloadUrl {
CN = 'https://gitcode.com/beyondkmp/tessdata/releases/download/4.1.0/',
GLOBAL = 'https://github.com/tesseract-ocr/tessdata/raw/main/'
}
export enum UpgradeChannel {
LATEST = 'latest', // 最新稳定版本
RC = 'rc', // 公测版本

16
src/main/services/ocr/tesseract/TesseractService.ts
View File

@ -1,6 +1,6 @@
import { loggerService } from '@logger'
import { getIpCountry } from '@main/utils/ipService'
import { MB, TesseractLangsDownloadUrl } from '@shared/config/constant'
import { MB } from '@shared/config/constant'
import { FileMetadata, ImageFileMetadata, isImageFile, OcrResult } from '@types'
import { app } from 'electron'
import fs from 'fs'
@ -114,13 +114,21 @@ const logger = loggerService.withContext('TesseractService')
// 'yi-us': 'yid'
// }
// config
const MB_SIZE_THRESHOLD = 50
const tesseractLangs = ['chi_sim', 'chi_tra', 'eng']
enum TesseractLangsDownloadUrl {
CN = 'https://gitcode.com/beyondkmp/tessdata/releases/download/4.1.0/',
GLOBAL = 'https://github.com/tesseract-ocr/tessdata/raw/main/'
}
export class TesseractService {
private worker: Tesseract.Worker | null = null
async getWorker(): Promise<Tesseract.Worker> {
if (!this.worker) {
// for now, only support limited languages
this.worker = await createWorker(['chi_sim', 'chi_tra', 'eng'], undefined, {
this.worker = await createWorker(tesseractLangs, undefined, {
langPath: await this._getLangPath(),
cachePath: await this._getCacheDir(),
gzip: false,
@ -133,8 +141,8 @@ export class TesseractService {
async imageOcr(file: ImageFileMetadata): Promise<OcrResult> {
const worker = await this.getWorker()
const stat = await fs.promises.stat(file.path)
if (stat.size > 50 * MB) {
throw new Error('This image is too large (max 50MB)')
if (stat.size > MB_SIZE_THRESHOLD * MB) {
throw new Error(`This image is too large (max ${MB_SIZE_THRESHOLD}MB)`)
}
const buffer = await fs.promises.readFile(file.path)
const result = await worker.recognize(buffer)