chore(dependency): update mac-system-ocr and refine electron-builder configuration

electron-builder.yml

@@ -27,12 +27,16 @@ files:
   - '!node_modules/@tavily/core/node_modules/js-tiktoken'
   - '!node_modules/pdf-parse/lib/pdf.js/{v1.9.426,v1.10.88,v2.0.550}'
   - '!node_modules/mammoth/{mammoth.browser.js,mammoth.browser.min.js}'
+  - '!node_modules/pdfjs-dist/web/**/*'
+  - '!node_modules/pdfjs-dist/legacy/web/*'
 asarUnpack:
   - resources/**
   - '**/*.{node,dll,metal,exp,lib}'
 win:
   executableName: Cherry Studio
   artifactName: ${productName}-${version}-portable.${ext}
+  files:
+    - '!node_modules/@cherrystudio/mac-system-ocr'
   target:
     - target: nsis
     - target: portable
@@ -64,6 +68,8 @@ mac:
         - x64
 linux:
   artifactName: ${productName}-${version}-${arch}.${ext}
+  files:
+    - '!node_modules/@cherrystudio/mac-system-ocr'
   target:
     - target: AppImage
       arch:

package.json

@@ -61,7 +61,7 @@
     "@cherrystudio/embedjs-loader-web": "^0.1.29",
     "@cherrystudio/embedjs-loader-xml": "^0.1.29",
     "@cherrystudio/embedjs-openai": "^0.1.29",
-    "@cherrystudio/mac-system-ocr": "^0.2.0",
+    "@cherrystudio/mac-system-ocr": "^0.2.1",
     "@electron-toolkit/utils": "^3.0.0",
     "@electron/notarize": "^2.5.0",
     "@google/generative-ai": "^0.21.0",

src/main/ocr/MistralOcrProvider.ts

@@ -31,15 +31,6 @@ export default class MistralOcrProvider extends BaseOcrProvider {
     if (isLocalFile(file)) {
       Logger.info(`OCR preupload started for local file: ${file.path}`)
 
-      const pdfInfo = await this.getPdfInfo(file.path)
-      if (pdfInfo.pageCount >= 1000) {
-        throw new Error(`PDF page count (${pdfInfo.pageCount}) exceeds the limit of 1000 pages`)
-      }
-      if (pdfInfo.fileSize >= 512 * 1024 * 1024) {
-        const fileSizeMB = Math.round(pdfInfo.fileSize / (1024 * 1024))
-        throw new Error(`PDF file size (${fileSizeMB}MB) exceeds the limit of 300MB`)
-      }
-
       if (file.ext.toLowerCase() === '.pdf') {
         const uploadResponse = await this.fileService.uploadFile(file)
 
@@ -47,12 +38,12 @@ export default class MistralOcrProvider extends BaseOcrProvider {
           Logger.error('File upload failed:', uploadResponse)
           throw new Error('Failed to upload file: ' + uploadResponse.displayName)
         }
-
+        await this.sendOcrProgress(file.id, 15)
         const fileUrl = await this.sdk.files.getSignedUrl({
           fileId: uploadResponse.fileId
         })
         Logger.info('Got signed URL:', fileUrl)
-
+        await this.sendOcrProgress(file.id, 20)
         document = {
           type: 'document_url',
           documentUrl: fileUrl.url
@@ -189,7 +180,7 @@ export default class MistralOcrProvider extends BaseOcrProvider {
     return {
       id: conversionId,
       name: file.name.replace(/\.[^/.]+$/, '.md'),
-      origin_name: mdFileName,
+      origin_name: (file as LocalFileSource).origin_name,
       path: mdFilePath,
       created_at: new Date().toISOString(),
       type: FileTypes.DOCUMENT,
@@ -197,6 +188,6 @@ export default class MistralOcrProvider extends BaseOcrProvider {
       size: fs.statSync(mdFilePath).size,
       count: result.pages.length,
       source: 'local'
-    }
+    } as LocalFileSource
   }
 }

yarn.lock

@@ -632,14 +632,14 @@ __metadata:
   languageName: node
   linkType: hard
 
-"@cherrystudio/mac-system-ocr@npm:^0.2.0":
-  version: 0.2.0
-  resolution: "@cherrystudio/mac-system-ocr@npm:0.2.0"
+"@cherrystudio/mac-system-ocr@npm:^0.2.1":
+  version: 0.2.1
+  resolution: "@cherrystudio/mac-system-ocr@npm:0.2.1"
   dependencies:
     bindings: "npm:^1.5.0"
     node-api-headers: "npm:^1.0.1"
     node-gyp: "npm:latest"
-  checksum: 10c0/3ece86759b9120958aa9682b2e666f6fe495fb15c10113f350a3f124a7116e8db84d186e7efb7fdbd02437b39e6f0ff763e1784d6ffe8d7179b81b458055711a
+  checksum: 10c0/fbba8f85c2244799f281b100fa6fbaf134f742a6f4d0a15f465ed4f0f048edc0acb60ed74908c462f0de424c2e3476892127f25d10ab805c123233e4922d415c
   languageName: node
   linkType: hard
 
@@ -3795,7 +3795,7 @@ __metadata:
     "@cherrystudio/embedjs-loader-web": "npm:^0.1.29"
     "@cherrystudio/embedjs-loader-xml": "npm:^0.1.29"
     "@cherrystudio/embedjs-openai": "npm:^0.1.29"
-    "@cherrystudio/mac-system-ocr": "npm:^0.2.0"
+    "@cherrystudio/mac-system-ocr": "npm:^0.2.1"
     "@electron-toolkit/eslint-config-prettier": "npm:^3.0.0"
     "@electron-toolkit/eslint-config-ts": "npm:^3.0.0"
     "@electron-toolkit/preload": "npm:^3.0.0"