@heripo/pdf-parser 0.1.40 → 0.1.42

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (3) hide show
  1. package/README.ko.md +19 -1
  2. package/README.md +19 -1
  3. package/package.json +10 -10
package/README.ko.md CHANGED
@@ -65,7 +65,7 @@
65
65
  brew install node
66
66
  ```
67
67
 
68
- #### 2. pnpm >= 11.0.0
68
+ #### 2. pnpm >= 11
69
69
 
70
70
  ```bash
71
71
  npm install -g pnpm
@@ -653,9 +653,13 @@ type PDFConvertOptions = {
653
653
  interface PDFCorrectionOptions {
654
654
  models: {
655
655
  textCorrection: LanguageModel; // 필수: 페이지 텍스트와 표 셀 OCR 보정
656
+ textCorrectionFallback?: LanguageModel; // 선택: textCorrection 실패 시 fallback
656
657
  pageGate: LanguageModel; // 필수: 구조 Review Assistance page gate
658
+ pageGateFallback?: LanguageModel; // 선택: pageGate 실패 시 fallback
657
659
  reviewAssistance: LanguageModel; // 필수: 기본 구조 review 모델
660
+ reviewAssistanceFallback?: LanguageModel; // 선택: reviewAssistance 공통 fallback
658
661
  tableCorrection?: LanguageModel; // 선택: 표 전용 보정 override
662
+ tableCorrectionFallback?: LanguageModel; // 선택: tableCorrection 실패 시 fallback
659
663
  reviewAssistanceTasks?: Partial<
660
664
  Record<
661
665
  | 'text_ocr_hanja'
@@ -667,6 +671,17 @@ interface PDFCorrectionOptions {
667
671
  LanguageModel
668
672
  >
669
673
  >;
674
+ reviewAssistanceTasksFallback?: Partial<
675
+ Record<
676
+ | 'text_ocr_hanja'
677
+ | 'text_integrity'
678
+ | 'text_role_footnote'
679
+ | 'tables'
680
+ | 'pictures_captions'
681
+ | 'layout_bbox_order',
682
+ LanguageModel
683
+ >
684
+ >;
670
685
  };
671
686
  concurrency?: {
672
687
  pages?: number; // 페이지 단위 text correction/page gate 동시성
@@ -687,6 +702,9 @@ interface PDFCorrectionOptions {
687
702
  };
688
703
  autoApplyThreshold?: number; // 자동 적용 최소 confidence (기본값: 0.85)
689
704
  proposalThreshold?: number; // sidecar proposal 최소 confidence (기본값: 0.5)
705
+ forceAutoApply?: boolean; // true일 경우, confidence 임계값에 관계없이 review-assistance 명령을 자동 적용 (기본값: false)
706
+ reviewAssistanceEnabled?: boolean; // false일 경우, review-assistance 단계를 완전히 건너뜀 (기본값: true)
707
+ tableCorrectionEnabled?: boolean; // false일 경우, table-correction 작업을 건너뜀 (기본값: true)
690
708
  temperature?: number; // VLM 생성 temperature (기본값: 0)
691
709
  }
692
710
  ```
package/README.md CHANGED
@@ -65,7 +65,7 @@
65
65
  brew install node
66
66
  ```
67
67
 
68
- #### 2. pnpm >= 11.0.0
68
+ #### 2. pnpm >= 11
69
69
 
70
70
  ```bash
71
71
  npm install -g pnpm
@@ -653,9 +653,13 @@ type PDFConvertOptions = {
653
653
  interface PDFCorrectionOptions {
654
654
  models: {
655
655
  textCorrection: LanguageModel; // Required: page text and table-cell OCR correction
656
+ textCorrectionFallback?: LanguageModel; // Optional: fallback for text correction
656
657
  pageGate: LanguageModel; // Required: structural Review Assistance page gate
658
+ pageGateFallback?: LanguageModel; // Optional: fallback for page gate
657
659
  reviewAssistance: LanguageModel; // Required: default structural review model
660
+ reviewAssistanceFallback?: LanguageModel; // Optional: common fallback for review assistance
658
661
  tableCorrection?: LanguageModel; // Optional: table-specific correction override
662
+ tableCorrectionFallback?: LanguageModel; // Optional: fallback for table correction
659
663
  reviewAssistanceTasks?: Partial<
660
664
  Record<
661
665
  | 'text_ocr_hanja'
@@ -667,6 +671,17 @@ interface PDFCorrectionOptions {
667
671
  LanguageModel
668
672
  >
669
673
  >;
674
+ reviewAssistanceTasksFallback?: Partial<
675
+ Record<
676
+ | 'text_ocr_hanja'
677
+ | 'text_integrity'
678
+ | 'text_role_footnote'
679
+ | 'tables'
680
+ | 'pictures_captions'
681
+ | 'layout_bbox_order',
682
+ LanguageModel
683
+ >
684
+ >;
670
685
  };
671
686
  concurrency?: {
672
687
  pages?: number; // Page-level text correction and page gate concurrency
@@ -687,6 +702,9 @@ interface PDFCorrectionOptions {
687
702
  };
688
703
  autoApplyThreshold?: number; // Minimum confidence for direct mutation (default: 0.85)
689
704
  proposalThreshold?: number; // Minimum confidence for sidecar proposal (default: 0.5)
705
+ forceAutoApply?: boolean; // When true, review-assistance commands auto-apply regardless of threshold (default: false)
706
+ reviewAssistanceEnabled?: boolean; // When false, review-assistance is skipped completely (default: true)
707
+ tableCorrectionEnabled?: boolean; // When false, table-correction is skipped (default: true)
690
708
  temperature?: number; // VLM generation temperature (default: 0)
691
709
  }
692
710
  ```
package/package.json CHANGED
@@ -2,7 +2,7 @@
2
2
  "name": "@heripo/pdf-parser",
3
3
  "private": false,
4
4
  "type": "module",
5
- "version": "0.1.40",
5
+ "version": "0.1.42",
6
6
  "description": "PDF parsing library using Docling SDK with OCR support for macOS",
7
7
  "main": "dist/index.js",
8
8
  "types": "dist/index.d.ts",
@@ -55,24 +55,24 @@
55
55
  "access": "public"
56
56
  },
57
57
  "dependencies": {
58
- "ai": "^7.0.14",
58
+ "ai": "^7.0.56",
59
59
  "docling-sdk": "^2.0.6",
60
- "es-toolkit": "^1.49.0",
60
+ "es-toolkit": "^1.50.0",
61
61
  "yauzl": "3.4.0",
62
62
  "zod": "^4.4.3",
63
- "@heripo/logger": "0.1.40",
64
- "@heripo/model": "0.1.40"
63
+ "@heripo/logger": "0.1.42",
64
+ "@heripo/model": "0.1.42"
65
65
  },
66
66
  "devDependencies": {
67
67
  "@types/yauzl": "^3.4.0",
68
- "@vitest/coverage-v8": "^4.1.9",
69
- "@vitest/expect": "^4.1.9",
68
+ "@vitest/coverage-v8": "^4.1.10",
69
+ "@vitest/expect": "^4.1.10",
70
70
  "tsup": "^8.5.1",
71
- "vitest": "^4.1.9",
71
+ "vitest": "^4.1.10",
72
72
  "@heripo/shared": "0.0.0",
73
- "@heripo/tsconfig": "0.0.0",
74
73
  "@heripo/tsup-config": "0.0.0",
75
- "@heripo/vitest-config": "0.0.0"
74
+ "@heripo/vitest-config": "0.0.0",
75
+ "@heripo/tsconfig": "0.0.0"
76
76
  },
77
77
  "scripts": {
78
78
  "clean": "rimraf dist",