From af05960cd37a5171a5820b90ac89fa083347bd94 Mon Sep 17 00:00:00 2001
From: icarus <eurfelux@gmail.com>
Date: Fri, 22 Aug 2025 16:54:57 +0800
Subject: [PATCH] =?UTF-8?q?feat(OCR=E6=9C=8D=E5=8A=A1):=20=E6=94=AF?=
 =?UTF-8?q?=E6=8C=81base64=E5=AD=97=E7=AC=A6=E4=B8=B2=E4=BD=9C=E4=B8=BAOCR?=
 =?UTF-8?q?=E8=BE=93=E5=85=A5?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

扩展tesseractOcr函数以接受base64字符串或图像文件作为输入
---
 src/main/services/ocr/OcrService.ts | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)

diff --git a/src/main/services/ocr/OcrService.ts b/src/main/services/ocr/OcrService.ts
index a54c009cf8..11f3a41ec0 100644
--- a/src/main/services/ocr/OcrService.ts
+++ b/src/main/services/ocr/OcrService.ts
@@ -19,19 +19,24 @@ const logger = loggerService.withContext('main:OcrService')
 
 /**
  * ocr by tesseract
- * @param file image file
+ * @param file image file or base64 string
  * @returns ocr result
  * @throws {Error}
  */
-const tesseractOcr = async (file: ImageFileMetadata): Promise<string> => {
+const tesseractOcr = async (file: ImageFileMetadata | string): Promise<string> => {
   try {
     const worker = await getTesseractWorker()
-    const stat = statSync(file.path)
-    if (stat.size > 50 * MB) {
-      throw new Error('This image is too large (max 50MB)')
+    let ret: Tesseract.RecognizeResult
+    if (typeof file === 'string') {
+      ret = await worker.recognize(file)
+    } else {
+      const stat = statSync(file.path)
+      if (stat.size > 50 * MB) {
+        throw new Error('This image is too large (max 50MB)')
+      }
+      const buffer = await readFile(file.path)
+      ret = await worker.recognize(buffer)
     }
-    const buffer = await readFile(file.path)
-    const ret = await worker.recognize(buffer)
     return ret.data.text
   } catch (e) {
     logger.error('Failed to ocr with tesseract.', e as Error)