kordoc 4.24.2 → 4.24.3

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 (124) hide show
  1. package/README.md +12 -10
  2. package/dist/{-SRNEV7LS.js → -Q56U3KCA.js} +11 -11
  3. package/dist/{chunk-E7B5CXYI.js → chunk-33NZN73M.js} +2 -2
  4. package/dist/{chunk-NJ5RUDCW.js → chunk-3EWXRIPH.js} +3 -3
  5. package/dist/{chunk-NMFFIUI6.js → chunk-42FSKT7R.js} +2 -2
  6. package/dist/{chunk-M46PB75E.js → chunk-4X73CGIL.js} +2 -2
  7. package/dist/{chunk-PNSALBXH.cjs → chunk-6ZRZLV3J.cjs} +2 -2
  8. package/dist/{chunk-PNSALBXH.cjs.map → chunk-6ZRZLV3J.cjs.map} +1 -1
  9. package/dist/{chunk-W24TJJNN.js → chunk-AEEQN6L3.js} +2 -2
  10. package/dist/{chunk-TW3V6FSH.js → chunk-EEXIFCXC.js} +2 -2
  11. package/dist/{chunk-LMEF7FN5.js → chunk-EWCJCWTT.js} +2 -2
  12. package/dist/{chunk-MHNJOJQV.js → chunk-FZT34MUI.js} +4 -4
  13. package/dist/{chunk-WDSTW3BD.js → chunk-HNX2I7D6.js} +2 -2
  14. package/dist/{chunk-PNX5FYHC.cjs → chunk-HTG3VVET.cjs} +24 -24
  15. package/dist/{chunk-PNX5FYHC.cjs.map → chunk-HTG3VVET.cjs.map} +1 -1
  16. package/dist/{chunk-KUC5I5MB.js → chunk-HU7POUBS.js} +2 -2
  17. package/dist/{chunk-WJAFE3Z6.js → chunk-I2QETSZN.js} +2 -2
  18. package/dist/{chunk-DHMBPMKC.js → chunk-J45OVEUM.js} +139 -15
  19. package/dist/chunk-J45OVEUM.js.map +1 -0
  20. package/dist/{chunk-UNICY546.cjs → chunk-JILIA5EV.cjs} +23 -23
  21. package/dist/{chunk-UNICY546.cjs.map → chunk-JILIA5EV.cjs.map} +1 -1
  22. package/dist/{chunk-TDMBTG7U.js → chunk-KXHNIHLC.js} +2 -2
  23. package/dist/{chunk-JFUUFJZW.cjs → chunk-N65EY5BI.cjs} +6 -6
  24. package/dist/{chunk-JFUUFJZW.cjs.map → chunk-N65EY5BI.cjs.map} +1 -1
  25. package/dist/{chunk-4XC4O7FJ.js → chunk-NKDPCALC.js} +4 -4
  26. package/dist/{chunk-C6ZVB547.js → chunk-OAOM7IEI.js} +11 -11
  27. package/dist/{chunk-4745MFQU.js → chunk-PHZDAHZC.js} +3 -3
  28. package/dist/{chunk-RIRPLFPN.js → chunk-PMX5CPK5.js} +2 -2
  29. package/dist/{chunk-E6VBLBOO.js → chunk-Q2QKRE4P.js} +2 -2
  30. package/dist/{chunk-7NTMM6PO.js → chunk-TZ753B6A.js} +2 -2
  31. package/dist/{chunk-HLTPMSYW.cjs → chunk-VFEI2GIE.cjs} +3 -3
  32. package/dist/{chunk-HLTPMSYW.cjs.map → chunk-VFEI2GIE.cjs.map} +1 -1
  33. package/dist/{chunk-3HN65YAL.js → chunk-W6O7KBOD.js} +2 -2
  34. package/dist/{chunk-VPSF5LEM.js → chunk-XAZ5RGVG.js} +3 -3
  35. package/dist/{chunk-R2GIGE5H.js → chunk-XNNMMYFO.js} +2 -2
  36. package/dist/{chunk-JMOAMKAR.js → chunk-XZOAE2Q7.js} +4 -4
  37. package/dist/{chunk-IGG63DFK.cjs → chunk-YG4YOPK6.cjs} +3 -3
  38. package/dist/{chunk-IGG63DFK.cjs.map → chunk-YG4YOPK6.cjs.map} +1 -1
  39. package/dist/chunks-REYSCBPB.js +10 -0
  40. package/dist/cli.js +30 -30
  41. package/dist/{formula-HK2MNZ3Q.cjs → formula-2BPHF6O7.cjs} +9 -9
  42. package/dist/{formula-HK2MNZ3Q.cjs.map → formula-2BPHF6O7.cjs.map} +1 -1
  43. package/dist/{formula-DSB574H7.js → formula-EA5GDCUD.js} +4 -4
  44. package/dist/{formula-U2LZ67KI.js → formula-IREEAC4T.js} +3 -3
  45. package/dist/{image-ocr-QNIMVE5A.cjs → image-ocr-CCTE7HX3.cjs} +17 -17
  46. package/dist/{image-ocr-QNIMVE5A.cjs.map → image-ocr-CCTE7HX3.cjs.map} +1 -1
  47. package/dist/{image-ocr-J54Q3YWP.js → image-ocr-JQJC5HWK.js} +6 -6
  48. package/dist/{image-ocr-3QB2HLSI.js → image-ocr-LGOFUS3O.js} +9 -9
  49. package/dist/index.cjs +517 -393
  50. package/dist/index.cjs.map +1 -1
  51. package/dist/index.js +139 -15
  52. package/dist/index.js.map +1 -1
  53. package/dist/mcp.js +23 -23
  54. package/dist/{model-bundle-CGWU4ALC.js → model-bundle-YW62Y6UD.js} +5 -5
  55. package/dist/{models-A3TF3BDV.js → models-AAL3IS2R.js} +5 -5
  56. package/dist/{parse-worker-v2-MNUD4WAR.js → parse-worker-v2-3JAHV7NY.js} +12 -12
  57. package/dist/{parser-C3ARF26I.js → parser-4FCKKJ2N.js} +18 -14
  58. package/dist/parser-4FCKKJ2N.js.map +1 -0
  59. package/dist/{parser-RPAHN3OR.cjs → parser-HE45CCSG.cjs} +163 -159
  60. package/dist/parser-HE45CCSG.cjs.map +1 -0
  61. package/dist/{parser-OBGWEWSU.js → parser-ZBP5GOQK.js} +15 -11
  62. package/dist/parser-ZBP5GOQK.js.map +1 -0
  63. package/dist/pdf-ocr-7MDCLWAA.cjs +31 -0
  64. package/dist/{pdf-ocr-GA5QJFNK.cjs.map → pdf-ocr-7MDCLWAA.cjs.map} +1 -1
  65. package/dist/{pdf-ocr-K53REDNY.js → pdf-ocr-HFAEOA7W.js} +6 -6
  66. package/dist/{pdf-ocr-ENLVEEYL.js → pdf-ocr-Q4TPGDDI.js} +9 -9
  67. package/dist/{profile-io-ZGFEO7CB.js → profile-io-EDQTKLLL.js} +3 -3
  68. package/dist/{rasterize-QMC2SAUT.js → rasterize-VD2XGHUJ.js} +3 -3
  69. package/dist/{redact-doc-JH6RZZBE.js → redact-doc-TBIBQJCT.js} +11 -11
  70. package/dist/{render-ZGKQLKPJ.js → render-J5KAZAQH.js} +7 -7
  71. package/dist/seal-H2JNAZU7.js +10 -0
  72. package/dist/{setup-LAQGCYQ2.js → setup-5LYHCQRU.js} +3 -3
  73. package/dist/visual-7EO6GR3M.js +21 -0
  74. package/dist/{watch-7YPB4TKC.js → watch-YP4IWM75.js} +12 -12
  75. package/package.json +1 -1
  76. package/dist/chunk-DHMBPMKC.js.map +0 -1
  77. package/dist/chunks-AL4P5TO5.js +0 -10
  78. package/dist/parser-C3ARF26I.js.map +0 -1
  79. package/dist/parser-OBGWEWSU.js.map +0 -1
  80. package/dist/parser-RPAHN3OR.cjs.map +0 -1
  81. package/dist/pdf-ocr-GA5QJFNK.cjs +0 -31
  82. package/dist/seal-KMGYTJAV.js +0 -10
  83. package/dist/visual-4OYJFGXL.js +0 -21
  84. /package/dist/{-SRNEV7LS.js.map → -Q56U3KCA.js.map} +0 -0
  85. /package/dist/{chunk-E7B5CXYI.js.map → chunk-33NZN73M.js.map} +0 -0
  86. /package/dist/{chunk-NJ5RUDCW.js.map → chunk-3EWXRIPH.js.map} +0 -0
  87. /package/dist/{chunk-NMFFIUI6.js.map → chunk-42FSKT7R.js.map} +0 -0
  88. /package/dist/{chunk-M46PB75E.js.map → chunk-4X73CGIL.js.map} +0 -0
  89. /package/dist/{chunk-W24TJJNN.js.map → chunk-AEEQN6L3.js.map} +0 -0
  90. /package/dist/{chunk-TW3V6FSH.js.map → chunk-EEXIFCXC.js.map} +0 -0
  91. /package/dist/{chunk-LMEF7FN5.js.map → chunk-EWCJCWTT.js.map} +0 -0
  92. /package/dist/{chunk-MHNJOJQV.js.map → chunk-FZT34MUI.js.map} +0 -0
  93. /package/dist/{chunk-WDSTW3BD.js.map → chunk-HNX2I7D6.js.map} +0 -0
  94. /package/dist/{chunk-KUC5I5MB.js.map → chunk-HU7POUBS.js.map} +0 -0
  95. /package/dist/{chunk-WJAFE3Z6.js.map → chunk-I2QETSZN.js.map} +0 -0
  96. /package/dist/{chunk-TDMBTG7U.js.map → chunk-KXHNIHLC.js.map} +0 -0
  97. /package/dist/{chunk-4XC4O7FJ.js.map → chunk-NKDPCALC.js.map} +0 -0
  98. /package/dist/{chunk-C6ZVB547.js.map → chunk-OAOM7IEI.js.map} +0 -0
  99. /package/dist/{chunk-4745MFQU.js.map → chunk-PHZDAHZC.js.map} +0 -0
  100. /package/dist/{chunk-RIRPLFPN.js.map → chunk-PMX5CPK5.js.map} +0 -0
  101. /package/dist/{chunk-E6VBLBOO.js.map → chunk-Q2QKRE4P.js.map} +0 -0
  102. /package/dist/{chunk-7NTMM6PO.js.map → chunk-TZ753B6A.js.map} +0 -0
  103. /package/dist/{chunk-3HN65YAL.js.map → chunk-W6O7KBOD.js.map} +0 -0
  104. /package/dist/{chunk-VPSF5LEM.js.map → chunk-XAZ5RGVG.js.map} +0 -0
  105. /package/dist/{chunk-R2GIGE5H.js.map → chunk-XNNMMYFO.js.map} +0 -0
  106. /package/dist/{chunk-JMOAMKAR.js.map → chunk-XZOAE2Q7.js.map} +0 -0
  107. /package/dist/{chunks-AL4P5TO5.js.map → chunks-REYSCBPB.js.map} +0 -0
  108. /package/dist/{formula-DSB574H7.js.map → formula-EA5GDCUD.js.map} +0 -0
  109. /package/dist/{formula-U2LZ67KI.js.map → formula-IREEAC4T.js.map} +0 -0
  110. /package/dist/{image-ocr-J54Q3YWP.js.map → image-ocr-JQJC5HWK.js.map} +0 -0
  111. /package/dist/{image-ocr-3QB2HLSI.js.map → image-ocr-LGOFUS3O.js.map} +0 -0
  112. /package/dist/{model-bundle-CGWU4ALC.js.map → model-bundle-YW62Y6UD.js.map} +0 -0
  113. /package/dist/{models-A3TF3BDV.js.map → models-AAL3IS2R.js.map} +0 -0
  114. /package/dist/{parse-worker-v2-MNUD4WAR.js.map → parse-worker-v2-3JAHV7NY.js.map} +0 -0
  115. /package/dist/{pdf-ocr-ENLVEEYL.js.map → pdf-ocr-HFAEOA7W.js.map} +0 -0
  116. /package/dist/{pdf-ocr-K53REDNY.js.map → pdf-ocr-Q4TPGDDI.js.map} +0 -0
  117. /package/dist/{profile-io-ZGFEO7CB.js.map → profile-io-EDQTKLLL.js.map} +0 -0
  118. /package/dist/{rasterize-QMC2SAUT.js.map → rasterize-VD2XGHUJ.js.map} +0 -0
  119. /package/dist/{redact-doc-JH6RZZBE.js.map → redact-doc-TBIBQJCT.js.map} +0 -0
  120. /package/dist/{render-ZGKQLKPJ.js.map → render-J5KAZAQH.js.map} +0 -0
  121. /package/dist/{seal-KMGYTJAV.js.map → seal-H2JNAZU7.js.map} +0 -0
  122. /package/dist/{setup-LAQGCYQ2.js.map → setup-5LYHCQRU.js.map} +0 -0
  123. /package/dist/{visual-4OYJFGXL.js.map → visual-7EO6GR3M.js.map} +0 -0
  124. /package/dist/{watch-7YPB4TKC.js.map → watch-YP4IWM75.js.map} +0 -0
package/README.md CHANGED
@@ -4,6 +4,7 @@
4
4
 
5
5
  [![npm version](https://img.shields.io/npm/v/kordoc.svg)](https://www.npmjs.com/package/kordoc)
6
6
  [![license](https://img.shields.io/npm/l/kordoc.svg)](LICENSE)
7
+ [![MCP Registry](https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fregistry.modelcontextprotocol.io%2Fv0%2Fservers%3Fsearch%3Dio.github.chrisryugj%2Fkordoc%26version%3Dlatest&query=%24.servers%5B0%5D.server.version&label=MCP%20Registry&color=blue)](https://registry.modelcontextprotocol.io/v0/servers?search=io.github.chrisryugj/kordoc)
7
8
 
8
9
  > *대한민국에서 둘째가라면 서러울 문서지옥. 거기서 7년 버틴 공무원이 만들었습니다.*
9
10
 
@@ -38,7 +39,7 @@ CLI는 설치 없이 `npx kordoc`으로 실행할 수 있습니다. PDF·OCR 의
38
39
  npx -y kordoc setup
39
40
  ```
40
41
 
41
- Claude Desktop·Claude Code·Cursor·Codex 등 설치된 클라이언트에 MCP를 등록합니다. 클라이언트를 재시작하면 파싱·비교·생성·검수·렌더 등 **18개 도구**를 사용할 수 있습니다.
42
+ Claude Desktop·Claude Code·Cursor·Codex 등 설치된 클라이언트에 MCP를 등록합니다. 공식 MCP 레지스트리에도 [`io.github.chrisryugj/kordoc`](https://registry.modelcontextprotocol.io/v0/servers?search=io.github.chrisryugj/kordoc)으로 등재돼 있습니다. 클라이언트를 재시작하면 파싱·비교·생성·검수·렌더 등 **18개 도구**를 사용할 수 있습니다.
42
43
 
43
44
  Claude Code 플러그인:
44
45
 
@@ -100,7 +101,7 @@ await writeFile("보고서.hwpx", Buffer.from(hwpx))
100
101
 
101
102
  ## 검증 결과
102
103
 
103
- **4.22.14 · 2026-10-19 검증**. 고정 코퍼스와 참조 기준으로 측정한 결과입니다.
104
+ **4.22.14 · 2026-10-19 검증**(렌더 배치는 4.24.3 · 2026-10-24). 고정 코퍼스와 참조 기준으로 측정한 결과입니다.
104
105
 
105
106
  | 대상 | 규모 | 결과 |
106
107
  | --- | --- | --- |
@@ -114,20 +115,21 @@ await writeFile("보고서.hwpx", Buffer.from(hwpx))
114
115
  | PDF 전체 | 1,911문서 중 1,729건 채점 | 글 커버리지 **99.80%** |
115
116
  | OCR | 53문서 · 102쪽 | CER **0.0409** · 글자 재현율 **99.02%** |
116
117
  | opendataloader-bench | PDF 200문서 | 종합 기본값 **0.964** · OCR 끔 **0.939** |
118
+ | HWPX 렌더 배치 (4.24.3) | 한컴 PDF 짝 · 글 조각 90,692 · 그림 4,107 | 글 위치(±1.5pt) **87,215 · 96.2%** · 그림 위치(±2pt) **4,017 · 97.8%** |
117
119
 
118
120
  표 구조 점수와 셀 내용·화면 재현은 별도 지표입니다. 중첩표 100%는 위 101표에 한정하며 전체 PDF·조판의 완전 일치를 뜻하지 않습니다. 채점 범위·제외 기준·옵션별 수치·재현 방법은 [벤치마크 상세](docs/benchmarks.md)에 있습니다.
119
121
 
120
122
  ## 최근 업데이트
121
123
 
122
- 최신 배포: **[4.24.2](https://github.com/chrisryugj/kordoc/releases/tag/v4.24.2)** · 2026-10-24
124
+ 최신 배포: **[4.24.3](https://github.com/chrisryugj/kordoc/releases/tag/v4.24.3)** · 2026-10-24
123
125
 
124
- | 버전 | 주요 변경 |
125
- | --- | --- |
126
- | 4.24.2 | 렌더를 한컴 PDF 와 더 가깝게: 쪽을 넘어온 떠 있는 표·칸 안 여러 개체 쌓기·세로 오프셋 이중 계산·마지막 행 높이·줄바꿈 앞 공백·쪽 번호(렌더 글 위치 86,508 → 86,954 / 90,550) |
127
- | 4.24.1 | 검수 수정: 블록 검수 희소 표 메모리 폭주·천원 규칙 이차 시간·예산 검산(목록 줄·소계(A)·비율 행), `generate --template` 견본 덮어쓰기 차단 |
128
- | 4.24.0 | 블록 단위 검수 `lintBlocks`: 열린 문서의 문단·표 칸을 블록으로 넘기면 지적을 블록 id·블록 안 위치로 돌려줌(CLI `lint-blocks`, MCP `lint_document`, 워커, Python SDK) |
129
- | 4.23.1 | 견본 생성 제목표 결재선 칸을 직위별로, 산출근거 줄·표 칸 천원 단위 오탐 제거 |
130
- | 4.23.0 | 부서 서식 견본으로 새 문서 생성(`generate --template`), 예산 문서 검수(`lint --budget`), MCP `lint_document` |
126
+ | 버전 | 주요 변경 | 수치 변화 |
127
+ | --- | --- | --- |
128
+ | 4.24.3 | 렌더: 한컴이 쪽에서 갈라 저장한 여러 행 칸, 엇갈린 병합 표 행 높이, 짝·홀 바탕쪽 물려받기, 묶음 안 회전 사진, 그림자 사진 · PDF: 쪽마다 갈린 행의 이어짐 보존 | 렌더 글 위치 86,954 → **87,215** · 그림 3,881 → **4,017** · 별표 PDF 재현율 0.99976 → **0.99982** |
129
+ | 4.24.2 | 렌더를 한컴 PDF 와 맞춤: 쪽을 넘어온 떠 있는 표, 칸 안 개체 쌓기, 세로 오프셋 이중 계산, 마지막 행 높이, 줄바꿈 앞 공백, 쪽 번호 | 렌더 글 위치 86,508 → **86,954** / 90,550 · 그림 3,872 → **3,881** / 3,976 |
130
+ | 4.24.1 | 블록 검수 희소 표 메모리 폭주, 천원 규칙 이차 시간, 예산 검산(목록 줄·소계(A)·비율 행), `generate --template` 견본 덮어쓰기 차단 | 20만 자 표기 검수 7.3초 → **10ms 안팎** |
131
+ | 4.24.0 | 블록 단위 검수 `lintBlocks`(CLI `lint-blocks`, MCP `lint_document`, 워커, Python SDK) | 새 기능 |
132
+ | 4.23.1 | 견본 생성 제목표 결재선 칸을 직위별로, 산출근거 줄·표 칸 천원 단위 오탐 제거 | 오탐 수정 |
131
133
 
132
134
  이전 버전과 상세 내용은 [변경 이력](CHANGELOG.md)에 있습니다.
133
135
 
@@ -58,10 +58,10 @@ import {
58
58
  usesGaejosikMunche,
59
59
  usesLogicStructure,
60
60
  validateHwpx
61
- } from "./chunk-C6ZVB547.js";
61
+ } from "./chunk-OAOM7IEI.js";
62
62
  import {
63
63
  blocksToChunks
64
- } from "./chunk-NMFFIUI6.js";
64
+ } from "./chunk-42FSKT7R.js";
65
65
  import {
66
66
  SPACE_EM_FIXED,
67
67
  SPACE_EM_FONT,
@@ -86,20 +86,20 @@ import {
86
86
  renderSceneToHtml,
87
87
  simulateWrap,
88
88
  simulateWrapKeepWord
89
- } from "./chunk-DHMBPMKC.js";
90
- import "./chunk-E6VBLBOO.js";
89
+ } from "./chunk-J45OVEUM.js";
90
+ import "./chunk-Q2QKRE4P.js";
91
91
  import {
92
92
  blocksToPages
93
- } from "./chunk-7NTMM6PO.js";
93
+ } from "./chunk-TZ753B6A.js";
94
94
  import "./chunk-VKVDMH7X.js";
95
95
  import "./chunk-VQFA55IC.js";
96
96
  import "./chunk-BYMPPVT3.js";
97
- import "./chunk-KUC5I5MB.js";
97
+ import "./chunk-HU7POUBS.js";
98
98
  import {
99
99
  blocksToMarkdown,
100
100
  flattenLayoutTables,
101
101
  hasStructuredCellContent
102
- } from "./chunk-WDSTW3BD.js";
102
+ } from "./chunk-HNX2I7D6.js";
103
103
  import {
104
104
  ValueCursor,
105
105
  applySplices,
@@ -113,11 +113,11 @@ import {
113
113
  isLabelCell,
114
114
  placeSealHwpx,
115
115
  scanSectionXml
116
- } from "./chunk-NJ5RUDCW.js";
117
- import "./chunk-RIRPLFPN.js";
116
+ } from "./chunk-3EWXRIPH.js";
117
+ import "./chunk-PMX5CPK5.js";
118
118
  import {
119
119
  VERSION
120
- } from "./chunk-LMEF7FN5.js";
120
+ } from "./chunk-EWCJCWTT.js";
121
121
  import {
122
122
  DEFAULT_REDACT_RULES,
123
123
  redactMarkdown,
@@ -227,4 +227,4 @@ export {
227
227
  usesLogicStructure,
228
228
  validateHwpx
229
229
  };
230
- //# sourceMappingURL=-SRNEV7LS.js.map
230
+ //# sourceMappingURL=-Q56U3KCA.js.map
@@ -1,6 +1,6 @@
1
1
  import {
2
2
  KordocError
3
- } from "./chunk-WJAFE3Z6.js";
3
+ } from "./chunk-I2QETSZN.js";
4
4
 
5
5
  // src/pdf/formula/models.ts
6
6
  import { createHash as createHash2 } from "crypto";
@@ -310,4 +310,4 @@ export {
310
310
  ensureFormulaModels,
311
311
  ensureSingleModel
312
312
  };
313
- //# sourceMappingURL=chunk-E7B5CXYI.js.map
313
+ //# sourceMappingURL=chunk-33NZN73M.js.map
@@ -2,11 +2,11 @@
2
2
  import {
3
3
  parseRenderStyles,
4
4
  readZipEntry
5
- } from "./chunk-RIRPLFPN.js";
5
+ } from "./chunk-PMX5CPK5.js";
6
6
  import {
7
7
  KordocError,
8
8
  precheckZipSize
9
- } from "./chunk-LMEF7FN5.js";
9
+ } from "./chunk-EWCJCWTT.js";
10
10
 
11
11
  // src/form/seal.ts
12
12
  import JSZip from "jszip";
@@ -1858,4 +1858,4 @@ export {
1858
1858
  patchZipEntries,
1859
1859
  placeSealHwpx
1860
1860
  };
1861
- //# sourceMappingURL=chunk-NJ5RUDCW.js.map
1861
+ //# sourceMappingURL=chunk-3EWXRIPH.js.map
@@ -2,7 +2,7 @@
2
2
  import {
3
3
  assignTableIds,
4
4
  blocksToMarkdown
5
- } from "./chunk-WDSTW3BD.js";
5
+ } from "./chunk-HNX2I7D6.js";
6
6
 
7
7
  // src/chunks.ts
8
8
  var LIST_MARKER_RE = /^(?:[□■◇◆○◎●◦ㅇ•▪▸▶※-]|\d{1,3}[.)]|[가나다라마바사아자차카타파하][.)]|\([가나다라마바사아자차카타파하0-9]{1,3}\)|[①-⑳㉮-㉻㈎-㈛])\s/;
@@ -82,4 +82,4 @@ function blocksToChunks(blocks, options) {
82
82
  export {
83
83
  blocksToChunks
84
84
  };
85
- //# sourceMappingURL=chunk-NMFFIUI6.js.map
85
+ //# sourceMappingURL=chunk-42FSKT7R.js.map
@@ -1,7 +1,7 @@
1
1
  #!/usr/bin/env node
2
2
  import {
3
3
  KordocError
4
- } from "./chunk-LMEF7FN5.js";
4
+ } from "./chunk-EWCJCWTT.js";
5
5
 
6
6
  // src/hwpx/profile-io.ts
7
7
  import { z } from "zod";
@@ -89,4 +89,4 @@ export {
89
89
  formatProfileSchema,
90
90
  parseFormatProfileJson
91
91
  };
92
- //# sourceMappingURL=chunk-M46PB75E.js.map
92
+ //# sourceMappingURL=chunk-4X73CGIL.js.map
@@ -1,5 +1,5 @@
1
1
  "use strict";Object.defineProperty(exports, "__esModule", {value: true}); function _nullishCoalesce(lhs, rhsFn) { if (lhs != null) { return lhs; } else { return rhsFn(); } } function _optionalChain(ops) { let lastAccessLHS = undefined; let value = ops[0]; let i = 1; while (i < ops.length) { const op = ops[i]; const fn = ops[i + 1]; i += 2; if ((op === 'optionalAccess' || op === 'optionalCall') && value == null) { return undefined; } if (op === 'access' || op === 'optionalAccess') { lastAccessLHS = value; value = fn(value); } else if (op === 'call' || op === 'optionalCall') { value = fn((...args) => value.call(lastAccessLHS, ...args)); lastAccessLHS = undefined; } } return value; }// src/utils.ts
2
- var VERSION = true ? "4.24.2" : "0.0.0-dev";
2
+ var VERSION = true ? "4.24.3" : "0.0.0-dev";
3
3
  function toArrayBuffer(buf) {
4
4
  if (buf.byteOffset === 0 && buf.byteLength === buf.buffer.byteLength) {
5
5
  return buf.buffer;
@@ -161,4 +161,4 @@ function partExtension(path) {
161
161
 
162
162
 
163
163
  exports.VERSION = VERSION; exports.toArrayBuffer = toArrayBuffer; exports.KordocError = KordocError; exports.sanitizeError = sanitizeError; exports.isPathTraversal = isPathTraversal; exports.normalizeSectionHref = normalizeSectionHref; exports.compareSectionPaths = compareSectionPaths; exports.unzipLimitBytes = unzipLimitBytes; exports.precheckZipSize = precheckZipSize; exports.stripDtd = stripDtd; exports.sanitizeHref = sanitizeHref; exports.escapeHtml = escapeHtml; exports.unescapeHtml = unescapeHtml; exports.safeMin = safeMin; exports.safeMax = safeMax; exports.OPTIONAL_DEP_INSTALL_HINT = OPTIONAL_DEP_INSTALL_HINT; exports.classifyError = classifyError; exports.partExtension = partExtension;
164
- //# sourceMappingURL=chunk-PNSALBXH.cjs.map
164
+ //# sourceMappingURL=chunk-6ZRZLV3J.cjs.map
@@ -1 +1 @@
1
- {"version":3,"sources":["/Users/mong-e/workspace/kordoc/dist/chunk-PNSALBXH.cjs","../src/utils.ts"],"names":[],"mappings":"AAAA;ACMO,IAAM,QAAA,EAAkB,KAAA,EAA4C,SAAA,EAAqB,WAAA;AAOzF,SAAS,aAAA,CAAc,GAAA,EAA0B;AACtD,EAAA,GAAA,CAAI,GAAA,CAAI,WAAA,IAAe,EAAA,GAAK,GAAA,CAAI,WAAA,IAAe,GAAA,CAAI,MAAA,CAAO,UAAA,EAAY;AACpE,IAAA,OAAO,GAAA,CAAI,MAAA;AAAA,EACb;AACA,EAAA,OAAO,GAAA,CAAI,MAAA,CAAO,KAAA,CAAM,GAAA,CAAI,UAAA,EAAY,GAAA,CAAI,WAAA,EAAa,GAAA,CAAI,UAAU,CAAA;AACzE;AAMO,IAAM,YAAA,EAAN,MAAA,QAA0B,MAAM;AAAA,EACrC,WAAA,CAAY,OAAA,EAAiB;AAC3B,IAAA,KAAA,CAAM,OAAO,CAAA;AACb,IAAA,IAAA,CAAK,KAAA,EAAO,aAAA;AAAA,EACd;AACF,CAAA;AAMA,SAAS,cAAA,CAAe,GAAA,EAAuB;AAC7C,EAAA,GAAA,CAAI,OAAO,IAAA,IAAQ,SAAA,GAAY,IAAA,IAAQ,IAAA,EAAM,OAAO,KAAA;AACpD,EAAA,GAAA,CAAK,GAAA,CAA8B,KAAA,IAAS,QAAA,EAAU,OAAO,IAAA;AAC7D,EAAA,MAAM,IAAA,EAAO,GAAA,CAAc,OAAA;AAC3B,EAAA,OAAO,OAAO,IAAA,IAAQ,SAAA,GAAY,mCAAA,CAAoC,IAAA,CAAK,GAAG,CAAA;AAChF;AAMO,SAAS,aAAA,CAAc,GAAA,EAAsB;AAClD,EAAA,GAAA,CAAI,IAAA,WAAe,WAAA,EAAa,OAAO,GAAA,CAAI,OAAA;AAI3C,EAAA,GAAA,CAAI,cAAA,CAAe,GAAG,CAAA,EAAG,OAAO,uGAAA;AAChC,EAAA,OAAO,0FAAA;AACT;AAMO,SAAS,eAAA,CAAgB,IAAA,EAAuB;AACrD,EAAA,GAAA,CAAI,IAAA,CAAK,QAAA,CAAS,IAAM,CAAA,EAAG,OAAO,IAAA;AAClC,EAAA,MAAM,WAAA,EAAa,IAAA,CAAK,OAAA,CAAQ,KAAA,EAAO,GAAG,CAAA;AAC1C,EAAA,MAAM,SAAA,EAAW,UAAA,CAAW,KAAA,CAAM,GAAG,CAAA;AACrC,EAAA,OAAO,QAAA,CAAS,IAAA,CAAK,CAAA,CAAA,EAAA,GAAK,EAAA,IAAM,IAAI,EAAA,GAAK,UAAA,CAAW,UAAA,CAAW,GAAG,EAAA,GAAK,YAAA,CAAa,IAAA,CAAK,UAAU,CAAA;AACrG;AASO,SAAS,oBAAA,CAAqB,IAAA,EAA6B;AAChE,EAAA,GAAA,CAAI,CAAC,IAAA,EAAM,OAAO,IAAA;AAClB,EAAA,IAAI,WAAA,EAAa,IAAA,CAAK,OAAA,CAAQ,KAAA,EAAO,GAAG,CAAA,CAAE,OAAA,CAAQ,MAAA,EAAQ,EAAE,CAAA;AAC5D,EAAA,GAAA,CAAI,eAAA,CAAgB,UAAU,CAAA,EAAG,OAAO,IAAA;AACxC,EAAA,GAAA,CAAI,sBAAA,CAAuB,IAAA,CAAK,UAAU,CAAA,EAAG,WAAA,EAAa,YAAA,EAAc,UAAA;AACxE,EAAA,OAAO,6BAAA,CAA8B,IAAA,CAAK,UAAU,EAAA,EAAI,WAAA,EAAa,IAAA;AACvE;AAGO,SAAS,mBAAA,CAAoB,CAAA,EAAW,CAAA,EAAmB;AAChE,EAAA,MAAM,GAAA,EAAK,MAAA,kCAAO,CAAA,qBAAE,KAAA,mBAAM,uBAAuB,CAAA,4BAAA,CAAI,CAAC,GAAA,UAAK,MAAA,CAAO,kBAAgB,CAAA;AAClF,EAAA,MAAM,GAAA,EAAK,MAAA,kCAAO,CAAA,qBAAE,KAAA,mBAAM,uBAAuB,CAAA,4BAAA,CAAI,CAAC,GAAA,UAAK,MAAA,CAAO,kBAAgB,CAAA;AAClF,EAAA,OAAO,GAAA,IAAO,GAAA,EAAK,CAAA,CAAE,aAAA,CAAc,CAAC,EAAA,EAAI,GAAA,EAAK,EAAA;AAC/C;AAKA,IAAM,iBAAA,EAAmB,IAAA;AAMlB,SAAS,eAAA,CAAgB,YAAA,EAA8B;AAC5D,EAAA,MAAM,GAAA,EAAK,MAAA,CAAO,OAAA,CAAQ,GAAA,CAAI,mBAAmB,CAAA;AACjD,EAAA,OAAO,MAAA,CAAO,QAAA,CAAS,EAAE,EAAA,GAAK,GAAA,EAAK,EAAA,EAAI,IAAA,CAAK,GAAA,CAAI,EAAA,EAAI,gBAAgB,EAAA,EAAI,KAAA,EAAO,KAAA,EAAO,YAAA;AACxF;AAMO,SAAS,eAAA,CACd,MAAA,EACA,oBAAA,EAAsB,IAAA,EAAM,KAAA,EAAO,IAAA,EACnC,WAAA,EAAa,GAAA,EAGb,KAAA,EACmD;AACnD,EAAA,IAAI;AACF,IAAA,MAAM,KAAA,EAAO,IAAI,QAAA,CAAS,MAAM,CAAA;AAChC,IAAA,MAAM,IAAA,EAAM,MAAA,CAAO,UAAA;AAEnB,IAAA,IAAI,WAAA,EAAa,CAAA,CAAA;AACjB,IAAA,IAAA,CAAA,IAAS,EAAA,EAAI,IAAA,EAAM,EAAA,EAAI,EAAA,GAAK,IAAA,CAAK,GAAA,CAAI,CAAA,EAAG,IAAA,EAAM,KAAK,CAAA,EAAG,CAAA,EAAA,EAAK;AACzD,MAAA,GAAA,CAAI,IAAA,CAAK,SAAA,CAAU,CAAA,EAAG,IAAI,EAAA,IAAM,SAAA,EAAY;AAAE,QAAA,WAAA,EAAa,CAAA;AAAG,QAAA,KAAA;AAAA,MAAM;AAAA,IACtE;AACA,IAAA,GAAA,CAAI,WAAA,EAAa,CAAA,EAAG,OAAO,EAAE,iBAAA,EAAmB,CAAA,EAAG,UAAA,EAAY,EAAE,CAAA;AAEjE,IAAA,MAAM,WAAA,EAAa,IAAA,CAAK,SAAA,CAAU,WAAA,EAAa,EAAA,EAAI,IAAI,CAAA;AACvD,IAAA,GAAA,CAAI,WAAA,EAAa,UAAA,EAAY;AAC3B,MAAA,MAAM,IAAI,WAAA,CAAY,CAAA,4CAAA,EAAiB,UAAU,CAAA,eAAA,EAAQ,UAAU,CAAA,CAAA,CAAG,CAAA;AAAA,IACxE;AAEA,IAAA,MAAM,OAAA,EAAS,IAAA,CAAK,SAAA,CAAU,WAAA,EAAa,EAAA,EAAI,IAAI,CAAA;AACnD,IAAA,MAAM,SAAA,EAAW,IAAA,CAAK,SAAA,CAAU,WAAA,EAAa,EAAA,EAAI,IAAI,CAAA;AACrD,IAAA,GAAA,CAAI,SAAA,EAAW,OAAA,EAAS,GAAA,EAAK,OAAO,EAAE,iBAAA,EAAmB,CAAA,EAAG,WAAW,CAAA;AAEvE,IAAA,IAAI,kBAAA,EAAoB,CAAA;AACxB,IAAA,IAAI,kBAAA,EAAoB,CAAA;AACxB,IAAA,IAAI,IAAA,EAAM,QAAA;AACV,IAAA,IAAA,CAAA,IAAS,EAAA,EAAI,CAAA,EAAG,EAAA,EAAI,WAAA,GAAc,IAAA,EAAM,GAAA,GAAM,SAAA,EAAW,MAAA,EAAQ,CAAA,EAAA,EAAK;AACpE,MAAA,GAAA,CAAI,IAAA,CAAK,SAAA,CAAU,GAAA,EAAK,IAAI,EAAA,IAAM,QAAA,EAAY,KAAA;AAG9C,MAAA,MAAM,SAAA,EAAW,IAAA,CAAK,SAAA,CAAU,IAAA,EAAM,EAAA,EAAI,IAAI,CAAA;AAC9C,MAAA,MAAM,UAAA,EAAY,SAAA,EAAW,GAAA,GAAM,IAAA,GAAO,IAAA,CAAK,SAAA,CAAU,QAAA,EAAU,IAAI,EAAA,IAAM,SAAA,EAAa,IAAA,CAAK,SAAA,CAAU,SAAA,EAAW,EAAA,EAAI,IAAI,EAAA,EAAI,CAAA;AAChI,MAAA,MAAM,KAAA,EAAO,IAAA,CAAK,GAAA,CAAI,IAAA,CAAK,SAAA,CAAU,IAAA,EAAM,EAAA,EAAI,IAAI,CAAA,EAAG,SAAS,CAAA;AAC/D,MAAA,MAAM,QAAA,EAAU,IAAA,CAAK,SAAA,CAAU,IAAA,EAAM,EAAA,EAAI,IAAI,CAAA;AAC7C,MAAA,MAAM,SAAA,EAAW,IAAA,CAAK,SAAA,CAAU,IAAA,EAAM,EAAA,EAAI,IAAI,CAAA;AAC9C,MAAA,MAAM,WAAA,EAAa,IAAA,CAAK,SAAA,CAAU,IAAA,EAAM,EAAA,EAAI,IAAI,CAAA;AAEhD,MAAA,MAAM,QAAA,EAAU,CAAC,CAAC,MAAA,GAAS,IAAA,EAAM,GAAA,EAAK,QAAA,GAAW,IAAA,GAC/C,KAAA,CAAM,EAAA,CAAG,IAAA,CAAK,IAAI,WAAA,CAAY,CAAA,CAAE,MAAA,CAAO,IAAI,UAAA,CAAW,MAAA,EAAQ,IAAA,EAAM,EAAA,EAAI,OAAO,CAAC,CAAC,CAAA;AACnF,MAAA,MAAM,MAAA,EAAQ,QAAA,GAAW,CAAC,iBAAC,KAAA,6BAAO,QAAA,GAAS,KAAA,CAAM,KAAA,CAAM,IAAA,CAAK,IAAI,WAAA,CAAY,CAAA,CAAE,MAAA,CAAO,IAAI,UAAA,CAAW,MAAA,EAAQ,IAAA,EAAM,EAAA,EAAI,OAAO,CAAC,CAAC,CAAA;AAC/H,MAAA,GAAA,CAAI,QAAA,GAAW,CAAC,KAAA,EAAO,kBAAA,GAAqB,IAAA;AAC5C,MAAA,GAAA,CAAI,CAAA,CAAE,QAAA,GAAA,iBAAY,KAAA,6BAAO,OAAA,GAAQ,KAAA,CAAA,CAAA,EAAS,kBAAA,GAAqB,IAAA;AAC/D,MAAA,IAAA,GAAO,GAAA,EAAK,QAAA,EAAU,SAAA,EAAW,UAAA;AAAA,IACnC;AAEA,IAAA,GAAA,CAAI,kBAAA,EAAoB,mBAAA,EAAqB;AAC3C,MAAA,MAAM,GAAA,EAAK,CAAC,CAAA,EAAA,GAAA,CAAe,EAAA,EAAI,KAAA,EAAO,IAAA,CAAA,CAAM,OAAA,CAAQ,CAAC,CAAA;AACrD,MAAA,MAAM,UAAA,EAAY,MAAA,GAAS,CAAC,KAAA,CAAM,KAAA,GAAQ,kBAAA,EAAoB,EAAA,EAC1D,CAAA,wDAAA,EAAgB,EAAA,CAAG,iBAAiB,CAAC,CAAA,oHAAA,EAAA,EAAqD,EAAA;AAC9F,MAAA,MAAM,IAAI,WAAA,CAAY,CAAA,kDAAA,EAAkB,EAAA,CAAG,iBAAiB,CAAC,CAAA,iBAAA,EAAU,oBAAA,EAAsB,KAAA,EAAO,IAAI,CAAA,GAAA,EAAM,SAAS,CAAA,CAAA;AACzH,IAAA;AAEuC,IAAA;AAC3B,EAAA;AAC0B,IAAA;AACO,IAAA;AAC/C,EAAA;AACF;AAS8C;AAC0C,EAAA;AACxF;AAGqB;AAC6G;AACxE;AAC9B,EAAA;AAC0B,EAAA;AAK+C,EAAA;AACrG;AAM+D;AACmB,EAAA;AACtC,EAAA;AAC5C;AAGmD;AAC8B,EAAA;AACjF;AACiG;AAiBlD;AACnC,EAAA;AACwD,EAAA;AAC3D,EAAA;AACT;AAG+C;AACnC,EAAA;AACwD,EAAA;AAC3D,EAAA;AACT;AAUE;AAGqD;AACjB,EAAA;AACpB,EAAA;AAEgB,EAAA;AAEA,EAAA;AAC0B,EAAA;AACV,EAAA;AAEmD,EAAA;AACG,EAAA;AACnB,EAAA;AAChD,EAAA;AAC8C,EAAA;AACrB,EAAA;AACrD,EAAA;AACT;AAIoD;AACqD,EAAA;AACzG;ADhI2K;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA","file":"/Users/mong-e/workspace/kordoc/dist/chunk-PNSALBXH.cjs","sourcesContent":[null,"/** kordoc 공용 유틸리티 */\n\nimport { format } from \"util\"\n\n/** 빌드 타임에 tsup define으로 주입되는 버전 */\ndeclare const __KORDOC_VERSION__: string\nexport const VERSION: string = typeof __KORDOC_VERSION__ !== \"undefined\" ? __KORDOC_VERSION__ : \"0.0.0-dev\"\n\n/**\n * Node.js Buffer → ArrayBuffer 변환\n * pool Buffer의 공유 ArrayBuffer 문제를 안전하게 처리.\n * offset=0이고 전체 ArrayBuffer를 차지하면 복사 없이 직접 반환.\n */\nexport function toArrayBuffer(buf: Buffer): ArrayBuffer {\n if (buf.byteOffset === 0 && buf.byteLength === buf.buffer.byteLength) {\n return buf.buffer as ArrayBuffer\n }\n return buf.buffer.slice(buf.byteOffset, buf.byteOffset + buf.byteLength) as ArrayBuffer\n}\n\n/**\n * kordoc 내부 에러 클래스 — 사용자에게 노출해도 안전한 메시지만 포함.\n * MCP 에러 정제에서 instanceof로 판별하여 allowlist 패턴 매칭 없이 안전하게 통과.\n */\nexport class KordocError extends Error {\n constructor(message: string) {\n super(message)\n this.name = \"KordocError\"\n }\n}\n\n/**\n * ENOENT 판별 — 입력 경로가 존재하지 않음.\n * 래핑된 KordocError 를 뚫고 올라온 경우도 있으므로 메시지까지 함께 본다.\n */\nfunction isFileNotFound(err: unknown): boolean {\n if (typeof err !== \"object\" || err === null) return false\n if ((err as NodeJS.ErrnoException).code === \"ENOENT\") return true\n const msg = (err as Error).message\n return typeof msg === \"string\" && /ENOENT|no such file or directory/i.test(msg)\n}\n\n/**\n * 에러 메시지 정제 — KordocError는 그대로, ENOENT 만 구분 문구, 나머지는 일반 메시지로 대체.\n * 파일시스템 경로, 스택 트레이스 등 내부 정보 노출 방지.\n */\nexport function sanitizeError(err: unknown): string {\n if (err instanceof KordocError) return err.message\n // 없는 파일은 문서 문제가 아니라 고칠 수 있는 입력 문제다 — 사유를 숨기지 않는다.\n // 경로는 노출하지 않는다(어느 파일인지는 stderr·실패 JSON 의 basename 이 알린다).\n // 문구는 MCP describeError 의 ENOENT 힌트와 같은 표현을 쓴다.\n if (isFileNotFound(err)) return \"파일 또는 디렉토리를 찾을 수 없습니다\"\n return \"문서 처리 중 오류가 발생했습니다\"\n}\n\n/**\n * ZIP 엔트리 경로의 경로 순회 여부 판별.\n * 백슬래시 정규화, .., 절대경로, Windows 드라이브 문자 모두 차단.\n */\nexport function isPathTraversal(name: string): boolean {\n if (name.includes(\"\\x00\")) return true\n const normalized = name.replace(/\\\\/g, \"/\")\n const segments = normalized.split(\"/\")\n return segments.some(s => s === \"..\") || normalized.startsWith(\"/\") || /^[A-Za-z]:/.test(normalized)\n}\n\n// ─── HWPX 섹션 href 해석 (parser/roundtrip 공용) ────────────────────\n\n/**\n * manifest href를 본문 섹션 경로(`Contents/sectionN.xml`)로 정규화.\n * 비본문 XML(header/script/settings)·path traversal·드라이브 경로는 null로 거른다.\n * 백슬래시 구분자를 슬래시로 통일해 parser와 roundtrip이 동일한 목록을 만든다.\n */\nexport function normalizeSectionHref(href: string): string | null {\n if (!href) return null\n let normalized = href.replace(/\\\\/g, \"/\").replace(/^\\/+/, \"\")\n if (isPathTraversal(normalized)) return null\n if (/^[Ss]ection\\d+\\.xml$/.test(normalized)) normalized = \"Contents/\" + normalized\n return /(?:^|\\/)[Ss]ection\\d+\\.xml$/.test(normalized) ? normalized : null\n}\n\n/** sectionN.xml을 N 숫자 순서로 정렬 (section10 > section2). */\nexport function compareSectionPaths(a: string, b: string): number {\n const ai = Number(a.match(/[Ss]ection(\\d+)\\.xml$/)?.[1] ?? Number.MAX_SAFE_INTEGER)\n const bi = Number(b.match(/[Ss]ection(\\d+)\\.xml$/)?.[1] ?? Number.MAX_SAFE_INTEGER)\n return ai === bi ? a.localeCompare(b) : ai - bi\n}\n\n// ─── ZIP 안전 로딩 (ZIP bomb 방지) ────────────────────\n\n/** ZIP 비압축 상한 env 의 최대값 (MB) — 오타로 사실상 무제한이 되지 않게 */\nconst MAX_UNZIP_ENV_MB = 8192\n\n/**\n * ZIP 비압축 크기 상한 — `KORDOC_MAX_UNZIP_MB` 가 양수면 그 값(MB, 최대 8192), 아니면 포맷 기본값.\n * 빈 행이 수백만 개인 XLSX 처럼 내용은 작고 XML 만 큰 실문서를 폐쇄망 대량 적재에서 받으려는 설정 (#91)\n */\nexport function unzipLimitBytes(defaultBytes: number): number {\n const mb = Number(process.env.KORDOC_MAX_UNZIP_MB)\n return Number.isFinite(mb) && mb > 0 ? Math.min(mb, MAX_UNZIP_ENV_MB) * 1024 * 1024 : defaultBytes\n}\n\n/**\n * ZIP bomb 사전 검사 — Central Directory에서 비압축 합계와 엔트리 수 확인.\n * HWPX/XLSX/DOCX 등 모든 ZIP 기반 포맷에서 공통 사용.\n */\nexport function precheckZipSize(\n buffer: ArrayBuffer,\n maxUncompressedSize = 256 * 1024 * 1024, // parser-shared MAX_DECOMPRESS_SIZE 와 동기\n maxEntries = 500,\n /** 그림·개체 파트(엔트리 이름 re) — skip 이면 합계에서 뺀다(파서가 풀지 않는다), 아니면 한도 초과 메시지에 크기를 적는다 (#108).\n * never — 파서가 어느 때도 풀지 않는 파트(동영상·음성·OLE 임베드)는 늘 뺀다: 110MB 동영상 하나로 발표 자료가 통째 거부됐다 */\n media?: { re: RegExp; skip: boolean; never?: RegExp },\n): { totalUncompressed: number; entryCount: number } {\n try {\n const data = new DataView(buffer)\n const len = buffer.byteLength\n // EOCD 시그니처 역방향 스캔\n let eocdOffset = -1\n for (let i = len - 22; i >= Math.max(0, len - 65557); i--) {\n if (data.getUint32(i, true) === 0x06054b50) { eocdOffset = i; break }\n }\n if (eocdOffset < 0) return { totalUncompressed: 0, entryCount: 0 }\n\n const entryCount = data.getUint16(eocdOffset + 10, true)\n if (entryCount > maxEntries) {\n throw new KordocError(`ZIP 엔트리 수 초과: ${entryCount} (최대 ${maxEntries})`)\n }\n\n const cdSize = data.getUint32(eocdOffset + 12, true)\n const cdOffset = data.getUint32(eocdOffset + 16, true)\n if (cdOffset + cdSize > len) return { totalUncompressed: 0, entryCount }\n\n let totalUncompressed = 0\n let mediaUncompressed = 0\n let pos = cdOffset\n for (let i = 0; i < entryCount && pos + 46 <= cdOffset + cdSize; i++) {\n if (data.getUint32(pos, true) !== 0x02014b50) break\n // 비압축 크기는 중앙 디렉터리와 지역 헤더 중 큰 값 — JSZip 은 지역 헤더 크기로 풀고(readZipEntry 가 그 크기를 넘으면 끊는다),\n // 한쪽만 작게 속여 합계 검사를 비켜 가지 못하게 둘 다 본다\n const localOff = data.getUint32(pos + 42, true)\n const localSize = localOff + 30 <= len && data.getUint32(localOff, true) === 0x04034b50 ? data.getUint32(localOff + 22, true) : 0\n const size = Math.max(data.getUint32(pos + 24, true), localSize)\n const nameLen = data.getUint16(pos + 28, true)\n const extraLen = data.getUint16(pos + 30, true)\n const commentLen = data.getUint16(pos + 32, true)\n // 이름이 버퍼 밖이면 그림 파트로 치지 않는다(합계에서 빠지지 않게)\n const isMedia = !!media && pos + 46 + nameLen <= len &&\n media.re.test(new TextDecoder().decode(new Uint8Array(buffer, pos + 46, nameLen)))\n const never = isMedia && !!media?.never && media.never.test(new TextDecoder().decode(new Uint8Array(buffer, pos + 46, nameLen)))\n if (isMedia && !never) mediaUncompressed += size\n if (!(isMedia && (media?.skip || never))) totalUncompressed += size\n pos += 46 + nameLen + extraLen + commentLen\n }\n\n if (totalUncompressed > maxUncompressedSize) {\n const mb = (n: number) => (n / 1024 / 1024).toFixed(1)\n const mediaNote = media && !media.skip && mediaUncompressed > 0\n ? ` — 그림·개체 파트가 ${mb(mediaUncompressed)}MB, 이미지 추출을 끄면(images: false·--no-images) 세지 않는다` : \"\"\n throw new KordocError(`ZIP 비압축 크기 초과: ${mb(totalUncompressed)}MB (최대 ${maxUncompressedSize / 1024 / 1024}MB)${mediaNote}`)\n }\n\n return { totalUncompressed, entryCount }\n } catch (err) {\n if (err instanceof KordocError) throw err\n return { totalUncompressed: 0, entryCount: 0 }\n }\n}\n\n/**\n * XML 파싱 전 프롤로그 정리 — 선두 BOM 제거 + DOCTYPE 제거(XXE/Billion Laughs 방지).\n *\n * BOM: 일부 OpenXML 라이터가 `[Content_Types].xml`·`.rels` 앞에 EF BB BF를 붙인다.\n * XML 스펙상 적법(엔티티는 BOM으로 시작할 수 있음)이고 엑셀·리브레오피스는 여는데,\n * xmldom은 \"선언이 문서 시작이 아님\" fatalError로 파일 전체를 거부했다 (#63).\n */\nexport function stripDtd(xml: string): string {\n return xml.replace(/^\\uFEFF/, \"\").replace(/<!DOCTYPE\\s[^[>]*(\\[[\\s\\S]*?\\])?\\s*>/gi, \"\")\n}\n\n/** 하이퍼링크 URL 살균 — javascript: 등 XSS 위험 스킴 차단 */\nconst SAFE_HREF_RE = /^(?:https?:|mailto:|tel:|#)/i\nconst HREF_ESCAPE: Record<string, string> = { \"(\": \"%28\", \")\": \"%29\", \"<\": \"%3C\", \">\": \"%3E\", \"\\\"\": \"%22\", \"'\": \"%27\", \"`\": \"%60\" }\nexport function sanitizeHref(href: string): string | null {\n const trimmed = href.trim()\n if (!trimmed || !SAFE_HREF_RE.test(trimmed)) return null\n // 괄호 percent-encoding — 마크다운 링크 목적지에서 불균형 ')'는 링크를 조기 종료시켜\n // 문법이 깨진다 (원문 하이퍼링크에 ')'가 박힌 실문서 존재). %28/%29는 의미 동일.\n // 공백·<>\"'` 도 — 스킴만 보고 통과시키면 \"mailto:a <img src=x onerror=…>\" 가 링크 목적지를 깨고 날 태그로\n // 렌더됐다(pdfjs 는 mailto·tel URI 를 정규화하지 않는다, v4.14.4 리뷰 재현). 인코딩해도 같은 주소다\n return trimmed.replace(/[()<>\"'`]/g, c => HREF_ESCAPE[c]).replace(/\\s/g, c => encodeURIComponent(c))\n}\n\n/**\n * HTML 엔티티 이스케이프 — 글은 `& < >`, 속성값(attr)은 `\"` 까지. HTML 표 경로(builder tableToHtml)와\n * 인쇄 렌더러가 같이 쓰는 한 벌. 읽는 쪽은 unescapeHtml 로 푼다\n */\nexport function escapeHtml(text: string, attr = false): string {\n const s = text.replace(/&/g, \"&amp;\").replace(/</g, \"&lt;\").replace(/>/g, \"&gt;\")\n return attr ? s.replace(/\"/g, \"&quot;\") : s\n}\n\n/** escapeHtml 의 역 (+ 손으로 쓴 HTML 표의 &#39;) — 한 번에 풀어 \"&amp;lt;\" 는 \"&lt;\" 로 남는다 */\nexport function unescapeHtml(text: string): string {\n return text.replace(/&(lt|gt|quot|#39|amp);/g, (_, e: string) => HTML_ENTITY[e])\n}\nconst HTML_ENTITY: Record<string, string> = { lt: \"<\", gt: \">\", quot: \"\\\"\", \"#39\": \"'\", amp: \"&\" }\n\n/**\n * console.log·info·warn·debug 를 stderr 로 — stdout 이 기계 출력 채널(MCP JSON-RPC·CLI 마크다운/JSON·parse-worker\n * NDJSON)인 진입점용. 라이브러리(pdfjs 등)가 console 로 찍는 경고가 그 채널을 깨지 않게 한다\n */\nexport function routeConsoleToStderr(): void {\n const toStderr = (...args: unknown[]): void => void process.stderr.write(format(...args) + \"\\n\")\n console.log = toStderr\n console.info = toStderr\n console.warn = toStderr\n console.debug = toStderr\n}\n\n// ─── 안전한 min/max (스택 오버플로 방지) ─────────────\n\n/** Math.min(...arr) 대체 — 대형 배열에서 스택 오버플로 방지 */\nexport function safeMin(arr: number[]): number {\n let min = Infinity\n for (let i = 0; i < arr.length; i++) if (arr[i] < min) min = arr[i]\n return min\n}\n\n/** Math.max(...arr) 대체 — 대형 배열에서 스택 오버플로 방지 */\nexport function safeMax(arr: number[]): number {\n let max = -Infinity\n for (let i = 0; i < arr.length; i++) if (arr[i] > max) max = arr[i]\n return max\n}\n\n// ─── 에러 분류 ──────────────────────────────────────\n\nimport type { ErrorCode } from \"./types.js\"\n\n/** 에러를 구조화된 ErrorCode로 분류 — KordocError 메시지 패턴 매칭 */\n/** 선택 의존 설치 실패 안내 꼬리 — 네트워크가 막힌 linux/x64 에서 onnxruntime-node 설치 스크립트가 CUDA 바이너리를 받다 실패하면\n * npm 이 sharp·@huggingface/tokenizers 까지 선택 의존 설치에서 뺀다(#99). npx 캐시 실행은 npm install 로 못 고친다 */\nexport const OPTIONAL_DEP_INSTALL_HINT =\n \" 네트워크가 제한된 linux/x64 에서 npx 로 설치했다면 onnxruntime-node 의 CUDA 다운로드 실패로 함께 빠졌을 수 있습니다 — \" +\n \"`ONNXRUNTIME_NODE_INSTALL=skip npx -y kordoc@^4 …` 로 다시 설치하세요.\"\n\nexport function classifyError(err: unknown): ErrorCode {\n if (!(err instanceof Error)) return \"PARSE_ERROR\"\n const msg = err.message\n // 입력 파일 없음 — 문서 내용 문제가 아니라 호출자의 경로 문제라 PARSE_ERROR 로 덮지 않는다.\n if (isFileNotFound(err)) return \"FILE_NOT_FOUND\"\n // DRM 을 암호화보다 먼저 — \"DRM 암호화된 HWPX…\" 류 메시지가 ENCRYPTED 로 떨어지지 않도록\n if (msg.includes(\"DRM\")) return \"DRM_PROTECTED\"\n if (msg.includes(\"암호화\") || msg.includes(\"암호로 보호\")) return \"ENCRYPTED\"\n if (msg.includes(\"optional dependency\")) return \"MISSING_DEPENDENCY\"\n // V8 문자열 한계 초과 — 이미지 base64 총량이 큰 문서의 JSON 직렬화 (#65)\n if (msg.includes(\"Invalid string length\") || msg.includes(\"Cannot create a string longer\")) return \"OUTPUT_TOO_LARGE\"\n if (msg.includes(\"ZIP bomb\") || msg.includes(\"ZIP 비압축 크기 초과\") || msg.includes(\"ZIP 엔트리 수 초과\")) return \"ZIP_BOMB\"\n if (msg.includes(\"bomb\") || msg.includes(\"크기 초과\") || msg.includes(\"압축 해제\")) return \"DECOMPRESSION_BOMB\"\n if (msg.includes(\"이미지 기반\")) return \"IMAGE_BASED_PDF\"\n if (msg.includes(\"섹션\") && (msg.includes(\"찾을 수 없\") || msg.includes(\"없음\"))) return \"NO_SECTIONS\"\n if (msg.includes(\"시그니처\") || msg.includes(\"복구할 수 없\")) return \"CORRUPTED\"\n return \"PARSE_ERROR\"\n}\n\n/** ZIP 파트 경로의 확장자 — 마지막 경로 조각에서만, 없으면 \"bin\". \"ppt/media.v2/image1\" 의 \"v2/image1\" 이 그림 파일 이름에 들어가\n * 저장 경로가 깨졌다 */\nexport function partExtension(path: string): string {\n return (/\\.([A-Za-z0-9]{1,5})$/.exec(path.slice(path.lastIndexOf(\"/\") + 1))?.[1] ?? \"bin\").toLowerCase()\n}\n"]}
1
+ {"version":3,"sources":["/Users/mong-e/workspace/kordoc/dist/chunk-6ZRZLV3J.cjs","../src/utils.ts"],"names":[],"mappings":"AAAA;ACMO,IAAM,QAAA,EAAkB,KAAA,EAA4C,SAAA,EAAqB,WAAA;AAOzF,SAAS,aAAA,CAAc,GAAA,EAA0B;AACtD,EAAA,GAAA,CAAI,GAAA,CAAI,WAAA,IAAe,EAAA,GAAK,GAAA,CAAI,WAAA,IAAe,GAAA,CAAI,MAAA,CAAO,UAAA,EAAY;AACpE,IAAA,OAAO,GAAA,CAAI,MAAA;AAAA,EACb;AACA,EAAA,OAAO,GAAA,CAAI,MAAA,CAAO,KAAA,CAAM,GAAA,CAAI,UAAA,EAAY,GAAA,CAAI,WAAA,EAAa,GAAA,CAAI,UAAU,CAAA;AACzE;AAMO,IAAM,YAAA,EAAN,MAAA,QAA0B,MAAM;AAAA,EACrC,WAAA,CAAY,OAAA,EAAiB;AAC3B,IAAA,KAAA,CAAM,OAAO,CAAA;AACb,IAAA,IAAA,CAAK,KAAA,EAAO,aAAA;AAAA,EACd;AACF,CAAA;AAMA,SAAS,cAAA,CAAe,GAAA,EAAuB;AAC7C,EAAA,GAAA,CAAI,OAAO,IAAA,IAAQ,SAAA,GAAY,IAAA,IAAQ,IAAA,EAAM,OAAO,KAAA;AACpD,EAAA,GAAA,CAAK,GAAA,CAA8B,KAAA,IAAS,QAAA,EAAU,OAAO,IAAA;AAC7D,EAAA,MAAM,IAAA,EAAO,GAAA,CAAc,OAAA;AAC3B,EAAA,OAAO,OAAO,IAAA,IAAQ,SAAA,GAAY,mCAAA,CAAoC,IAAA,CAAK,GAAG,CAAA;AAChF;AAMO,SAAS,aAAA,CAAc,GAAA,EAAsB;AAClD,EAAA,GAAA,CAAI,IAAA,WAAe,WAAA,EAAa,OAAO,GAAA,CAAI,OAAA;AAI3C,EAAA,GAAA,CAAI,cAAA,CAAe,GAAG,CAAA,EAAG,OAAO,uGAAA;AAChC,EAAA,OAAO,0FAAA;AACT;AAMO,SAAS,eAAA,CAAgB,IAAA,EAAuB;AACrD,EAAA,GAAA,CAAI,IAAA,CAAK,QAAA,CAAS,IAAM,CAAA,EAAG,OAAO,IAAA;AAClC,EAAA,MAAM,WAAA,EAAa,IAAA,CAAK,OAAA,CAAQ,KAAA,EAAO,GAAG,CAAA;AAC1C,EAAA,MAAM,SAAA,EAAW,UAAA,CAAW,KAAA,CAAM,GAAG,CAAA;AACrC,EAAA,OAAO,QAAA,CAAS,IAAA,CAAK,CAAA,CAAA,EAAA,GAAK,EAAA,IAAM,IAAI,EAAA,GAAK,UAAA,CAAW,UAAA,CAAW,GAAG,EAAA,GAAK,YAAA,CAAa,IAAA,CAAK,UAAU,CAAA;AACrG;AASO,SAAS,oBAAA,CAAqB,IAAA,EAA6B;AAChE,EAAA,GAAA,CAAI,CAAC,IAAA,EAAM,OAAO,IAAA;AAClB,EAAA,IAAI,WAAA,EAAa,IAAA,CAAK,OAAA,CAAQ,KAAA,EAAO,GAAG,CAAA,CAAE,OAAA,CAAQ,MAAA,EAAQ,EAAE,CAAA;AAC5D,EAAA,GAAA,CAAI,eAAA,CAAgB,UAAU,CAAA,EAAG,OAAO,IAAA;AACxC,EAAA,GAAA,CAAI,sBAAA,CAAuB,IAAA,CAAK,UAAU,CAAA,EAAG,WAAA,EAAa,YAAA,EAAc,UAAA;AACxE,EAAA,OAAO,6BAAA,CAA8B,IAAA,CAAK,UAAU,EAAA,EAAI,WAAA,EAAa,IAAA;AACvE;AAGO,SAAS,mBAAA,CAAoB,CAAA,EAAW,CAAA,EAAmB;AAChE,EAAA,MAAM,GAAA,EAAK,MAAA,kCAAO,CAAA,qBAAE,KAAA,mBAAM,uBAAuB,CAAA,4BAAA,CAAI,CAAC,GAAA,UAAK,MAAA,CAAO,kBAAgB,CAAA;AAClF,EAAA,MAAM,GAAA,EAAK,MAAA,kCAAO,CAAA,qBAAE,KAAA,mBAAM,uBAAuB,CAAA,4BAAA,CAAI,CAAC,GAAA,UAAK,MAAA,CAAO,kBAAgB,CAAA;AAClF,EAAA,OAAO,GAAA,IAAO,GAAA,EAAK,CAAA,CAAE,aAAA,CAAc,CAAC,EAAA,EAAI,GAAA,EAAK,EAAA;AAC/C;AAKA,IAAM,iBAAA,EAAmB,IAAA;AAMlB,SAAS,eAAA,CAAgB,YAAA,EAA8B;AAC5D,EAAA,MAAM,GAAA,EAAK,MAAA,CAAO,OAAA,CAAQ,GAAA,CAAI,mBAAmB,CAAA;AACjD,EAAA,OAAO,MAAA,CAAO,QAAA,CAAS,EAAE,EAAA,GAAK,GAAA,EAAK,EAAA,EAAI,IAAA,CAAK,GAAA,CAAI,EAAA,EAAI,gBAAgB,EAAA,EAAI,KAAA,EAAO,KAAA,EAAO,YAAA;AACxF;AAMO,SAAS,eAAA,CACd,MAAA,EACA,oBAAA,EAAsB,IAAA,EAAM,KAAA,EAAO,IAAA,EACnC,WAAA,EAAa,GAAA,EAGb,KAAA,EACmD;AACnD,EAAA,IAAI;AACF,IAAA,MAAM,KAAA,EAAO,IAAI,QAAA,CAAS,MAAM,CAAA;AAChC,IAAA,MAAM,IAAA,EAAM,MAAA,CAAO,UAAA;AAEnB,IAAA,IAAI,WAAA,EAAa,CAAA,CAAA;AACjB,IAAA,IAAA,CAAA,IAAS,EAAA,EAAI,IAAA,EAAM,EAAA,EAAI,EAAA,GAAK,IAAA,CAAK,GAAA,CAAI,CAAA,EAAG,IAAA,EAAM,KAAK,CAAA,EAAG,CAAA,EAAA,EAAK;AACzD,MAAA,GAAA,CAAI,IAAA,CAAK,SAAA,CAAU,CAAA,EAAG,IAAI,EAAA,IAAM,SAAA,EAAY;AAAE,QAAA,WAAA,EAAa,CAAA;AAAG,QAAA,KAAA;AAAA,MAAM;AAAA,IACtE;AACA,IAAA,GAAA,CAAI,WAAA,EAAa,CAAA,EAAG,OAAO,EAAE,iBAAA,EAAmB,CAAA,EAAG,UAAA,EAAY,EAAE,CAAA;AAEjE,IAAA,MAAM,WAAA,EAAa,IAAA,CAAK,SAAA,CAAU,WAAA,EAAa,EAAA,EAAI,IAAI,CAAA;AACvD,IAAA,GAAA,CAAI,WAAA,EAAa,UAAA,EAAY;AAC3B,MAAA,MAAM,IAAI,WAAA,CAAY,CAAA,4CAAA,EAAiB,UAAU,CAAA,eAAA,EAAQ,UAAU,CAAA,CAAA,CAAG,CAAA;AAAA,IACxE;AAEA,IAAA,MAAM,OAAA,EAAS,IAAA,CAAK,SAAA,CAAU,WAAA,EAAa,EAAA,EAAI,IAAI,CAAA;AACnD,IAAA,MAAM,SAAA,EAAW,IAAA,CAAK,SAAA,CAAU,WAAA,EAAa,EAAA,EAAI,IAAI,CAAA;AACrD,IAAA,GAAA,CAAI,SAAA,EAAW,OAAA,EAAS,GAAA,EAAK,OAAO,EAAE,iBAAA,EAAmB,CAAA,EAAG,WAAW,CAAA;AAEvE,IAAA,IAAI,kBAAA,EAAoB,CAAA;AACxB,IAAA,IAAI,kBAAA,EAAoB,CAAA;AACxB,IAAA,IAAI,IAAA,EAAM,QAAA;AACV,IAAA,IAAA,CAAA,IAAS,EAAA,EAAI,CAAA,EAAG,EAAA,EAAI,WAAA,GAAc,IAAA,EAAM,GAAA,GAAM,SAAA,EAAW,MAAA,EAAQ,CAAA,EAAA,EAAK;AACpE,MAAA,GAAA,CAAI,IAAA,CAAK,SAAA,CAAU,GAAA,EAAK,IAAI,EAAA,IAAM,QAAA,EAAY,KAAA;AAG9C,MAAA,MAAM,SAAA,EAAW,IAAA,CAAK,SAAA,CAAU,IAAA,EAAM,EAAA,EAAI,IAAI,CAAA;AAC9C,MAAA,MAAM,UAAA,EAAY,SAAA,EAAW,GAAA,GAAM,IAAA,GAAO,IAAA,CAAK,SAAA,CAAU,QAAA,EAAU,IAAI,EAAA,IAAM,SAAA,EAAa,IAAA,CAAK,SAAA,CAAU,SAAA,EAAW,EAAA,EAAI,IAAI,EAAA,EAAI,CAAA;AAChI,MAAA,MAAM,KAAA,EAAO,IAAA,CAAK,GAAA,CAAI,IAAA,CAAK,SAAA,CAAU,IAAA,EAAM,EAAA,EAAI,IAAI,CAAA,EAAG,SAAS,CAAA;AAC/D,MAAA,MAAM,QAAA,EAAU,IAAA,CAAK,SAAA,CAAU,IAAA,EAAM,EAAA,EAAI,IAAI,CAAA;AAC7C,MAAA,MAAM,SAAA,EAAW,IAAA,CAAK,SAAA,CAAU,IAAA,EAAM,EAAA,EAAI,IAAI,CAAA;AAC9C,MAAA,MAAM,WAAA,EAAa,IAAA,CAAK,SAAA,CAAU,IAAA,EAAM,EAAA,EAAI,IAAI,CAAA;AAEhD,MAAA,MAAM,QAAA,EAAU,CAAC,CAAC,MAAA,GAAS,IAAA,EAAM,GAAA,EAAK,QAAA,GAAW,IAAA,GAC/C,KAAA,CAAM,EAAA,CAAG,IAAA,CAAK,IAAI,WAAA,CAAY,CAAA,CAAE,MAAA,CAAO,IAAI,UAAA,CAAW,MAAA,EAAQ,IAAA,EAAM,EAAA,EAAI,OAAO,CAAC,CAAC,CAAA;AACnF,MAAA,MAAM,MAAA,EAAQ,QAAA,GAAW,CAAC,iBAAC,KAAA,6BAAO,QAAA,GAAS,KAAA,CAAM,KAAA,CAAM,IAAA,CAAK,IAAI,WAAA,CAAY,CAAA,CAAE,MAAA,CAAO,IAAI,UAAA,CAAW,MAAA,EAAQ,IAAA,EAAM,EAAA,EAAI,OAAO,CAAC,CAAC,CAAA;AAC/H,MAAA,GAAA,CAAI,QAAA,GAAW,CAAC,KAAA,EAAO,kBAAA,GAAqB,IAAA;AAC5C,MAAA,GAAA,CAAI,CAAA,CAAE,QAAA,GAAA,iBAAY,KAAA,6BAAO,OAAA,GAAQ,KAAA,CAAA,CAAA,EAAS,kBAAA,GAAqB,IAAA;AAC/D,MAAA,IAAA,GAAO,GAAA,EAAK,QAAA,EAAU,SAAA,EAAW,UAAA;AAAA,IACnC;AAEA,IAAA,GAAA,CAAI,kBAAA,EAAoB,mBAAA,EAAqB;AAC3C,MAAA,MAAM,GAAA,EAAK,CAAC,CAAA,EAAA,GAAA,CAAe,EAAA,EAAI,KAAA,EAAO,IAAA,CAAA,CAAM,OAAA,CAAQ,CAAC,CAAA;AACrD,MAAA,MAAM,UAAA,EAAY,MAAA,GAAS,CAAC,KAAA,CAAM,KAAA,GAAQ,kBAAA,EAAoB,EAAA,EAC1D,CAAA,wDAAA,EAAgB,EAAA,CAAG,iBAAiB,CAAC,CAAA,oHAAA,EAAA,EAAqD,EAAA;AAC9F,MAAA,MAAM,IAAI,WAAA,CAAY,CAAA,kDAAA,EAAkB,EAAA,CAAG,iBAAiB,CAAC,CAAA,iBAAA,EAAU,oBAAA,EAAsB,KAAA,EAAO,IAAI,CAAA,GAAA,EAAM,SAAS,CAAA,CAAA;AACzH,IAAA;AAEuC,IAAA;AAC3B,EAAA;AAC0B,IAAA;AACO,IAAA;AAC/C,EAAA;AACF;AAS8C;AAC0C,EAAA;AACxF;AAGqB;AAC6G;AACxE;AAC9B,EAAA;AAC0B,EAAA;AAK+C,EAAA;AACrG;AAM+D;AACmB,EAAA;AACtC,EAAA;AAC5C;AAGmD;AAC8B,EAAA;AACjF;AACiG;AAiBlD;AACnC,EAAA;AACwD,EAAA;AAC3D,EAAA;AACT;AAG+C;AACnC,EAAA;AACwD,EAAA;AAC3D,EAAA;AACT;AAUE;AAGqD;AACjB,EAAA;AACpB,EAAA;AAEgB,EAAA;AAEA,EAAA;AAC0B,EAAA;AACV,EAAA;AAEmD,EAAA;AACG,EAAA;AACnB,EAAA;AAChD,EAAA;AAC8C,EAAA;AACrB,EAAA;AACrD,EAAA;AACT;AAIoD;AACqD,EAAA;AACzG;ADhI2K;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA","file":"/Users/mong-e/workspace/kordoc/dist/chunk-6ZRZLV3J.cjs","sourcesContent":[null,"/** kordoc 공용 유틸리티 */\n\nimport { format } from \"util\"\n\n/** 빌드 타임에 tsup define으로 주입되는 버전 */\ndeclare const __KORDOC_VERSION__: string\nexport const VERSION: string = typeof __KORDOC_VERSION__ !== \"undefined\" ? __KORDOC_VERSION__ : \"0.0.0-dev\"\n\n/**\n * Node.js Buffer → ArrayBuffer 변환\n * pool Buffer의 공유 ArrayBuffer 문제를 안전하게 처리.\n * offset=0이고 전체 ArrayBuffer를 차지하면 복사 없이 직접 반환.\n */\nexport function toArrayBuffer(buf: Buffer): ArrayBuffer {\n if (buf.byteOffset === 0 && buf.byteLength === buf.buffer.byteLength) {\n return buf.buffer as ArrayBuffer\n }\n return buf.buffer.slice(buf.byteOffset, buf.byteOffset + buf.byteLength) as ArrayBuffer\n}\n\n/**\n * kordoc 내부 에러 클래스 — 사용자에게 노출해도 안전한 메시지만 포함.\n * MCP 에러 정제에서 instanceof로 판별하여 allowlist 패턴 매칭 없이 안전하게 통과.\n */\nexport class KordocError extends Error {\n constructor(message: string) {\n super(message)\n this.name = \"KordocError\"\n }\n}\n\n/**\n * ENOENT 판별 — 입력 경로가 존재하지 않음.\n * 래핑된 KordocError 를 뚫고 올라온 경우도 있으므로 메시지까지 함께 본다.\n */\nfunction isFileNotFound(err: unknown): boolean {\n if (typeof err !== \"object\" || err === null) return false\n if ((err as NodeJS.ErrnoException).code === \"ENOENT\") return true\n const msg = (err as Error).message\n return typeof msg === \"string\" && /ENOENT|no such file or directory/i.test(msg)\n}\n\n/**\n * 에러 메시지 정제 — KordocError는 그대로, ENOENT 만 구분 문구, 나머지는 일반 메시지로 대체.\n * 파일시스템 경로, 스택 트레이스 등 내부 정보 노출 방지.\n */\nexport function sanitizeError(err: unknown): string {\n if (err instanceof KordocError) return err.message\n // 없는 파일은 문서 문제가 아니라 고칠 수 있는 입력 문제다 — 사유를 숨기지 않는다.\n // 경로는 노출하지 않는다(어느 파일인지는 stderr·실패 JSON 의 basename 이 알린다).\n // 문구는 MCP describeError 의 ENOENT 힌트와 같은 표현을 쓴다.\n if (isFileNotFound(err)) return \"파일 또는 디렉토리를 찾을 수 없습니다\"\n return \"문서 처리 중 오류가 발생했습니다\"\n}\n\n/**\n * ZIP 엔트리 경로의 경로 순회 여부 판별.\n * 백슬래시 정규화, .., 절대경로, Windows 드라이브 문자 모두 차단.\n */\nexport function isPathTraversal(name: string): boolean {\n if (name.includes(\"\\x00\")) return true\n const normalized = name.replace(/\\\\/g, \"/\")\n const segments = normalized.split(\"/\")\n return segments.some(s => s === \"..\") || normalized.startsWith(\"/\") || /^[A-Za-z]:/.test(normalized)\n}\n\n// ─── HWPX 섹션 href 해석 (parser/roundtrip 공용) ────────────────────\n\n/**\n * manifest href를 본문 섹션 경로(`Contents/sectionN.xml`)로 정규화.\n * 비본문 XML(header/script/settings)·path traversal·드라이브 경로는 null로 거른다.\n * 백슬래시 구분자를 슬래시로 통일해 parser와 roundtrip이 동일한 목록을 만든다.\n */\nexport function normalizeSectionHref(href: string): string | null {\n if (!href) return null\n let normalized = href.replace(/\\\\/g, \"/\").replace(/^\\/+/, \"\")\n if (isPathTraversal(normalized)) return null\n if (/^[Ss]ection\\d+\\.xml$/.test(normalized)) normalized = \"Contents/\" + normalized\n return /(?:^|\\/)[Ss]ection\\d+\\.xml$/.test(normalized) ? normalized : null\n}\n\n/** sectionN.xml을 N 숫자 순서로 정렬 (section10 > section2). */\nexport function compareSectionPaths(a: string, b: string): number {\n const ai = Number(a.match(/[Ss]ection(\\d+)\\.xml$/)?.[1] ?? Number.MAX_SAFE_INTEGER)\n const bi = Number(b.match(/[Ss]ection(\\d+)\\.xml$/)?.[1] ?? Number.MAX_SAFE_INTEGER)\n return ai === bi ? a.localeCompare(b) : ai - bi\n}\n\n// ─── ZIP 안전 로딩 (ZIP bomb 방지) ────────────────────\n\n/** ZIP 비압축 상한 env 의 최대값 (MB) — 오타로 사실상 무제한이 되지 않게 */\nconst MAX_UNZIP_ENV_MB = 8192\n\n/**\n * ZIP 비압축 크기 상한 — `KORDOC_MAX_UNZIP_MB` 가 양수면 그 값(MB, 최대 8192), 아니면 포맷 기본값.\n * 빈 행이 수백만 개인 XLSX 처럼 내용은 작고 XML 만 큰 실문서를 폐쇄망 대량 적재에서 받으려는 설정 (#91)\n */\nexport function unzipLimitBytes(defaultBytes: number): number {\n const mb = Number(process.env.KORDOC_MAX_UNZIP_MB)\n return Number.isFinite(mb) && mb > 0 ? Math.min(mb, MAX_UNZIP_ENV_MB) * 1024 * 1024 : defaultBytes\n}\n\n/**\n * ZIP bomb 사전 검사 — Central Directory에서 비압축 합계와 엔트리 수 확인.\n * HWPX/XLSX/DOCX 등 모든 ZIP 기반 포맷에서 공통 사용.\n */\nexport function precheckZipSize(\n buffer: ArrayBuffer,\n maxUncompressedSize = 256 * 1024 * 1024, // parser-shared MAX_DECOMPRESS_SIZE 와 동기\n maxEntries = 500,\n /** 그림·개체 파트(엔트리 이름 re) — skip 이면 합계에서 뺀다(파서가 풀지 않는다), 아니면 한도 초과 메시지에 크기를 적는다 (#108).\n * never — 파서가 어느 때도 풀지 않는 파트(동영상·음성·OLE 임베드)는 늘 뺀다: 110MB 동영상 하나로 발표 자료가 통째 거부됐다 */\n media?: { re: RegExp; skip: boolean; never?: RegExp },\n): { totalUncompressed: number; entryCount: number } {\n try {\n const data = new DataView(buffer)\n const len = buffer.byteLength\n // EOCD 시그니처 역방향 스캔\n let eocdOffset = -1\n for (let i = len - 22; i >= Math.max(0, len - 65557); i--) {\n if (data.getUint32(i, true) === 0x06054b50) { eocdOffset = i; break }\n }\n if (eocdOffset < 0) return { totalUncompressed: 0, entryCount: 0 }\n\n const entryCount = data.getUint16(eocdOffset + 10, true)\n if (entryCount > maxEntries) {\n throw new KordocError(`ZIP 엔트리 수 초과: ${entryCount} (최대 ${maxEntries})`)\n }\n\n const cdSize = data.getUint32(eocdOffset + 12, true)\n const cdOffset = data.getUint32(eocdOffset + 16, true)\n if (cdOffset + cdSize > len) return { totalUncompressed: 0, entryCount }\n\n let totalUncompressed = 0\n let mediaUncompressed = 0\n let pos = cdOffset\n for (let i = 0; i < entryCount && pos + 46 <= cdOffset + cdSize; i++) {\n if (data.getUint32(pos, true) !== 0x02014b50) break\n // 비압축 크기는 중앙 디렉터리와 지역 헤더 중 큰 값 — JSZip 은 지역 헤더 크기로 풀고(readZipEntry 가 그 크기를 넘으면 끊는다),\n // 한쪽만 작게 속여 합계 검사를 비켜 가지 못하게 둘 다 본다\n const localOff = data.getUint32(pos + 42, true)\n const localSize = localOff + 30 <= len && data.getUint32(localOff, true) === 0x04034b50 ? data.getUint32(localOff + 22, true) : 0\n const size = Math.max(data.getUint32(pos + 24, true), localSize)\n const nameLen = data.getUint16(pos + 28, true)\n const extraLen = data.getUint16(pos + 30, true)\n const commentLen = data.getUint16(pos + 32, true)\n // 이름이 버퍼 밖이면 그림 파트로 치지 않는다(합계에서 빠지지 않게)\n const isMedia = !!media && pos + 46 + nameLen <= len &&\n media.re.test(new TextDecoder().decode(new Uint8Array(buffer, pos + 46, nameLen)))\n const never = isMedia && !!media?.never && media.never.test(new TextDecoder().decode(new Uint8Array(buffer, pos + 46, nameLen)))\n if (isMedia && !never) mediaUncompressed += size\n if (!(isMedia && (media?.skip || never))) totalUncompressed += size\n pos += 46 + nameLen + extraLen + commentLen\n }\n\n if (totalUncompressed > maxUncompressedSize) {\n const mb = (n: number) => (n / 1024 / 1024).toFixed(1)\n const mediaNote = media && !media.skip && mediaUncompressed > 0\n ? ` — 그림·개체 파트가 ${mb(mediaUncompressed)}MB, 이미지 추출을 끄면(images: false·--no-images) 세지 않는다` : \"\"\n throw new KordocError(`ZIP 비압축 크기 초과: ${mb(totalUncompressed)}MB (최대 ${maxUncompressedSize / 1024 / 1024}MB)${mediaNote}`)\n }\n\n return { totalUncompressed, entryCount }\n } catch (err) {\n if (err instanceof KordocError) throw err\n return { totalUncompressed: 0, entryCount: 0 }\n }\n}\n\n/**\n * XML 파싱 전 프롤로그 정리 — 선두 BOM 제거 + DOCTYPE 제거(XXE/Billion Laughs 방지).\n *\n * BOM: 일부 OpenXML 라이터가 `[Content_Types].xml`·`.rels` 앞에 EF BB BF를 붙인다.\n * XML 스펙상 적법(엔티티는 BOM으로 시작할 수 있음)이고 엑셀·리브레오피스는 여는데,\n * xmldom은 \"선언이 문서 시작이 아님\" fatalError로 파일 전체를 거부했다 (#63).\n */\nexport function stripDtd(xml: string): string {\n return xml.replace(/^\\uFEFF/, \"\").replace(/<!DOCTYPE\\s[^[>]*(\\[[\\s\\S]*?\\])?\\s*>/gi, \"\")\n}\n\n/** 하이퍼링크 URL 살균 — javascript: 등 XSS 위험 스킴 차단 */\nconst SAFE_HREF_RE = /^(?:https?:|mailto:|tel:|#)/i\nconst HREF_ESCAPE: Record<string, string> = { \"(\": \"%28\", \")\": \"%29\", \"<\": \"%3C\", \">\": \"%3E\", \"\\\"\": \"%22\", \"'\": \"%27\", \"`\": \"%60\" }\nexport function sanitizeHref(href: string): string | null {\n const trimmed = href.trim()\n if (!trimmed || !SAFE_HREF_RE.test(trimmed)) return null\n // 괄호 percent-encoding — 마크다운 링크 목적지에서 불균형 ')'는 링크를 조기 종료시켜\n // 문법이 깨진다 (원문 하이퍼링크에 ')'가 박힌 실문서 존재). %28/%29는 의미 동일.\n // 공백·<>\"'` 도 — 스킴만 보고 통과시키면 \"mailto:a <img src=x onerror=…>\" 가 링크 목적지를 깨고 날 태그로\n // 렌더됐다(pdfjs 는 mailto·tel URI 를 정규화하지 않는다, v4.14.4 리뷰 재현). 인코딩해도 같은 주소다\n return trimmed.replace(/[()<>\"'`]/g, c => HREF_ESCAPE[c]).replace(/\\s/g, c => encodeURIComponent(c))\n}\n\n/**\n * HTML 엔티티 이스케이프 — 글은 `& < >`, 속성값(attr)은 `\"` 까지. HTML 표 경로(builder tableToHtml)와\n * 인쇄 렌더러가 같이 쓰는 한 벌. 읽는 쪽은 unescapeHtml 로 푼다\n */\nexport function escapeHtml(text: string, attr = false): string {\n const s = text.replace(/&/g, \"&amp;\").replace(/</g, \"&lt;\").replace(/>/g, \"&gt;\")\n return attr ? s.replace(/\"/g, \"&quot;\") : s\n}\n\n/** escapeHtml 의 역 (+ 손으로 쓴 HTML 표의 &#39;) — 한 번에 풀어 \"&amp;lt;\" 는 \"&lt;\" 로 남는다 */\nexport function unescapeHtml(text: string): string {\n return text.replace(/&(lt|gt|quot|#39|amp);/g, (_, e: string) => HTML_ENTITY[e])\n}\nconst HTML_ENTITY: Record<string, string> = { lt: \"<\", gt: \">\", quot: \"\\\"\", \"#39\": \"'\", amp: \"&\" }\n\n/**\n * console.log·info·warn·debug 를 stderr 로 — stdout 이 기계 출력 채널(MCP JSON-RPC·CLI 마크다운/JSON·parse-worker\n * NDJSON)인 진입점용. 라이브러리(pdfjs 등)가 console 로 찍는 경고가 그 채널을 깨지 않게 한다\n */\nexport function routeConsoleToStderr(): void {\n const toStderr = (...args: unknown[]): void => void process.stderr.write(format(...args) + \"\\n\")\n console.log = toStderr\n console.info = toStderr\n console.warn = toStderr\n console.debug = toStderr\n}\n\n// ─── 안전한 min/max (스택 오버플로 방지) ─────────────\n\n/** Math.min(...arr) 대체 — 대형 배열에서 스택 오버플로 방지 */\nexport function safeMin(arr: number[]): number {\n let min = Infinity\n for (let i = 0; i < arr.length; i++) if (arr[i] < min) min = arr[i]\n return min\n}\n\n/** Math.max(...arr) 대체 — 대형 배열에서 스택 오버플로 방지 */\nexport function safeMax(arr: number[]): number {\n let max = -Infinity\n for (let i = 0; i < arr.length; i++) if (arr[i] > max) max = arr[i]\n return max\n}\n\n// ─── 에러 분류 ──────────────────────────────────────\n\nimport type { ErrorCode } from \"./types.js\"\n\n/** 에러를 구조화된 ErrorCode로 분류 — KordocError 메시지 패턴 매칭 */\n/** 선택 의존 설치 실패 안내 꼬리 — 네트워크가 막힌 linux/x64 에서 onnxruntime-node 설치 스크립트가 CUDA 바이너리를 받다 실패하면\n * npm 이 sharp·@huggingface/tokenizers 까지 선택 의존 설치에서 뺀다(#99). npx 캐시 실행은 npm install 로 못 고친다 */\nexport const OPTIONAL_DEP_INSTALL_HINT =\n \" 네트워크가 제한된 linux/x64 에서 npx 로 설치했다면 onnxruntime-node 의 CUDA 다운로드 실패로 함께 빠졌을 수 있습니다 — \" +\n \"`ONNXRUNTIME_NODE_INSTALL=skip npx -y kordoc@^4 …` 로 다시 설치하세요.\"\n\nexport function classifyError(err: unknown): ErrorCode {\n if (!(err instanceof Error)) return \"PARSE_ERROR\"\n const msg = err.message\n // 입력 파일 없음 — 문서 내용 문제가 아니라 호출자의 경로 문제라 PARSE_ERROR 로 덮지 않는다.\n if (isFileNotFound(err)) return \"FILE_NOT_FOUND\"\n // DRM 을 암호화보다 먼저 — \"DRM 암호화된 HWPX…\" 류 메시지가 ENCRYPTED 로 떨어지지 않도록\n if (msg.includes(\"DRM\")) return \"DRM_PROTECTED\"\n if (msg.includes(\"암호화\") || msg.includes(\"암호로 보호\")) return \"ENCRYPTED\"\n if (msg.includes(\"optional dependency\")) return \"MISSING_DEPENDENCY\"\n // V8 문자열 한계 초과 — 이미지 base64 총량이 큰 문서의 JSON 직렬화 (#65)\n if (msg.includes(\"Invalid string length\") || msg.includes(\"Cannot create a string longer\")) return \"OUTPUT_TOO_LARGE\"\n if (msg.includes(\"ZIP bomb\") || msg.includes(\"ZIP 비압축 크기 초과\") || msg.includes(\"ZIP 엔트리 수 초과\")) return \"ZIP_BOMB\"\n if (msg.includes(\"bomb\") || msg.includes(\"크기 초과\") || msg.includes(\"압축 해제\")) return \"DECOMPRESSION_BOMB\"\n if (msg.includes(\"이미지 기반\")) return \"IMAGE_BASED_PDF\"\n if (msg.includes(\"섹션\") && (msg.includes(\"찾을 수 없\") || msg.includes(\"없음\"))) return \"NO_SECTIONS\"\n if (msg.includes(\"시그니처\") || msg.includes(\"복구할 수 없\")) return \"CORRUPTED\"\n return \"PARSE_ERROR\"\n}\n\n/** ZIP 파트 경로의 확장자 — 마지막 경로 조각에서만, 없으면 \"bin\". \"ppt/media.v2/image1\" 의 \"v2/image1\" 이 그림 파일 이름에 들어가\n * 저장 경로가 깨졌다 */\nexport function partExtension(path: string): string {\n return (/\\.([A-Za-z0-9]{1,5})$/.exec(path.slice(path.lastIndexOf(\"/\") + 1))?.[1] ?? \"bin\").toLowerCase()\n}\n"]}
@@ -1,7 +1,7 @@
1
1
  #!/usr/bin/env node
2
2
  import {
3
3
  assertNetworkAllowed
4
- } from "./chunk-TDMBTG7U.js";
4
+ } from "./chunk-KXHNIHLC.js";
5
5
 
6
6
  // src/pdf/formula/models.ts
7
7
  import { createHash as createHash2 } from "crypto";
@@ -296,4 +296,4 @@ export {
296
296
  ensureFormulaModels,
297
297
  ensureSingleModel
298
298
  };
299
- //# sourceMappingURL=chunk-W24TJJNN.js.map
299
+ //# sourceMappingURL=chunk-AEEQN6L3.js.map
@@ -1,7 +1,7 @@
1
1
  import {
2
2
  escapeHtml,
3
3
  sanitizeHref
4
- } from "./chunk-WJAFE3Z6.js";
4
+ } from "./chunk-I2QETSZN.js";
5
5
 
6
6
  // src/script-tags.ts
7
7
  var LITERAL_TAG = /<(?=\/?(?:u|sup|sub)>)/g;
@@ -1539,4 +1539,4 @@ export {
1539
1539
  HEADING_RATIO_H3,
1540
1540
  validateOcrOptions
1541
1541
  };
1542
- //# sourceMappingURL=chunk-TW3V6FSH.js.map
1542
+ //# sourceMappingURL=chunk-EEXIFCXC.js.map
@@ -2,7 +2,7 @@
2
2
 
3
3
  // src/utils.ts
4
4
  import { format } from "util";
5
- var VERSION = true ? "4.24.2" : "0.0.0-dev";
5
+ var VERSION = true ? "4.24.3" : "0.0.0-dev";
6
6
  function toArrayBuffer(buf) {
7
7
  if (buf.byteOffset === 0 && buf.byteLength === buf.buffer.byteLength) {
8
8
  return buf.buffer;
@@ -172,4 +172,4 @@ export {
172
172
  classifyError,
173
173
  partExtension
174
174
  };
175
- //# sourceMappingURL=chunk-LMEF7FN5.js.map
175
+ //# sourceMappingURL=chunk-EWCJCWTT.js.map
@@ -10,14 +10,14 @@ import {
10
10
  ocrSourcePath,
11
11
  parseCharacterDict,
12
12
  resolveOcrModels
13
- } from "./chunk-JMOAMKAR.js";
13
+ } from "./chunk-XZOAE2Q7.js";
14
14
  import {
15
15
  validateOcrOptions
16
- } from "./chunk-TW3V6FSH.js";
16
+ } from "./chunk-EEXIFCXC.js";
17
17
  import {
18
18
  KordocError,
19
19
  OPTIONAL_DEP_INSTALL_HINT
20
- } from "./chunk-WJAFE3Z6.js";
20
+ } from "./chunk-I2QETSZN.js";
21
21
 
22
22
  // src/ocr/ruling-lines.ts
23
23
  var INK_LUMA_MAX = 205;
@@ -2212,4 +2212,4 @@ export {
2212
2212
  groupOcrLines,
2213
2213
  ocrItemsToBlocks
2214
2214
  };
2215
- //# sourceMappingURL=chunk-MHNJOJQV.js.map
2215
+ //# sourceMappingURL=chunk-FZT34MUI.js.map
@@ -2,7 +2,7 @@
2
2
  import {
3
3
  escapeHtml,
4
4
  sanitizeHref
5
- } from "./chunk-LMEF7FN5.js";
5
+ } from "./chunk-EWCJCWTT.js";
6
6
 
7
7
  // src/shared/symbol-fonts.ts
8
8
  var WINGDINGS = [
@@ -1179,4 +1179,4 @@ export {
1179
1179
  visibleText,
1180
1180
  escapeHtmlCellText
1181
1181
  };
1182
- //# sourceMappingURL=chunk-WDSTW3BD.js.map
1182
+ //# sourceMappingURL=chunk-HNX2I7D6.js.map
@@ -10,14 +10,14 @@ var _chunkTK3FDAEEcjs = require('./chunk-TK3FDAEE.cjs');
10
10
 
11
11
 
12
12
 
13
- var _chunkUNICY546cjs = require('./chunk-UNICY546.cjs');
13
+ var _chunkJILIA5EVcjs = require('./chunk-JILIA5EV.cjs');
14
14
 
15
15
 
16
- var _chunkJFUUFJZWcjs = require('./chunk-JFUUFJZW.cjs');
16
+ var _chunkN65EY5BIcjs = require('./chunk-N65EY5BI.cjs');
17
17
 
18
18
 
19
19
 
20
- var _chunkPNSALBXHcjs = require('./chunk-PNSALBXH.cjs');
20
+ var _chunk6ZRZLV3Jcjs = require('./chunk-6ZRZLV3J.cjs');
21
21
 
22
22
  // src/ocr/ruling-lines.ts
23
23
  var INK_LUMA_MAX = 205;
@@ -1410,7 +1410,7 @@ var OcrEngine = class _OcrEngine {
1410
1410
  this.ort = parts.ort;
1411
1411
  this.sharp = parts.sharp;
1412
1412
  }
1413
- static async create(models = _chunkUNICY546cjs.DEFAULT_OCR_MODELS) {
1413
+ static async create(models = _chunkJILIA5EVcjs.DEFAULT_OCR_MODELS) {
1414
1414
  _chunkTK3FDAEEcjs.disableOrtTelemetry.call(void 0, );
1415
1415
  const [ortMod, sharpModRaw] = await Promise.all([
1416
1416
  tryImport("onnxruntime-node", () => Promise.resolve().then(() => _interopRequireWildcard(require("onnxruntime-node")))),
@@ -1430,16 +1430,16 @@ var OcrEngine = class _OcrEngine {
1430
1430
  const local = (s) => "spec" in s ? Promise.resolve(null) : verifiedBytes(s);
1431
1431
  const [detBytes, recBytes, dictBytes] = await Promise.all([local(models.det), local(models.rec), local(models.dict)]);
1432
1432
  const session = (s, bytes) => bytes ? ortMod.InferenceSession.create(bytes, sessionOpts).catch(() => {
1433
- throw new (0, _chunkPNSALBXHcjs.KordocError)(`OCR \uBAA8\uB378\uC744 \uC5F4\uC9C0 \uBABB\uD588\uC2B5\uB2C8\uB2E4: ${_path.basename.call(void 0, _chunkUNICY546cjs.ocrSourcePath.call(void 0, s))} \u2014 ONNX \uBAA8\uB378\uC774 \uC544\uB2C8\uAC70\uB098 \uC190\uC0C1\uB410\uC2B5\uB2C8\uB2E4`);
1434
- }) : ortMod.InferenceSession.create(_chunkUNICY546cjs.ocrSourcePath.call(void 0, s), sessionOpts);
1433
+ throw new (0, _chunk6ZRZLV3Jcjs.KordocError)(`OCR \uBAA8\uB378\uC744 \uC5F4\uC9C0 \uBABB\uD588\uC2B5\uB2C8\uB2E4: ${_path.basename.call(void 0, _chunkJILIA5EVcjs.ocrSourcePath.call(void 0, s))} \u2014 ONNX \uBAA8\uB378\uC774 \uC544\uB2C8\uAC70\uB098 \uC190\uC0C1\uB410\uC2B5\uB2C8\uB2E4`);
1434
+ }) : ortMod.InferenceSession.create(_chunkJILIA5EVcjs.ocrSourcePath.call(void 0, s), sessionOpts);
1435
1435
  const [det, rec, dictYml] = await Promise.all([
1436
1436
  session(models.det, detBytes),
1437
1437
  session(models.rec, recBytes),
1438
- dictBytes ? new TextDecoder().decode(dictBytes) : _promises.readFile.call(void 0, _chunkUNICY546cjs.ocrSourcePath.call(void 0, models.dict), "utf-8")
1438
+ dictBytes ? new TextDecoder().decode(dictBytes) : _promises.readFile.call(void 0, _chunkJILIA5EVcjs.ocrSourcePath.call(void 0, models.dict), "utf-8")
1439
1439
  ]);
1440
- const dict = _chunkUNICY546cjs.parseCharacterDict.call(void 0, dictYml);
1440
+ const dict = _chunkJILIA5EVcjs.parseCharacterDict.call(void 0, dictYml);
1441
1441
  if (dict.length === 0) {
1442
- throw "spec" in models.dict ? new Error("OCR \uC0AC\uC804 \uD30C\uC2F1 \uC2E4\uD328 \u2014 \uBAA8\uB378 \uCE90\uC2DC\uB97C \uC0AD\uC81C \uD6C4 \uC7AC\uB2E4\uC6B4\uB85C\uB4DC\uD558\uC138\uC694") : new (0, _chunkPNSALBXHcjs.KordocError)(`OCR \uC0AC\uC804 \uD30C\uC2F1 \uC2E4\uD328 \u2014 ${_path.basename.call(void 0, models.dict.file.path)} \uC5D0 PostProcess.character_dict \uBAA9\uB85D\uC774 \uC5C6\uC2B5\uB2C8\uB2E4`);
1442
+ throw "spec" in models.dict ? new Error("OCR \uC0AC\uC804 \uD30C\uC2F1 \uC2E4\uD328 \u2014 \uBAA8\uB378 \uCE90\uC2DC\uB97C \uC0AD\uC81C \uD6C4 \uC7AC\uB2E4\uC6B4\uB85C\uB4DC\uD558\uC138\uC694") : new (0, _chunk6ZRZLV3Jcjs.KordocError)(`OCR \uC0AC\uC804 \uD30C\uC2F1 \uC2E4\uD328 \u2014 ${_path.basename.call(void 0, models.dict.file.path)} \uC5D0 PostProcess.character_dict \uBAA9\uB85D\uC774 \uC5C6\uC2B5\uB2C8\uB2E4`);
1443
1443
  }
1444
1444
  return new _OcrEngine({ det, rec, dict, ort: ortMod, sharp: sharpMod });
1445
1445
  }
@@ -1648,7 +1648,7 @@ var OcrEngine = class _OcrEngine {
1648
1648
  const out = await this.rec.run({ [this.rec.inputNames[0]]: tensor });
1649
1649
  const logits = out[this.rec.outputNames[0]];
1650
1650
  const [, T, C] = logits.dims;
1651
- if (C !== this.dict.length + 2) throw new (0, _chunkPNSALBXHcjs.KordocError)(`OCR rec \uCD9C\uB825 \uD074\uB798\uC2A4 ${C}\uAC1C\uAC00 \uC0AC\uC804 ${this.dict.length}\uC790 + 2 \uC640 \uB2E4\uB985\uB2C8\uB2E4 \u2014 \uBAA8\uB378\uC5D0 \uB9DE\uB294 \uC0AC\uC804(ocrRecDict)\uC744 \uC9C0\uC815\uD558\uC138\uC694`);
1651
+ if (C !== this.dict.length + 2) throw new (0, _chunk6ZRZLV3Jcjs.KordocError)(`OCR rec \uCD9C\uB825 \uD074\uB798\uC2A4 ${C}\uAC1C\uAC00 \uC0AC\uC804 ${this.dict.length}\uC790 + 2 \uC640 \uB2E4\uB985\uB2C8\uB2E4 \u2014 \uBAA8\uB378\uC5D0 \uB9DE\uB294 \uC0AC\uC804(ocrRecDict)\uC744 \uC9C0\uC815\uD558\uC138\uC694`);
1652
1652
  const data = logits.data;
1653
1653
  idx.forEach((ci, k) => {
1654
1654
  const r = ctcDecode(data.subarray(k * T * C, (k + 1) * T * C), T, C, this.dict);
@@ -1771,13 +1771,13 @@ async function tryImport(name, loader) {
1771
1771
  return await loader();
1772
1772
  } catch (e) {
1773
1773
  throw new Error(
1774
- `\uB0B4\uC7A5 OCR \uC744 \uC0AC\uC6A9\uD558\uB824\uBA74 optional dependency '${name}' \uC774 \uD544\uC694\uD569\uB2C8\uB2E4. \`npm install ${name}\` \uD6C4 \uB2E4\uC2DC \uC2E4\uD589\uD558\uC138\uC694.${_chunkPNSALBXHcjs.OPTIONAL_DEP_INSTALL_HINT} \uC6D0\uC778: ${e.message}`
1774
+ `\uB0B4\uC7A5 OCR \uC744 \uC0AC\uC6A9\uD558\uB824\uBA74 optional dependency '${name}' \uC774 \uD544\uC694\uD569\uB2C8\uB2E4. \`npm install ${name}\` \uD6C4 \uB2E4\uC2DC \uC2E4\uD589\uD558\uC138\uC694.${_chunk6ZRZLV3Jcjs.OPTIONAL_DEP_INSTALL_HINT} \uC6D0\uC778: ${e.message}`
1775
1775
  );
1776
1776
  }
1777
1777
  }
1778
1778
  async function verifiedBytes(s) {
1779
1779
  const name = _path.basename.call(void 0, s.file.path);
1780
- const fail = (why2) => new (0, _chunkPNSALBXHcjs.KordocError)(`OCR \uBAA8\uB378 \uD30C\uC77C\uC744 \uC77D\uC9C0 \uBABB\uD588\uC2B5\uB2C8\uB2E4: ${name} (${why2})`);
1780
+ const fail = (why2) => new (0, _chunk6ZRZLV3Jcjs.KordocError)(`OCR \uBAA8\uB378 \uD30C\uC77C\uC744 \uC77D\uC9C0 \uBABB\uD588\uC2B5\uB2C8\uB2E4: ${name} (${why2})`);
1781
1781
  const why = (e) => {
1782
1782
  const code = e.code;
1783
1783
  return code === "ENOENT" ? "\uD30C\uC77C\uC774 \uC5C6\uC74C" : code === "EACCES" || code === "EPERM" ? "\uAD8C\uD55C \uC5C6\uC74C" : "\uC77D\uAE30 \uC2E4\uD328";
@@ -1798,13 +1798,13 @@ async function verifiedBytes(s) {
1798
1798
  }
1799
1799
  const actual = _crypto.createHash.call(void 0, "sha256").update(bytes).digest("hex");
1800
1800
  if (actual !== s.file.sha256) {
1801
- throw new (0, _chunkPNSALBXHcjs.KordocError)(`OCR \uBAA8\uB378 SHA256 mismatch: ${_path.basename.call(void 0, s.file.path)} \u2014 expected ${s.file.sha256}, got ${actual} (\uB85C\uB4DC \uAC70\uBD80)`);
1801
+ throw new (0, _chunk6ZRZLV3Jcjs.KordocError)(`OCR \uBAA8\uB378 SHA256 mismatch: ${_path.basename.call(void 0, s.file.path)} \u2014 expected ${s.file.sha256}, got ${actual} (\uB85C\uB4DC \uAC70\uBD80)`);
1802
1802
  }
1803
1803
  return bytes;
1804
1804
  }
1805
1805
  var engines = /* @__PURE__ */ new Map();
1806
- function getOcrEngine(models = _chunkUNICY546cjs.DEFAULT_OCR_MODELS) {
1807
- const key = _chunkUNICY546cjs.ocrSelectionKey.call(void 0, models);
1806
+ function getOcrEngine(models = _chunkJILIA5EVcjs.DEFAULT_OCR_MODELS) {
1807
+ const key = _chunkJILIA5EVcjs.ocrSelectionKey.call(void 0, models);
1808
1808
  let p = engines.get(key);
1809
1809
  if (!p) {
1810
1810
  p = OcrEngine.create(models).catch((err) => {
@@ -1907,13 +1907,13 @@ function rotateRgba(rgba, width, height, deg) {
1907
1907
  var OCR_RENDER_SCALE = 3;
1908
1908
  var MAX_OCR_PIXELS = 24e6;
1909
1909
  var PAGE_TIMEOUT_MS = 12e4;
1910
- var DEFAULT_BUILTIN_OCR = Object.freeze({ models: _chunkUNICY546cjs.DEFAULT_OCR_MODELS, tuning: DEFAULT_OCR_TUNING });
1910
+ var DEFAULT_BUILTIN_OCR = Object.freeze({ models: _chunkJILIA5EVcjs.DEFAULT_OCR_MODELS, tuning: DEFAULT_OCR_TUNING });
1911
1911
  function builtinOcrConfig(o) {
1912
1912
  if (!o || o.ocrDet === void 0 && o.ocrDetLongSide === void 0 && o.ocrRec === void 0 && o.ocrRecDict === void 0 && !o.ocrLines) return DEFAULT_BUILTIN_OCR;
1913
- const bad = _chunkJFUUFJZWcjs.validateOcrOptions.call(void 0, o);
1914
- if (bad) throw new (0, _chunkPNSALBXHcjs.KordocError)(bad);
1913
+ const bad = _chunkN65EY5BIcjs.validateOcrOptions.call(void 0, o);
1914
+ if (bad) throw new (0, _chunk6ZRZLV3Jcjs.KordocError)(bad);
1915
1915
  return {
1916
- models: _chunkUNICY546cjs.resolveOcrModels.call(void 0, o),
1916
+ models: _chunkJILIA5EVcjs.resolveOcrModels.call(void 0, o),
1917
1917
  tuning: o.ocrDetLongSide === void 0 ? DEFAULT_OCR_TUNING : { ...DEFAULT_OCR_TUNING, detLongSide: o.ocrDetLongSide },
1918
1918
  lines: o.ocrLines === "all" ? "all" : o.ocrLines ? "kept" : void 0
1919
1919
  };
@@ -1939,7 +1939,7 @@ async function runPdfOcr(buffer, targets, mode, warnings, onProgress, detectTabl
1939
1939
  "@hyzyla/pdfium",
1940
1940
  () => Promise.resolve().then(() => _interopRequireWildcard(require("@hyzyla/pdfium")))
1941
1941
  );
1942
- if (mode === "builtin") await _chunkUNICY546cjs.ensureOcrModelsFor.call(void 0, config.models, reportModelDownload);
1942
+ if (mode === "builtin") await _chunkJILIA5EVcjs.ensureOcrModelsFor.call(void 0, config.models, reportModelDownload);
1943
1943
  const engine = mode === "builtin" ? await getOcrEngine(config.models) : null;
1944
1944
  const pdfium = await pdfiumMod.PDFiumLibrary.init();
1945
1945
  const doc = await pdfium.loadDocument(new Uint8Array(buffer));
@@ -2023,7 +2023,7 @@ async function ocrOnePage(page, pageNo, mode, engine, warnings, detectTables, ve
2023
2023
  const x1 = r.x1 * scale - 2, x2 = r.x2 * scale + 2, y1 = (pdfH - r.y2) * scale - 2, y2 = (pdfH - r.y1) * scale + 2;
2024
2024
  const columnRules = _nullishCoalesce(_optionalChain([ruling, 'optionalAccess', _26 => _26.verticals, 'access', _27 => _27.filter, 'call', _28 => _28((s) => Math.min(s.x1, s.x2) >= x1 && Math.max(s.x1, s.x2) <= x2 && Math.min(s.y1, s.y2) >= y1 && Math.max(s.y1, s.y2) <= y2), 'access', _29 => _29.length]), () => ( 0));
2025
2025
  if (columnRules < 2) {
2026
- for (const b of blocks) if (b.type === "table") _chunkUNICY546cjs.UNRULED_REGION_TABLES.add(b);
2026
+ for (const b of blocks) if (b.type === "table") _chunkJILIA5EVcjs.UNRULED_REGION_TABLES.add(b);
2027
2027
  }
2028
2028
  return blocks;
2029
2029
  });
@@ -2159,14 +2159,14 @@ function ocrItemsToBlocks(items, pageNumber, pdfW, pdfH, scale, extraLines, dete
2159
2159
  ...xs[i].space ? { hasSpaceBefore: true } : {}
2160
2160
  };
2161
2161
  });
2162
- return _chunkUNICY546cjs.extractPageBlocksWithLines.call(void 0, norm, pageNumber, opList, pdfW, pdfH, extraLines, detectTables);
2162
+ return _chunkJILIA5EVcjs.extractPageBlocksWithLines.call(void 0, norm, pageNumber, opList, pdfW, pdfH, extraLines, detectTables);
2163
2163
  }
2164
2164
  async function tryImport2(name, loader) {
2165
2165
  try {
2166
2166
  return await loader();
2167
2167
  } catch (e) {
2168
2168
  throw new Error(
2169
- `OCR \uC744 \uC0AC\uC6A9\uD558\uB824\uBA74 optional dependency '${name}' \uC774 \uD544\uC694\uD569\uB2C8\uB2E4. \`npm install ${name}\` \uD6C4 \uB2E4\uC2DC \uC2E4\uD589\uD558\uC138\uC694.${_chunkPNSALBXHcjs.OPTIONAL_DEP_INSTALL_HINT} \uC6D0\uC778: ${e.message}`
2169
+ `OCR \uC744 \uC0AC\uC6A9\uD558\uB824\uBA74 optional dependency '${name}' \uC774 \uD544\uC694\uD569\uB2C8\uB2E4. \`npm install ${name}\` \uD6C4 \uB2E4\uC2DC \uC2E4\uD589\uD558\uC138\uC694.${_chunk6ZRZLV3Jcjs.OPTIONAL_DEP_INSTALL_HINT} \uC6D0\uC778: ${e.message}`
2170
2170
  );
2171
2171
  }
2172
2172
  }
@@ -2212,4 +2212,4 @@ function bgraToRgba(bgra) {
2212
2212
 
2213
2213
 
2214
2214
  exports.getOcrEngine = getOcrEngine; exports.detectRulingLines = detectRulingLines; exports.rulingToPdfLines = rulingToPdfLines; exports.deskewPage = deskewPage; exports.MAX_OCR_PIXELS = MAX_OCR_PIXELS; exports.DEFAULT_BUILTIN_OCR = DEFAULT_BUILTIN_OCR; exports.builtinOcrConfig = builtinOcrConfig; exports.reportModelDownload = reportModelDownload; exports.toOcrLine = toOcrLine; exports.pageLines = pageLines; exports.runPdfOcr = runPdfOcr; exports.ocrOnePage = ocrOnePage; exports.groupOcrLines = groupOcrLines; exports.ocrItemsToBlocks = ocrItemsToBlocks;
2215
- //# sourceMappingURL=chunk-PNX5FYHC.cjs.map
2215
+ //# sourceMappingURL=chunk-HTG3VVET.cjs.map