kordoc 4.5.0 → 4.6.0

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 (184) hide show
  1. package/LICENSE +21 -21
  2. package/NOTICE +176 -176
  3. package/README.md +958 -951
  4. package/THIRD_PARTY/apache-2.0.LICENSE +201 -201
  5. package/THIRD_PARTY/claw-hwp.LICENSE +21 -21
  6. package/THIRD_PARTY/hml-equation-parser.LICENSE +201 -201
  7. package/THIRD_PARTY/hml-equation-parser.txt +35 -35
  8. package/THIRD_PARTY/opendataloader-pdf.txt +32 -32
  9. package/THIRD_PARTY/pix2text.txt +17 -17
  10. package/THIRD_PARTY/rhwp-forms.txt +35 -35
  11. package/dist/{-CWYIC4ES.js → -DEVLQKIT.js} +12 -12
  12. package/dist/{chunk-M4I5F6GK.js → chunk-2CABQYTT.js} +9 -9
  13. package/dist/chunk-2CABQYTT.js.map +1 -0
  14. package/dist/{chunk-NBOVSNSR.js → chunk-2CENWSAW.js} +10 -10
  15. package/dist/chunk-2CENWSAW.js.map +1 -0
  16. package/dist/{chunk-3H5X63BQ.js → chunk-3FFPHTXD.js} +1 -1
  17. package/dist/chunk-3FFPHTXD.js.map +1 -0
  18. package/dist/{chunk-MOL7MDBG.js → chunk-3TBUDJDE.js} +1 -1
  19. package/dist/chunk-3TBUDJDE.js.map +1 -0
  20. package/dist/{chunk-5RPYBC2Q.js → chunk-6GBLFQMA.js} +1 -1
  21. package/dist/chunk-6GBLFQMA.js.map +1 -0
  22. package/dist/{chunk-GZDOWYJM.js → chunk-6XPXIG57.js} +2 -2
  23. package/dist/chunk-6XPXIG57.js.map +1 -0
  24. package/dist/{chunk-45EUJ2P6.js → chunk-7BTO44Q7.js} +37 -21
  25. package/dist/chunk-7BTO44Q7.js.map +1 -0
  26. package/dist/{chunk-GE43BE46.js → chunk-AIQ3ISQU.js} +1 -1
  27. package/dist/chunk-AIQ3ISQU.js.map +1 -0
  28. package/dist/{chunk-GLPLLT4Q.cjs → chunk-CARNWWGI.cjs} +4 -4
  29. package/dist/chunk-CARNWWGI.cjs.map +1 -0
  30. package/dist/chunk-GS7T56RP.cjs.map +1 -1
  31. package/dist/{chunk-TCPO6XAY.cjs → chunk-HPRGVU6A.cjs} +11 -11
  32. package/dist/chunk-HPRGVU6A.cjs.map +1 -0
  33. package/dist/{chunk-U3ZNF4SO.cjs → chunk-I42DND3B.cjs} +17 -17
  34. package/dist/chunk-I42DND3B.cjs.map +1 -0
  35. package/dist/{chunk-ZHNUH2IW.js → chunk-IBVMJIHR.js} +2 -2
  36. package/dist/chunk-IBVMJIHR.js.map +1 -0
  37. package/dist/{chunk-XYSTZIIL.js → chunk-IHK5LCYX.js} +2 -2
  38. package/dist/chunk-IHK5LCYX.js.map +1 -0
  39. package/dist/{chunk-TP4P7M4I.js → chunk-KC63W72C.js} +2 -2
  40. package/dist/chunk-KC63W72C.js.map +1 -0
  41. package/dist/{chunk-GCHH4WHT.js → chunk-KQOTL4WI.js} +2 -2
  42. package/dist/chunk-KQOTL4WI.js.map +1 -0
  43. package/dist/{chunk-ZV5OKQ3Z.cjs → chunk-MGGU3ZRK.cjs} +1 -1
  44. package/dist/chunk-MGGU3ZRK.cjs.map +1 -0
  45. package/dist/{chunk-FT2EXMMP.js → chunk-MSMSHU2F.js} +9 -9
  46. package/dist/chunk-MSMSHU2F.js.map +1 -0
  47. package/dist/{chunk-DZIXKL7E.js → chunk-O5EK52TG.js} +1 -1
  48. package/dist/chunk-O5EK52TG.js.map +1 -0
  49. package/dist/{chunk-CBUC73JM.js → chunk-OPL22SYF.js} +3 -3
  50. package/dist/chunk-OPL22SYF.js.map +1 -0
  51. package/dist/{chunk-2CF2WXBG.js → chunk-PQDNCWAZ.js} +11 -11
  52. package/dist/chunk-PQDNCWAZ.js.map +1 -0
  53. package/dist/{chunk-UEWW3DH2.js → chunk-PSAHY5XW.js} +1 -1
  54. package/dist/chunk-PSAHY5XW.js.map +1 -0
  55. package/dist/{chunk-EYLOFSA5.js → chunk-UGKXJA3W.js} +4 -3
  56. package/dist/chunk-UGKXJA3W.js.map +1 -0
  57. package/dist/{chunk-WDIPTY3Q.js → chunk-UQ4NSK66.js} +3 -3
  58. package/dist/chunk-UQ4NSK66.js.map +1 -0
  59. package/dist/{chunk-7ZQKECQT.js → chunk-VNEV6VKD.js} +2 -2
  60. package/dist/chunk-VNEV6VKD.js.map +1 -0
  61. package/dist/{chunk-7FCNOKOR.js → chunk-W34QSJFF.js} +1 -1
  62. package/dist/chunk-W34QSJFF.js.map +1 -0
  63. package/dist/{chunk-UQ6KDJWE.cjs → chunk-WX7WQTOT.cjs} +4 -3
  64. package/dist/chunk-WX7WQTOT.cjs.map +1 -0
  65. package/dist/{chunk-DCZVOIEO.cjs → chunk-X34YWRL5.cjs} +1 -1
  66. package/dist/chunk-X34YWRL5.cjs.map +1 -0
  67. package/dist/{chunk-UAFSPWU7.js → chunk-YNJDNP44.js} +1 -1
  68. package/dist/chunk-YNJDNP44.js.map +1 -0
  69. package/dist/{chunk-TTEHX4FN.js → chunk-ZYQ5X6PG.js} +4 -3
  70. package/dist/chunk-ZYQ5X6PG.js.map +1 -0
  71. package/dist/chunks-Y2VE2WCJ.js +10 -0
  72. package/dist/cli.js +54 -29
  73. package/dist/cli.js.map +1 -1
  74. package/dist/{formula-JLYS73QL.js → formula-ETB6MESN.js} +2 -2
  75. package/dist/formula-ETB6MESN.js.map +1 -0
  76. package/dist/{formula-LIVQSCMM.js → formula-KW4NI5IO.js} +2 -2
  77. package/dist/formula-KW4NI5IO.js.map +1 -0
  78. package/dist/{formula-CG4S7TWO.cjs → formula-WKGP3D3Y.cjs} +8 -8
  79. package/dist/formula-WKGP3D3Y.cjs.map +1 -0
  80. package/dist/{image-ocr-BWQIUUVC.js → image-ocr-C6U7YHRC.js} +6 -6
  81. package/dist/image-ocr-C6U7YHRC.js.map +1 -0
  82. package/dist/{image-ocr-EDZOUU6Q.js → image-ocr-YIBJCOUQ.js} +8 -8
  83. package/dist/image-ocr-YIBJCOUQ.js.map +1 -0
  84. package/dist/{image-ocr-KZ4MGVJ3.cjs → image-ocr-YID6TDN5.cjs} +11 -11
  85. package/dist/image-ocr-YID6TDN5.cjs.map +1 -0
  86. package/dist/index.cjs +209 -193
  87. package/dist/index.cjs.map +1 -1
  88. package/dist/index.d.cts +13 -1
  89. package/dist/index.d.ts +13 -1
  90. package/dist/index.js +31 -15
  91. package/dist/index.js.map +1 -1
  92. package/dist/mcp.js +25 -23
  93. package/dist/mcp.js.map +1 -1
  94. package/dist/{models-6SIKLKRH.js → models-6VMAVZR6.js} +3 -3
  95. package/dist/page-range-CIRRJPXJ.js +7 -0
  96. package/dist/page-range-OF5I4PQY.js +8 -0
  97. package/dist/page-range-TPIRSA3J.cjs +8 -0
  98. package/dist/page-range-TPIRSA3J.cjs.map +1 -0
  99. package/dist/{parser-7G2QFIYZ.cjs → parser-EYCVDUJA.cjs} +33 -33
  100. package/dist/parser-EYCVDUJA.cjs.map +1 -0
  101. package/dist/{parser-JLEKXA2Z.js → parser-GFXZV2VJ.js} +11 -11
  102. package/dist/parser-GFXZV2VJ.js.map +1 -0
  103. package/dist/{parser-6OJOFJ64.js → parser-SHM3JEH6.js} +9 -9
  104. package/dist/parser-SHM3JEH6.js.map +1 -0
  105. package/dist/pdf-ocr-TMAWPOHC.cjs +13 -0
  106. package/dist/pdf-ocr-TMAWPOHC.cjs.map +1 -0
  107. package/dist/pdf-ocr-UKFYODLP.js +15 -0
  108. package/dist/pdf-ocr-ZG4UVUIT.js +12 -0
  109. package/dist/{profile-io-FMPT6VEZ.js → profile-io-FQZNNA3O.js} +3 -3
  110. package/dist/{rasterize-LARCX6CS.js → rasterize-QGKFNTXO.js} +2 -2
  111. package/dist/rasterize-QGKFNTXO.js.map +1 -0
  112. package/dist/{redact-7ILEAUTS.js → redact-W5K3ILQH.js} +2 -2
  113. package/dist/render-YFIWQYVB.js +10 -0
  114. package/dist/seal-GZ7NDWLR.js +10 -0
  115. package/dist/{setup-GUQT5ELJ.js → setup-I4I75QJU.js} +1 -1
  116. package/dist/setup-I4I75QJU.js.map +1 -0
  117. package/dist/{watch-Y2VKD2MN.js → watch-OIVXH2RN.js} +12 -12
  118. package/dist/watch-OIVXH2RN.js.map +1 -0
  119. package/package.json +103 -103
  120. package/templates//352/260/204/354/235/264/352/270/260/354/225/210/353/254/270_/354/230/210/354/213/234/352/260/222.json +9 -9
  121. package/templates//354/235/274/353/260/230/352/270/260/354/225/210/353/254/270_/354/230/210/354/213/234/352/260/222.json +24 -24
  122. package/dist/chunk-2CF2WXBG.js.map +0 -1
  123. package/dist/chunk-3H5X63BQ.js.map +0 -1
  124. package/dist/chunk-45EUJ2P6.js.map +0 -1
  125. package/dist/chunk-5RPYBC2Q.js.map +0 -1
  126. package/dist/chunk-7FCNOKOR.js.map +0 -1
  127. package/dist/chunk-7ZQKECQT.js.map +0 -1
  128. package/dist/chunk-CBUC73JM.js.map +0 -1
  129. package/dist/chunk-DCZVOIEO.cjs.map +0 -1
  130. package/dist/chunk-DZIXKL7E.js.map +0 -1
  131. package/dist/chunk-EYLOFSA5.js.map +0 -1
  132. package/dist/chunk-FT2EXMMP.js.map +0 -1
  133. package/dist/chunk-GCHH4WHT.js.map +0 -1
  134. package/dist/chunk-GE43BE46.js.map +0 -1
  135. package/dist/chunk-GLPLLT4Q.cjs.map +0 -1
  136. package/dist/chunk-GZDOWYJM.js.map +0 -1
  137. package/dist/chunk-M4I5F6GK.js.map +0 -1
  138. package/dist/chunk-MOL7MDBG.js.map +0 -1
  139. package/dist/chunk-NBOVSNSR.js.map +0 -1
  140. package/dist/chunk-TCPO6XAY.cjs.map +0 -1
  141. package/dist/chunk-TP4P7M4I.js.map +0 -1
  142. package/dist/chunk-TTEHX4FN.js.map +0 -1
  143. package/dist/chunk-U3ZNF4SO.cjs.map +0 -1
  144. package/dist/chunk-UAFSPWU7.js.map +0 -1
  145. package/dist/chunk-UEWW3DH2.js.map +0 -1
  146. package/dist/chunk-UQ6KDJWE.cjs.map +0 -1
  147. package/dist/chunk-WDIPTY3Q.js.map +0 -1
  148. package/dist/chunk-XYSTZIIL.js.map +0 -1
  149. package/dist/chunk-ZHNUH2IW.js.map +0 -1
  150. package/dist/chunk-ZV5OKQ3Z.cjs.map +0 -1
  151. package/dist/chunks-EYDY4Y6O.js +0 -10
  152. package/dist/formula-CG4S7TWO.cjs.map +0 -1
  153. package/dist/formula-JLYS73QL.js.map +0 -1
  154. package/dist/formula-LIVQSCMM.js.map +0 -1
  155. package/dist/image-ocr-BWQIUUVC.js.map +0 -1
  156. package/dist/image-ocr-EDZOUU6Q.js.map +0 -1
  157. package/dist/image-ocr-KZ4MGVJ3.cjs.map +0 -1
  158. package/dist/page-range-737B4EZW.js +0 -8
  159. package/dist/page-range-LNMXJU6W.js +0 -7
  160. package/dist/page-range-P7SDW6LR.cjs +0 -8
  161. package/dist/page-range-P7SDW6LR.cjs.map +0 -1
  162. package/dist/parser-6OJOFJ64.js.map +0 -1
  163. package/dist/parser-7G2QFIYZ.cjs.map +0 -1
  164. package/dist/parser-JLEKXA2Z.js.map +0 -1
  165. package/dist/pdf-ocr-364VNE6N.cjs +0 -13
  166. package/dist/pdf-ocr-364VNE6N.cjs.map +0 -1
  167. package/dist/pdf-ocr-OWUZLWQ3.js +0 -15
  168. package/dist/pdf-ocr-YQGMCT53.js +0 -12
  169. package/dist/rasterize-LARCX6CS.js.map +0 -1
  170. package/dist/render-4WHQ7JCX.js +0 -10
  171. package/dist/seal-BX6YA6PI.js +0 -10
  172. package/dist/setup-GUQT5ELJ.js.map +0 -1
  173. package/dist/watch-Y2VKD2MN.js.map +0 -1
  174. /package/dist/{-CWYIC4ES.js.map → -DEVLQKIT.js.map} +0 -0
  175. /package/dist/{chunks-EYDY4Y6O.js.map → chunks-Y2VE2WCJ.js.map} +0 -0
  176. /package/dist/{models-6SIKLKRH.js.map → models-6VMAVZR6.js.map} +0 -0
  177. /package/dist/{page-range-737B4EZW.js.map → page-range-CIRRJPXJ.js.map} +0 -0
  178. /package/dist/{page-range-LNMXJU6W.js.map → page-range-OF5I4PQY.js.map} +0 -0
  179. /package/dist/{pdf-ocr-OWUZLWQ3.js.map → pdf-ocr-UKFYODLP.js.map} +0 -0
  180. /package/dist/{pdf-ocr-YQGMCT53.js.map → pdf-ocr-ZG4UVUIT.js.map} +0 -0
  181. /package/dist/{profile-io-FMPT6VEZ.js.map → profile-io-FQZNNA3O.js.map} +0 -0
  182. /package/dist/{redact-7ILEAUTS.js.map → redact-W5K3ILQH.js.map} +0 -0
  183. /package/dist/{render-4WHQ7JCX.js.map → render-YFIWQYVB.js.map} +0 -0
  184. /package/dist/{seal-BX6YA6PI.js.map → seal-GZ7NDWLR.js.map} +0 -0
@@ -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 ? "4.5.0" : "0.0.0-dev";
2
+ var VERSION = true ? "4.6.0" : "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;
@@ -73,7 +73,7 @@ function precheckZipSize(buffer, maxUncompressedSize = 256 * 1024 * 1024, maxEnt
73
73
  }
74
74
  }
75
75
  function stripDtd(xml) {
76
- return xml.replace(/<!DOCTYPE\s[^[>]*(\[[\s\S]*?\])?\s*>/gi, "");
76
+ return xml.replace(/^\uFEFF/, "").replace(/<!DOCTYPE\s[^[>]*(\[[\s\S]*?\])?\s*>/gi, "");
77
77
  }
78
78
  var SAFE_HREF_RE = /^(?:https?:|mailto:|tel:|#)/i;
79
79
  function sanitizeHref(href) {
@@ -97,6 +97,7 @@ function classifyError(err) {
97
97
  if (msg.includes("DRM")) return "DRM_PROTECTED";
98
98
  if (msg.includes("\uC554\uD638\uD654") || msg.includes("\uC554\uD638\uB85C \uBCF4\uD638")) return "ENCRYPTED";
99
99
  if (msg.includes("optional dependency")) return "MISSING_DEPENDENCY";
100
+ if (msg.includes("Invalid string length") || msg.includes("Cannot create a string longer")) return "OUTPUT_TOO_LARGE";
100
101
  if (msg.includes("ZIP bomb") || msg.includes("ZIP \uBE44\uC555\uCD95 \uD06C\uAE30 \uCD08\uACFC") || msg.includes("ZIP \uC5D4\uD2B8\uB9AC \uC218 \uCD08\uACFC")) return "ZIP_BOMB";
101
102
  if (msg.includes("bomb") || msg.includes("\uD06C\uAE30 \uCD08\uACFC") || msg.includes("\uC555\uCD95 \uD574\uC81C")) return "DECOMPRESSION_BOMB";
102
103
  if (msg.includes("\uC774\uBBF8\uC9C0 \uAE30\uBC18")) return "IMAGE_BASED_PDF";
@@ -128,4 +129,4 @@ var HEADING_RATIO_H3 = 1.15;
128
129
 
129
130
 
130
131
  exports.VERSION = VERSION; exports.toArrayBuffer = toArrayBuffer; exports.KordocError = KordocError; exports.sanitizeError = sanitizeError; exports.isPathTraversal = isPathTraversal; exports.normalizeSectionHref = normalizeSectionHref; exports.compareSectionPaths = compareSectionPaths; exports.precheckZipSize = precheckZipSize; exports.stripDtd = stripDtd; exports.sanitizeHref = sanitizeHref; exports.safeMin = safeMin; exports.safeMax = safeMax; exports.classifyError = classifyError; exports.HEADING_RATIO_H1 = HEADING_RATIO_H1; exports.HEADING_RATIO_H2 = HEADING_RATIO_H2; exports.HEADING_RATIO_H3 = HEADING_RATIO_H3;
131
- //# sourceMappingURL=chunk-UQ6KDJWE.cjs.map
132
+ //# sourceMappingURL=chunk-WX7WQTOT.cjs.map
@@ -0,0 +1 @@
1
+ {"version":3,"sources":["d:\\AI_Project\\kordoc\\dist\\chunk-WX7WQTOT.cjs"],"names":[],"mappings":"AAAA;AACA,IAAI,QAAQ,EAAE,KAAK,EAAE,QAAQ,EAAE,WAAW;AAC1C,SAAS,aAAa,CAAC,GAAG,EAAE;AAC5B,EAAE,GAAG,CAAC,GAAG,CAAC,WAAW,IAAI,EAAE,GAAG,GAAG,CAAC,WAAW,IAAI,GAAG,CAAC,MAAM,CAAC,UAAU,EAAE;AACxE,IAAI,OAAO,GAAG,CAAC,MAAM;AACrB,EAAE;AACF,EAAE,OAAO,GAAG,CAAC,MAAM,CAAC,KAAK,CAAC,GAAG,CAAC,UAAU,EAAE,GAAG,CAAC,WAAW,EAAE,GAAG,CAAC,UAAU,CAAC;AAC1E;AACA,IAAI,YAAY,EAAE,MAAM,QAAQ,MAAM;AACtC,EAAE,WAAW,CAAC,OAAO,EAAE;AACvB,IAAI,KAAK,CAAC,OAAO,CAAC;AAClB,IAAI,IAAI,CAAC,KAAK,EAAE,aAAa;AAC7B,EAAE;AACF,CAAC;AACD,SAAS,aAAa,CAAC,GAAG,EAAE;AAC5B,EAAE,GAAG,CAAC,IAAI,WAAW,WAAW,EAAE,OAAO,GAAG,CAAC,OAAO;AACpD,EAAE,OAAO,0FAA0F;AACnG;AACA,SAAS,eAAe,CAAC,IAAI,EAAE;AAC/B,EAAE,GAAG,CAAC,IAAI,CAAC,QAAQ,CAAC,IAAI,CAAC,EAAE,OAAO,IAAI;AACtC,EAAE,MAAM,WAAW,EAAE,IAAI,CAAC,OAAO,CAAC,KAAK,EAAE,GAAG,CAAC;AAC7C,EAAE,MAAM,SAAS,EAAE,UAAU,CAAC,KAAK,CAAC,GAAG,CAAC;AACxC,EAAE,OAAO,QAAQ,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,GAAG,EAAE,IAAI,IAAI,EAAE,GAAG,UAAU,CAAC,UAAU,CAAC,GAAG,EAAE,GAAG,YAAY,CAAC,IAAI,CAAC,UAAU,CAAC;AACxG;AACA,SAAS,oBAAoB,CAAC,IAAI,EAAE;AACpC,EAAE,GAAG,CAAC,CAAC,IAAI,EAAE,OAAO,IAAI;AACxB,EAAE,IAAI,WAAW,EAAE,IAAI,CAAC,OAAO,CAAC,KAAK,EAAE,GAAG,CAAC,CAAC,OAAO,CAAC,MAAM,EAAE,EAAE,CAAC;AAC/D,EAAE,GAAG,CAAC,eAAe,CAAC,UAAU,CAAC,EAAE,OAAO,IAAI;AAC9C,EAAE,GAAG,CAAC,sBAAsB,CAAC,IAAI,CAAC,UAAU,CAAC,EAAE,WAAW,EAAE,YAAY,EAAE,UAAU;AACpF,EAAE,OAAO,6BAA6B,CAAC,IAAI,CAAC,UAAU,EAAE,EAAE,WAAW,EAAE,IAAI;AAC3E;AACA,SAAS,mBAAmB,CAAC,CAAC,EAAE,CAAC,EAAE;AACnC,EAAE,MAAM,GAAG,EAAE,MAAM,kCAAC,CAAC,mBAAC,KAAK,mBAAC,uBAAuB,CAAC,4BAAE,CAAC,CAAC,GAAE,UAAG,MAAM,CAAC,kBAAgB,CAAC;AACrF,EAAE,MAAM,GAAG,EAAE,MAAM,kCAAC,CAAC,qBAAC,KAAK,mBAAC,uBAAuB,CAAC,4BAAE,CAAC,CAAC,GAAE,UAAG,MAAM,CAAC,kBAAgB,CAAC;AACrF,EAAE,OAAO,GAAG,IAAI,GAAG,EAAE,CAAC,CAAC,aAAa,CAAC,CAAC,EAAE,EAAE,GAAG,EAAE,EAAE;AACjD;AACA,SAAS,eAAe,CAAC,MAAM,EAAE,oBAAoB,EAAE,IAAI,EAAE,KAAK,EAAE,IAAI,EAAE,WAAW,EAAE,GAAG,EAAE;AAC5F,EAAE,IAAI;AACN,IAAI,MAAM,KAAK,EAAE,IAAI,QAAQ,CAAC,MAAM,CAAC;AACrC,IAAI,MAAM,IAAI,EAAE,MAAM,CAAC,UAAU;AACjC,IAAI,IAAI,WAAW,EAAE,CAAC,CAAC;AACvB,IAAI,IAAI,CAAC,IAAI,EAAE,EAAE,IAAI,EAAE,EAAE,EAAE,EAAE,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,IAAI,EAAE,KAAK,CAAC,EAAE,CAAC,EAAE,EAAE;AAC/D,MAAM,GAAG,CAAC,IAAI,CAAC,SAAS,CAAC,CAAC,EAAE,IAAI,EAAE,IAAI,SAAS,EAAE;AACjD,QAAQ,WAAW,EAAE,CAAC;AACtB,QAAQ,KAAK;AACb,MAAM;AACN,IAAI;AACJ,IAAI,GAAG,CAAC,WAAW,EAAE,CAAC,EAAE,OAAO,EAAE,iBAAiB,EAAE,CAAC,EAAE,UAAU,EAAE,EAAE,CAAC;AACtE,IAAI,MAAM,WAAW,EAAE,IAAI,CAAC,SAAS,CAAC,WAAW,EAAE,EAAE,EAAE,IAAI,CAAC;AAC5D,IAAI,GAAG,CAAC,WAAW,EAAE,UAAU,EAAE;AACjC,MAAM,MAAM,IAAI,WAAW,CAAC,CAAC,4CAA4C,EAAE,UAAU,CAAC,eAAe,EAAE,UAAU,CAAC,CAAC,CAAC,CAAC;AACrH,IAAI;AACJ,IAAI,MAAM,OAAO,EAAE,IAAI,CAAC,SAAS,CAAC,WAAW,EAAE,EAAE,EAAE,IAAI,CAAC;AACxD,IAAI,MAAM,SAAS,EAAE,IAAI,CAAC,SAAS,CAAC,WAAW,EAAE,EAAE,EAAE,IAAI,CAAC;AAC1D,IAAI,GAAG,CAAC,SAAS,EAAE,OAAO,EAAE,GAAG,EAAE,OAAO,EAAE,iBAAiB,EAAE,CAAC,EAAE,WAAW,CAAC;AAC5E,IAAI,IAAI,kBAAkB,EAAE,CAAC;AAC7B,IAAI,IAAI,IAAI,EAAE,QAAQ;AACtB,IAAI,IAAI,CAAC,IAAI,EAAE,EAAE,CAAC,EAAE,EAAE,EAAE,WAAW,GAAG,IAAI,EAAE,GAAG,GAAG,SAAS,EAAE,MAAM,EAAE,CAAC,EAAE,EAAE;AAC1E,MAAM,GAAG,CAAC,IAAI,CAAC,SAAS,CAAC,GAAG,EAAE,IAAI,EAAE,IAAI,QAAQ,EAAE,KAAK;AACvD,MAAM,kBAAkB,GAAG,IAAI,CAAC,SAAS,CAAC,IAAI,EAAE,EAAE,EAAE,IAAI,CAAC;AACzD,MAAM,MAAM,QAAQ,EAAE,IAAI,CAAC,SAAS,CAAC,IAAI,EAAE,EAAE,EAAE,IAAI,CAAC;AACpD,MAAM,MAAM,SAAS,EAAE,IAAI,CAAC,SAAS,CAAC,IAAI,EAAE,EAAE,EAAE,IAAI,CAAC;AACrD,MAAM,MAAM,WAAW,EAAE,IAAI,CAAC,SAAS,CAAC,IAAI,EAAE,EAAE,EAAE,IAAI,CAAC;AACvD,MAAM,IAAI,GAAG,GAAG,EAAE,QAAQ,EAAE,SAAS,EAAE,UAAU;AACjD,IAAI;AACJ,IAAI,GAAG,CAAC,kBAAkB,EAAE,mBAAmB,EAAE;AACjD,MAAM,MAAM,IAAI,WAAW,CAAC,CAAC,kDAAkD,EAAE,CAAC,kBAAkB,EAAE,KAAK,EAAE,IAAI,CAAC,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,iBAAiB,EAAE,oBAAoB,EAAE,KAAK,EAAE,IAAI,CAAC,GAAG,CAAC,CAAC;AACxL,IAAI;AACJ,IAAI,OAAO,EAAE,iBAAiB,EAAE,WAAW,CAAC;AAC5C,EAAE,EAAE,MAAM,CAAC,GAAG,EAAE;AAChB,IAAI,GAAG,CAAC,IAAI,WAAW,WAAW,EAAE,MAAM,GAAG;AAC7C,IAAI,OAAO,EAAE,iBAAiB,EAAE,CAAC,EAAE,UAAU,EAAE,EAAE,CAAC;AAClD,EAAE;AACF;AACA,SAAS,QAAQ,CAAC,GAAG,EAAE;AACvB,EAAE,OAAO,GAAG,CAAC,OAAO,CAAC,SAAS,EAAE,EAAE,CAAC,CAAC,OAAO,CAAC,wCAAwC,EAAE,EAAE,CAAC;AACzF;AACA,IAAI,aAAa,EAAE,8BAA8B;AACjD,SAAS,YAAY,CAAC,IAAI,EAAE;AAC5B,EAAE,MAAM,QAAQ,EAAE,IAAI,CAAC,IAAI,CAAC,CAAC;AAC7B,EAAE,GAAG,CAAC,CAAC,QAAQ,GAAG,CAAC,YAAY,CAAC,IAAI,CAAC,OAAO,CAAC,EAAE,OAAO,IAAI;AAC1D,EAAE,OAAO,OAAO,CAAC,OAAO,CAAC,KAAK,EAAE,KAAK,CAAC,CAAC,OAAO,CAAC,KAAK,EAAE,KAAK,CAAC;AAC5D;AACA,SAAS,OAAO,CAAC,GAAG,EAAE;AACtB,EAAE,IAAI,IAAI,EAAE,QAAQ;AACpB,EAAE,IAAI,CAAC,IAAI,EAAE,EAAE,CAAC,EAAE,EAAE,EAAE,GAAG,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,GAAG,CAAC,GAAG,CAAC,CAAC,EAAE,EAAE,GAAG,EAAE,IAAI,EAAE,GAAG,CAAC,CAAC,CAAC;AACrE,EAAE,OAAO,GAAG;AACZ;AACA,SAAS,OAAO,CAAC,GAAG,EAAE;AACtB,EAAE,IAAI,IAAI,EAAE,CAAC,QAAQ;AACrB,EAAE,IAAI,CAAC,IAAI,EAAE,EAAE,CAAC,EAAE,EAAE,EAAE,GAAG,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,GAAG,CAAC,GAAG,CAAC,CAAC,EAAE,EAAE,GAAG,EAAE,IAAI,EAAE,GAAG,CAAC,CAAC,CAAC;AACrE,EAAE,OAAO,GAAG;AACZ;AACA,SAAS,aAAa,CAAC,GAAG,EAAE;AAC5B,EAAE,GAAG,CAAC,CAAC,CAAC,IAAI,WAAW,KAAK,CAAC,EAAE,OAAO,aAAa;AACnD,EAAE,MAAM,IAAI,EAAE,GAAG,CAAC,OAAO;AACzB,EAAE,GAAG,CAAC,GAAG,CAAC,QAAQ,CAAC,KAAK,CAAC,EAAE,OAAO,eAAe;AACjD,EAAE,GAAG,CAAC,GAAG,CAAC,QAAQ,CAAC,oBAAoB,EAAE,GAAG,GAAG,CAAC,QAAQ,CAAC,iCAAiC,CAAC,EAAE,OAAO,WAAW;AAC/G,EAAE,GAAG,CAAC,GAAG,CAAC,QAAQ,CAAC,qBAAqB,CAAC,EAAE,OAAO,oBAAoB;AACtE,EAAE,GAAG,CAAC,GAAG,CAAC,QAAQ,CAAC,uBAAuB,EAAE,GAAG,GAAG,CAAC,QAAQ,CAAC,+BAA+B,CAAC,EAAE,OAAO,kBAAkB;AACvH,EAAE,GAAG,CAAC,GAAG,CAAC,QAAQ,CAAC,UAAU,EAAE,GAAG,GAAG,CAAC,QAAQ,CAAC,kDAAkD,EAAE,GAAG,GAAG,CAAC,QAAQ,CAAC,4CAA4C,CAAC,EAAE,OAAO,UAAU;AACnL,EAAE,GAAG,CAAC,GAAG,CAAC,QAAQ,CAAC,MAAM,EAAE,GAAG,GAAG,CAAC,QAAQ,CAAC,2BAA2B,EAAE,GAAG,GAAG,CAAC,QAAQ,CAAC,2BAA2B,CAAC,EAAE,OAAO,oBAAoB;AACjJ,EAAE,GAAG,CAAC,GAAG,CAAC,QAAQ,CAAC,iCAAiC,CAAC,EAAE,OAAO,iBAAiB;AAC/E,EAAE,GAAG,CAAC,GAAG,CAAC,QAAQ,CAAC,cAAc,EAAE,GAAG,CAAC,GAAG,CAAC,QAAQ,CAAC,4BAA4B,EAAE,GAAG,GAAG,CAAC,QAAQ,CAAC,cAAc,CAAC,CAAC,EAAE,OAAO,aAAa;AACxI,EAAE,GAAG,CAAC,GAAG,CAAC,QAAQ,CAAC,0BAA0B,EAAE,GAAG,GAAG,CAAC,QAAQ,CAAC,kCAAkC,CAAC,EAAE,OAAO,WAAW;AACtH,EAAE,OAAO,aAAa;AACtB;AACA;AACA;AACA,IAAI,iBAAiB,EAAE,GAAG;AAC1B,IAAI,iBAAiB,EAAE,GAAG;AAC1B,IAAI,iBAAiB,EAAE,IAAI;AAC3B;AACA;AACE;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACF,+mBAAC","file":"D:\\AI_Project\\kordoc\\dist\\chunk-WX7WQTOT.cjs","sourcesContent":[null]}
@@ -30,4 +30,4 @@ function parsePageRange(spec, maxPages) {
30
30
 
31
31
 
32
32
  exports.parsePageRange = parsePageRange;
33
- //# sourceMappingURL=chunk-DCZVOIEO.cjs.map
33
+ //# sourceMappingURL=chunk-X34YWRL5.cjs.map
@@ -0,0 +1 @@
1
+ {"version":3,"sources":["d:\\AI_Project\\kordoc\\dist\\chunk-X34YWRL5.cjs"],"names":[],"mappings":"AAAA;AACA,SAAS,cAAc,CAAC,IAAI,EAAE,QAAQ,EAAE;AACxC,EAAE,MAAM,OAAO,kBAAkB,IAAI,GAAG,CAAC,CAAC;AAC1C,EAAE,GAAG,CAAC,SAAS,GAAG,CAAC,EAAE,OAAO,MAAM;AAClC,EAAE,GAAG,CAAC,KAAK,CAAC,OAAO,CAAC,IAAI,CAAC,EAAE;AAC3B,IAAI,IAAI,CAAC,MAAM,EAAE,GAAG,IAAI,EAAE;AAC1B,MAAM,MAAM,KAAK,EAAE,IAAI,CAAC,KAAK,CAAC,CAAC,CAAC;AAChC,MAAM,GAAG,CAAC,KAAK,GAAG,EAAE,GAAG,KAAK,GAAG,QAAQ,EAAE,MAAM,CAAC,GAAG,CAAC,IAAI,CAAC;AACzD,IAAI;AACJ,IAAI,OAAO,MAAM;AACjB,EAAE;AACF,EAAE,GAAG,CAAC,OAAO,KAAK,IAAI,SAAS,GAAG,IAAI,CAAC,IAAI,CAAC,EAAE,IAAI,EAAE,EAAE,OAAO,MAAM;AACnE,EAAE,MAAM,MAAM,EAAE,IAAI,CAAC,KAAK,CAAC,GAAG,CAAC;AAC/B,EAAE,IAAI,CAAC,MAAM,KAAK,GAAG,KAAK,EAAE;AAC5B,IAAI,MAAM,QAAQ,EAAE,IAAI,CAAC,IAAI,CAAC,CAAC;AAC/B,IAAI,GAAG,CAAC,CAAC,OAAO,EAAE,QAAQ;AAC1B,IAAI,MAAM,WAAW,EAAE,OAAO,CAAC,KAAK,CAAC,qBAAqB,CAAC;AAC3D,IAAI,GAAG,CAAC,UAAU,EAAE;AACpB,MAAM,MAAM,MAAM,EAAE,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,QAAQ,CAAC,UAAU,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC;AAC5D,MAAM,MAAM,IAAI,EAAE,IAAI,CAAC,GAAG,CAAC,QAAQ,EAAE,QAAQ,CAAC,UAAU,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC;AACjE,MAAM,IAAI,CAAC,IAAI,EAAE,EAAE,KAAK,EAAE,EAAE,GAAG,GAAG,EAAE,CAAC,EAAE,EAAE,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC;AACtD,IAAI,EAAE,KAAK;AACX,MAAM,MAAM,KAAK,EAAE,QAAQ,CAAC,OAAO,EAAE,EAAE,CAAC;AACxC,MAAM,GAAG,CAAC,CAAC,KAAK,CAAC,IAAI,EAAE,GAAG,KAAK,GAAG,EAAE,GAAG,KAAK,GAAG,QAAQ,EAAE,MAAM,CAAC,GAAG,CAAC,IAAI,CAAC;AACzE,IAAI;AACJ,EAAE;AACF,EAAE,OAAO,MAAM;AACf;AACA;AACA;AACE;AACF,wCAAC","file":"D:\\AI_Project\\kordoc\\dist\\chunk-X34YWRL5.cjs","sourcesContent":[null]}
@@ -243,4 +243,4 @@ export {
243
243
  ensureFormulaModels,
244
244
  ensureSingleModel
245
245
  };
246
- //# sourceMappingURL=chunk-UAFSPWU7.js.map
246
+ //# sourceMappingURL=chunk-YNJDNP44.js.map
@@ -0,0 +1 @@
1
+ {"version":3,"sources":["../src/pdf/formula/models.ts"],"sourcesContent":["/**\r\n * 수식 OCR 모델 자동 다운로드 + SHA-256 검증\r\n *\r\n * 캐시 위치: `~/.cache/kordoc/models/pix2text/`\r\n * 총 ~155MB (MFD 44MB + MFR encoder 87MB + MFR decoder 30MB + tokenizer 40KB)\r\n *\r\n * HF URL + SHA 는 Docufinder archive/rust-formula-ocr-wip 에서 검증 완료됨.\r\n * 모델 버전 갱신 시 URL + SHA 를 함께 갱신해야 한다.\r\n */\r\n\r\nimport { createHash } from \"crypto\"\r\nimport { createReadStream } from \"fs\"\r\nimport { mkdir, stat, unlink, rename } from \"fs/promises\"\r\nimport { createWriteStream } from \"fs\"\r\nimport { homedir } from \"os\"\r\nimport { join, dirname } from \"path\"\r\nimport { pipeline } from \"stream/promises\"\r\nimport { Readable } from \"stream\"\r\n\r\nexport interface ModelSpec {\r\n name: string\r\n filename: string\r\n url: string\r\n sha256: string\r\n sizeMb: number\r\n}\r\n\r\n/** 모델 스펙 (SHA-256 은 실제 다운로드로 검증됨 — 변경 금지) */\r\nexport const MFD_MODEL: ModelSpec = {\r\n name: \"Pix2Text MFD\",\r\n filename: \"mfd.onnx\",\r\n url: \"https://huggingface.co/breezedeus/pix2text-mfd/resolve/main/mfd-v20240618.onnx\",\r\n sha256: \"51a8854743b17ae654729af8db82a630c1ccfa06debf4856c8b28055f87d02c1\",\r\n sizeMb: 42,\r\n}\r\n\r\nexport const MFR_ENCODER_MODEL: ModelSpec = {\r\n name: \"Pix2Text MFR encoder\",\r\n filename: \"encoder_model.onnx\",\r\n url: \"https://huggingface.co/breezedeus/pix2text-mfr/resolve/main/encoder_model.onnx\",\r\n sha256: \"bd8d5c322792e9ec45793af5569e9748f82a3d728a9e00213dbfc56c1486f37d\",\r\n sizeMb: 87,\r\n}\r\n\r\nexport const MFR_DECODER_MODEL: ModelSpec = {\r\n name: \"Pix2Text MFR decoder\",\r\n filename: \"decoder_model.onnx\",\r\n url: \"https://huggingface.co/breezedeus/pix2text-mfr/resolve/main/decoder_model.onnx\",\r\n sha256: \"fd0f92d7a012f3dae41e1ac79421aea0ea888b5a66cb3f9a004e424f82f3daed\",\r\n sizeMb: 30,\r\n}\r\n\r\nexport const MFR_TOKENIZER: ModelSpec = {\r\n name: \"Pix2Text MFR tokenizer\",\r\n filename: \"tokenizer.json\",\r\n url: \"https://huggingface.co/breezedeus/pix2text-mfr/resolve/main/tokenizer.json\",\r\n sha256: \"3e2ab757277d22639bec28c9d7972e352d3d1dba223051fa674002dc5ab64df3\",\r\n sizeMb: 1,\r\n}\r\n\r\nexport const ALL_FORMULA_MODELS: ReadonlyArray<ModelSpec> = [\r\n MFD_MODEL,\r\n MFR_ENCODER_MODEL,\r\n MFR_DECODER_MODEL,\r\n MFR_TOKENIZER,\r\n]\r\n\r\n/**\r\n * 모델 캐시 하위 디렉토리 경로. 환경 변수 `KORDOC_MODEL_CACHE` 가 설정되면\r\n * 해당 경로를 우선 사용, 없으면 `~/.cache/kordoc/models/<subdir>/`.\r\n * (수식 OCR 외 텍스트 OCR 등 다른 모델군도 공용 — subdir 로 격리)\r\n */\r\nexport function getModelsDir(subdir: string): string {\r\n const override = process.env.KORDOC_MODEL_CACHE\r\n if (override && override.trim()) {\r\n return join(override, subdir)\r\n }\r\n return join(homedir(), \".cache\", \"kordoc\", \"models\", subdir)\r\n}\r\n\r\nexport function getFormulaModelsDir(): string {\r\n return getModelsDir(\"pix2text\")\r\n}\r\n\r\nexport interface ModelStatus {\r\n spec: ModelSpec\r\n localPath: string\r\n exists: boolean\r\n /** SHA 검증까지 끝났는가 (exists && valid) */\r\n verified: boolean\r\n /** 검증 실패 시 이유 */\r\n invalidReason?: string\r\n}\r\n\r\n/** 지정 모델군의 존재/유효성 상태 반환 (다운로드 없이 확인만) */\r\nexport async function getModelStatusIn(dir: string, specs: ReadonlyArray<ModelSpec>): Promise<ModelStatus[]> {\r\n const result: ModelStatus[] = []\r\n for (const spec of specs) {\r\n const localPath = join(dir, spec.filename)\r\n let exists = false\r\n try {\r\n const s = await stat(localPath)\r\n exists = s.isFile() && s.size > 0\r\n } catch {\r\n exists = false\r\n }\r\n if (!exists) {\r\n result.push({ spec, localPath, exists: false, verified: false })\r\n continue\r\n }\r\n try {\r\n const actual = await sha256OfFile(localPath)\r\n if (actual === spec.sha256) {\r\n result.push({ spec, localPath, exists: true, verified: true })\r\n } else {\r\n result.push({\r\n spec,\r\n localPath,\r\n exists: true,\r\n verified: false,\r\n invalidReason: `SHA256 mismatch: expected ${spec.sha256}, got ${actual}`,\r\n })\r\n }\r\n } catch (e) {\r\n result.push({\r\n spec,\r\n localPath,\r\n exists: true,\r\n verified: false,\r\n invalidReason: `SHA compute failed: ${(e as Error).message}`,\r\n })\r\n }\r\n }\r\n return result\r\n}\r\n\r\n/** 모든 수식 모델의 존재/유효성 상태 반환 (다운로드 없이 확인만) */\r\nexport async function getFormulaModelStatus(): Promise<ModelStatus[]> {\r\n return getModelStatusIn(getFormulaModelsDir(), ALL_FORMULA_MODELS)\r\n}\r\n\r\nexport interface DownloadProgress {\r\n spec: ModelSpec\r\n /** 현재까지 다운로드한 바이트 */\r\n downloaded: number\r\n /** 전체 바이트 (알 수 있으면) */\r\n total: number | null\r\n /** \"download\" | \"verify\" | \"done\" | \"skip\" */\r\n phase: \"download\" | \"verify\" | \"done\" | \"skip\" | \"error\"\r\n message?: string\r\n}\r\n\r\nexport type ProgressHandler = (p: DownloadProgress) => void\r\n\r\n/**\r\n * 지정 모델군을 다운로드/검증한다.\r\n * - 이미 있고 SHA 일치 → skip\r\n * - 없음 → 다운로드 후 SHA 검증. 실패 시 파일 삭제.\r\n */\r\nexport async function ensureModelsIn(dir: string, specs: ReadonlyArray<ModelSpec>, onProgress?: ProgressHandler): Promise<void> {\r\n await mkdir(dir, { recursive: true })\r\n\r\n for (const spec of specs) {\r\n const localPath = join(dir, spec.filename)\r\n\r\n if (await isExistingValid(localPath, spec.sha256)) {\r\n onProgress?.({\r\n spec,\r\n downloaded: 0,\r\n total: null,\r\n phase: \"skip\",\r\n message: \"이미 존재 + SHA 일치\",\r\n })\r\n continue\r\n }\r\n\r\n // 기존 파일 있지만 SHA 불일치 → 삭제\r\n try {\r\n await unlink(localPath)\r\n } catch {\r\n // 없을 수 있음\r\n }\r\n\r\n await downloadToFile(spec, localPath, onProgress)\r\n }\r\n}\r\n\r\n/** 필요한 모든 수식 모델을 다운로드/검증한다. */\r\nexport async function ensureFormulaModels(onProgress?: ProgressHandler): Promise<void> {\r\n return ensureModelsIn(getFormulaModelsDir(), ALL_FORMULA_MODELS, onProgress)\r\n}\r\n\r\n/** 단일 모델만 확인/다운로드 (진단 UI 에서 개별 상태 새로고침용) */\r\nexport async function ensureSingleModel(spec: ModelSpec, onProgress?: ProgressHandler): Promise<void> {\r\n const dir = getFormulaModelsDir()\r\n await mkdir(dir, { recursive: true })\r\n const localPath = join(dir, spec.filename)\r\n if (await isExistingValid(localPath, spec.sha256)) {\r\n onProgress?.({ spec, downloaded: 0, total: null, phase: \"skip\" })\r\n return\r\n }\r\n try {\r\n await unlink(localPath)\r\n } catch {\r\n // 재다운로드 전 손상/잔존 파일 정리 best-effort — 애초에 없으면 실패해도 무방\r\n }\r\n await downloadToFile(spec, localPath, onProgress)\r\n}\r\n\r\nasync function isExistingValid(localPath: string, sha256Expected: string): Promise<boolean> {\r\n try {\r\n const s = await stat(localPath)\r\n if (!s.isFile() || s.size === 0) return false\r\n } catch {\r\n return false\r\n }\r\n try {\r\n const actual = await sha256OfFile(localPath)\r\n return actual === sha256Expected\r\n } catch {\r\n return false\r\n }\r\n}\r\n\r\nasync function downloadToFile(\r\n spec: ModelSpec,\r\n localPath: string,\r\n onProgress?: ProgressHandler,\r\n): Promise<void> {\r\n // 먼저 .part 로 받고 검증 후 rename — 중단된 다운로드가 \"정상 파일\"로 오인되는 걸 방지\r\n const partPath = `${localPath}.part`\r\n await mkdir(dirname(localPath), { recursive: true })\r\n\r\n const resp = await fetch(spec.url, {\r\n headers: {\r\n // HF CDN 은 UA 없으면 가끔 403 을 뱉는다\r\n \"User-Agent\": \"kordoc-formula-ocr/1.0 (+https://github.com/chrisryugj/kordoc)\",\r\n },\r\n })\r\n if (!resp.ok || !resp.body) {\r\n throw new Error(\r\n `${spec.name} 다운로드 실패: HTTP ${resp.status} ${resp.statusText} (${spec.url})`,\r\n )\r\n }\r\n\r\n const lenHeader = resp.headers.get(\"content-length\")\r\n const total = lenHeader ? Number.parseInt(lenHeader, 10) : null\r\n let downloaded = 0\r\n\r\n const ws = createWriteStream(partPath)\r\n try {\r\n const reader = Readable.fromWeb(resp.body as unknown as import(\"stream/web\").ReadableStream)\r\n reader.on(\"data\", (chunk: Buffer | Uint8Array) => {\r\n downloaded += chunk.length\r\n onProgress?.({\r\n spec,\r\n downloaded,\r\n total,\r\n phase: \"download\",\r\n })\r\n })\r\n await pipeline(reader, ws)\r\n } catch (e) {\r\n try {\r\n await unlink(partPath)\r\n } catch {\r\n // .part 임시파일 정리 best-effort — 본 에러(스트리밍 실패)를 가리지 않도록 무시\r\n }\r\n throw new Error(`${spec.name} 스트리밍 실패: ${(e as Error).message}`)\r\n }\r\n\r\n onProgress?.({\r\n spec,\r\n downloaded,\r\n total,\r\n phase: \"verify\",\r\n })\r\n\r\n // SHA 검증\r\n let actual: string\r\n try {\r\n actual = await sha256OfFile(partPath)\r\n } catch (e) {\r\n try {\r\n await unlink(partPath)\r\n } catch {\r\n // .part 임시파일 정리 best-effort — 본 에러(SHA 계산 실패)를 가리지 않도록 무시\r\n }\r\n throw new Error(`${spec.name} SHA 계산 실패: ${(e as Error).message}`)\r\n }\r\n\r\n if (actual !== spec.sha256) {\r\n try {\r\n await unlink(partPath)\r\n } catch {\r\n // 오염된 .part 정리 best-effort — 본 에러(SHA mismatch)를 가리지 않도록 무시\r\n }\r\n throw new Error(\r\n `${spec.name} SHA256 mismatch: expected ${spec.sha256}, got ${actual} — 모델 URL 이 오염되었거나 전송 중 손상되었습니다.`,\r\n )\r\n }\r\n\r\n await rename(partPath, localPath)\r\n onProgress?.({\r\n spec,\r\n downloaded,\r\n total,\r\n phase: \"done\",\r\n })\r\n}\r\n\r\nasync function sha256OfFile(p: string): Promise<string> {\r\n const h = createHash(\"sha256\")\r\n const stream = createReadStream(p)\r\n await pipeline(stream, async function* (src) {\r\n for await (const chunk of src) {\r\n h.update(chunk)\r\n // yield 안 함 (소비만)\r\n }\r\n })\r\n return h.digest(\"hex\")\r\n}\r\n"],"mappings":";;;AAUA,SAAS,kBAAkB;AAC3B,SAAS,wBAAwB;AACjC,SAAS,OAAO,MAAM,QAAQ,cAAc;AAC5C,SAAS,yBAAyB;AAClC,SAAS,eAAe;AACxB,SAAS,MAAM,eAAe;AAC9B,SAAS,gBAAgB;AACzB,SAAS,gBAAgB;AAWlB,IAAM,YAAuB;AAAA,EAClC,MAAM;AAAA,EACN,UAAU;AAAA,EACV,KAAK;AAAA,EACL,QAAQ;AAAA,EACR,QAAQ;AACV;AAEO,IAAM,oBAA+B;AAAA,EAC1C,MAAM;AAAA,EACN,UAAU;AAAA,EACV,KAAK;AAAA,EACL,QAAQ;AAAA,EACR,QAAQ;AACV;AAEO,IAAM,oBAA+B;AAAA,EAC1C,MAAM;AAAA,EACN,UAAU;AAAA,EACV,KAAK;AAAA,EACL,QAAQ;AAAA,EACR,QAAQ;AACV;AAEO,IAAM,gBAA2B;AAAA,EACtC,MAAM;AAAA,EACN,UAAU;AAAA,EACV,KAAK;AAAA,EACL,QAAQ;AAAA,EACR,QAAQ;AACV;AAEO,IAAM,qBAA+C;AAAA,EAC1D;AAAA,EACA;AAAA,EACA;AAAA,EACA;AACF;AAOO,SAAS,aAAa,QAAwB;AACnD,QAAM,WAAW,QAAQ,IAAI;AAC7B,MAAI,YAAY,SAAS,KAAK,GAAG;AAC/B,WAAO,KAAK,UAAU,MAAM;AAAA,EAC9B;AACA,SAAO,KAAK,QAAQ,GAAG,UAAU,UAAU,UAAU,MAAM;AAC7D;AAEO,SAAS,sBAA8B;AAC5C,SAAO,aAAa,UAAU;AAChC;AAaA,eAAsB,iBAAiB,KAAa,OAAyD;AAC3G,QAAM,SAAwB,CAAC;AAC/B,aAAW,QAAQ,OAAO;AACxB,UAAM,YAAY,KAAK,KAAK,KAAK,QAAQ;AACzC,QAAI,SAAS;AACb,QAAI;AACF,YAAM,IAAI,MAAM,KAAK,SAAS;AAC9B,eAAS,EAAE,OAAO,KAAK,EAAE,OAAO;AAAA,IAClC,QAAQ;AACN,eAAS;AAAA,IACX;AACA,QAAI,CAAC,QAAQ;AACX,aAAO,KAAK,EAAE,MAAM,WAAW,QAAQ,OAAO,UAAU,MAAM,CAAC;AAC/D;AAAA,IACF;AACA,QAAI;AACF,YAAM,SAAS,MAAM,aAAa,SAAS;AAC3C,UAAI,WAAW,KAAK,QAAQ;AAC1B,eAAO,KAAK,EAAE,MAAM,WAAW,QAAQ,MAAM,UAAU,KAAK,CAAC;AAAA,MAC/D,OAAO;AACL,eAAO,KAAK;AAAA,UACV;AAAA,UACA;AAAA,UACA,QAAQ;AAAA,UACR,UAAU;AAAA,UACV,eAAe,6BAA6B,KAAK,MAAM,SAAS,MAAM;AAAA,QACxE,CAAC;AAAA,MACH;AAAA,IACF,SAAS,GAAG;AACV,aAAO,KAAK;AAAA,QACV;AAAA,QACA;AAAA,QACA,QAAQ;AAAA,QACR,UAAU;AAAA,QACV,eAAe,uBAAwB,EAAY,OAAO;AAAA,MAC5D,CAAC;AAAA,IACH;AAAA,EACF;AACA,SAAO;AACT;AAGA,eAAsB,wBAAgD;AACpE,SAAO,iBAAiB,oBAAoB,GAAG,kBAAkB;AACnE;AAoBA,eAAsB,eAAe,KAAa,OAAiC,YAA6C;AAC9H,QAAM,MAAM,KAAK,EAAE,WAAW,KAAK,CAAC;AAEpC,aAAW,QAAQ,OAAO;AACxB,UAAM,YAAY,KAAK,KAAK,KAAK,QAAQ;AAEzC,QAAI,MAAM,gBAAgB,WAAW,KAAK,MAAM,GAAG;AACjD,mBAAa;AAAA,QACX;AAAA,QACA,YAAY;AAAA,QACZ,OAAO;AAAA,QACP,OAAO;AAAA,QACP,SAAS;AAAA,MACX,CAAC;AACD;AAAA,IACF;AAGA,QAAI;AACF,YAAM,OAAO,SAAS;AAAA,IACxB,QAAQ;AAAA,IAER;AAEA,UAAM,eAAe,MAAM,WAAW,UAAU;AAAA,EAClD;AACF;AAGA,eAAsB,oBAAoB,YAA6C;AACrF,SAAO,eAAe,oBAAoB,GAAG,oBAAoB,UAAU;AAC7E;AAGA,eAAsB,kBAAkB,MAAiB,YAA6C;AACpG,QAAM,MAAM,oBAAoB;AAChC,QAAM,MAAM,KAAK,EAAE,WAAW,KAAK,CAAC;AACpC,QAAM,YAAY,KAAK,KAAK,KAAK,QAAQ;AACzC,MAAI,MAAM,gBAAgB,WAAW,KAAK,MAAM,GAAG;AACjD,iBAAa,EAAE,MAAM,YAAY,GAAG,OAAO,MAAM,OAAO,OAAO,CAAC;AAChE;AAAA,EACF;AACA,MAAI;AACF,UAAM,OAAO,SAAS;AAAA,EACxB,QAAQ;AAAA,EAER;AACA,QAAM,eAAe,MAAM,WAAW,UAAU;AAClD;AAEA,eAAe,gBAAgB,WAAmB,gBAA0C;AAC1F,MAAI;AACF,UAAM,IAAI,MAAM,KAAK,SAAS;AAC9B,QAAI,CAAC,EAAE,OAAO,KAAK,EAAE,SAAS,EAAG,QAAO;AAAA,EAC1C,QAAQ;AACN,WAAO;AAAA,EACT;AACA,MAAI;AACF,UAAM,SAAS,MAAM,aAAa,SAAS;AAC3C,WAAO,WAAW;AAAA,EACpB,QAAQ;AACN,WAAO;AAAA,EACT;AACF;AAEA,eAAe,eACb,MACA,WACA,YACe;AAEf,QAAM,WAAW,GAAG,SAAS;AAC7B,QAAM,MAAM,QAAQ,SAAS,GAAG,EAAE,WAAW,KAAK,CAAC;AAEnD,QAAM,OAAO,MAAM,MAAM,KAAK,KAAK;AAAA,IACjC,SAAS;AAAA;AAAA,MAEP,cAAc;AAAA,IAChB;AAAA,EACF,CAAC;AACD,MAAI,CAAC,KAAK,MAAM,CAAC,KAAK,MAAM;AAC1B,UAAM,IAAI;AAAA,MACR,GAAG,KAAK,IAAI,gDAAkB,KAAK,MAAM,IAAI,KAAK,UAAU,KAAK,KAAK,GAAG;AAAA,IAC3E;AAAA,EACF;AAEA,QAAM,YAAY,KAAK,QAAQ,IAAI,gBAAgB;AACnD,QAAM,QAAQ,YAAY,OAAO,SAAS,WAAW,EAAE,IAAI;AAC3D,MAAI,aAAa;AAEjB,QAAM,KAAK,kBAAkB,QAAQ;AACrC,MAAI;AACF,UAAM,SAAS,SAAS,QAAQ,KAAK,IAAsD;AAC3F,WAAO,GAAG,QAAQ,CAAC,UAA+B;AAChD,oBAAc,MAAM;AACpB,mBAAa;AAAA,QACX;AAAA,QACA;AAAA,QACA;AAAA,QACA,OAAO;AAAA,MACT,CAAC;AAAA,IACH,CAAC;AACD,UAAM,SAAS,QAAQ,EAAE;AAAA,EAC3B,SAAS,GAAG;AACV,QAAI;AACF,YAAM,OAAO,QAAQ;AAAA,IACvB,QAAQ;AAAA,IAER;AACA,UAAM,IAAI,MAAM,GAAG,KAAK,IAAI,2CAAc,EAAY,OAAO,EAAE;AAAA,EACjE;AAEA,eAAa;AAAA,IACX;AAAA,IACA;AAAA,IACA;AAAA,IACA,OAAO;AAAA,EACT,CAAC;AAGD,MAAI;AACJ,MAAI;AACF,aAAS,MAAM,aAAa,QAAQ;AAAA,EACtC,SAAS,GAAG;AACV,QAAI;AACF,YAAM,OAAO,QAAQ;AAAA,IACvB,QAAQ;AAAA,IAER;AACA,UAAM,IAAI,MAAM,GAAG,KAAK,IAAI,mCAAgB,EAAY,OAAO,EAAE;AAAA,EACnE;AAEA,MAAI,WAAW,KAAK,QAAQ;AAC1B,QAAI;AACF,YAAM,OAAO,QAAQ;AAAA,IACvB,QAAQ;AAAA,IAER;AACA,UAAM,IAAI;AAAA,MACR,GAAG,KAAK,IAAI,8BAA8B,KAAK,MAAM,SAAS,MAAM;AAAA,IACtE;AAAA,EACF;AAEA,QAAM,OAAO,UAAU,SAAS;AAChC,eAAa;AAAA,IACX;AAAA,IACA;AAAA,IACA;AAAA,IACA,OAAO;AAAA,EACT,CAAC;AACH;AAEA,eAAe,aAAa,GAA4B;AACtD,QAAM,IAAI,WAAW,QAAQ;AAC7B,QAAM,SAAS,iBAAiB,CAAC;AACjC,QAAM,SAAS,QAAQ,iBAAiB,KAAK;AAC3C,qBAAiB,SAAS,KAAK;AAC7B,QAAE,OAAO,KAAK;AAAA,IAEhB;AAAA,EACF,CAAC;AACD,SAAO,EAAE,OAAO,KAAK;AACvB;","names":[]}
@@ -1,7 +1,7 @@
1
1
  #!/usr/bin/env node
2
2
 
3
3
  // src/utils.ts
4
- var VERSION = true ? "4.5.0" : "0.0.0-dev";
4
+ var VERSION = true ? "4.6.0" : "0.0.0-dev";
5
5
  function toArrayBuffer(buf) {
6
6
  if (buf.byteOffset === 0 && buf.byteLength === buf.buffer.byteLength) {
7
7
  return buf.buffer;
@@ -75,7 +75,7 @@ function precheckZipSize(buffer, maxUncompressedSize = 256 * 1024 * 1024, maxEnt
75
75
  }
76
76
  }
77
77
  function stripDtd(xml) {
78
- return xml.replace(/<!DOCTYPE\s[^[>]*(\[[\s\S]*?\])?\s*>/gi, "");
78
+ return xml.replace(/^\uFEFF/, "").replace(/<!DOCTYPE\s[^[>]*(\[[\s\S]*?\])?\s*>/gi, "");
79
79
  }
80
80
  var SAFE_HREF_RE = /^(?:https?:|mailto:|tel:|#)/i;
81
81
  function sanitizeHref(href) {
@@ -99,6 +99,7 @@ function classifyError(err) {
99
99
  if (msg.includes("DRM")) return "DRM_PROTECTED";
100
100
  if (msg.includes("\uC554\uD638\uD654") || msg.includes("\uC554\uD638\uB85C \uBCF4\uD638")) return "ENCRYPTED";
101
101
  if (msg.includes("optional dependency")) return "MISSING_DEPENDENCY";
102
+ if (msg.includes("Invalid string length") || msg.includes("Cannot create a string longer")) return "OUTPUT_TOO_LARGE";
102
103
  if (msg.includes("ZIP bomb") || msg.includes("ZIP \uBE44\uC555\uCD95 \uD06C\uAE30 \uCD08\uACFC") || msg.includes("ZIP \uC5D4\uD2B8\uB9AC \uC218 \uCD08\uACFC")) return "ZIP_BOMB";
103
104
  if (msg.includes("bomb") || msg.includes("\uD06C\uAE30 \uCD08\uACFC") || msg.includes("\uC555\uCD95 \uD574\uC81C")) return "DECOMPRESSION_BOMB";
104
105
  if (msg.includes("\uC774\uBBF8\uC9C0 \uAE30\uBC18")) return "IMAGE_BASED_PDF";
@@ -122,4 +123,4 @@ export {
122
123
  safeMax,
123
124
  classifyError
124
125
  };
125
- //# sourceMappingURL=chunk-TTEHX4FN.js.map
126
+ //# sourceMappingURL=chunk-ZYQ5X6PG.js.map
@@ -0,0 +1 @@
1
+ {"version":3,"sources":["../src/utils.ts"],"sourcesContent":["/** kordoc 공용 유틸리티 */\r\n\r\n/** 빌드 타임에 tsup define으로 주입되는 버전 */\r\ndeclare const __KORDOC_VERSION__: string\r\nexport const VERSION: string = typeof __KORDOC_VERSION__ !== \"undefined\" ? __KORDOC_VERSION__ : \"0.0.0-dev\"\r\n\r\n/**\r\n * Node.js Buffer → ArrayBuffer 변환\r\n * pool Buffer의 공유 ArrayBuffer 문제를 안전하게 처리.\r\n * offset=0이고 전체 ArrayBuffer를 차지하면 복사 없이 직접 반환.\r\n */\r\nexport function toArrayBuffer(buf: Buffer): ArrayBuffer {\r\n if (buf.byteOffset === 0 && buf.byteLength === buf.buffer.byteLength) {\r\n return buf.buffer as ArrayBuffer\r\n }\r\n return buf.buffer.slice(buf.byteOffset, buf.byteOffset + buf.byteLength) as ArrayBuffer\r\n}\r\n\r\n/**\r\n * kordoc 내부 에러 클래스 — 사용자에게 노출해도 안전한 메시지만 포함.\r\n * MCP 에러 정제에서 instanceof로 판별하여 allowlist 패턴 매칭 없이 안전하게 통과.\r\n */\r\nexport class KordocError extends Error {\r\n constructor(message: string) {\r\n super(message)\r\n this.name = \"KordocError\"\r\n }\r\n}\r\n\r\n/**\r\n * 에러 메시지 정제 — KordocError는 그대로, 나머지는 일반 메시지로 대체.\r\n * 파일시스템 경로, 스택 트레이스 등 내부 정보 노출 방지.\r\n */\r\nexport function sanitizeError(err: unknown): string {\r\n if (err instanceof KordocError) return err.message\r\n return \"문서 처리 중 오류가 발생했습니다\"\r\n}\r\n\r\n/**\r\n * ZIP 엔트리 경로의 경로 순회 여부 판별.\r\n * 백슬래시 정규화, .., 절대경로, Windows 드라이브 문자 모두 차단.\r\n */\r\nexport function isPathTraversal(name: string): boolean {\r\n if (name.includes(\"\\x00\")) return true\r\n const normalized = name.replace(/\\\\/g, \"/\")\r\n const segments = normalized.split(\"/\")\r\n return segments.some(s => s === \"..\") || normalized.startsWith(\"/\") || /^[A-Za-z]:/.test(normalized)\r\n}\r\n\r\n// ─── HWPX 섹션 href 해석 (parser/roundtrip 공용) ────────────────────\r\n\r\n/**\r\n * manifest href를 본문 섹션 경로(`Contents/sectionN.xml`)로 정규화.\r\n * 비본문 XML(header/script/settings)·path traversal·드라이브 경로는 null로 거른다.\r\n * 백슬래시 구분자를 슬래시로 통일해 parser와 roundtrip이 동일한 목록을 만든다.\r\n */\r\nexport function normalizeSectionHref(href: string): string | null {\r\n if (!href) return null\r\n let normalized = href.replace(/\\\\/g, \"/\").replace(/^\\/+/, \"\")\r\n if (isPathTraversal(normalized)) return null\r\n if (/^[Ss]ection\\d+\\.xml$/.test(normalized)) normalized = \"Contents/\" + normalized\r\n return /(?:^|\\/)[Ss]ection\\d+\\.xml$/.test(normalized) ? normalized : null\r\n}\r\n\r\n/** sectionN.xml을 N 숫자 순서로 정렬 (section10 > section2). */\r\nexport function compareSectionPaths(a: string, b: string): number {\r\n const ai = Number(a.match(/[Ss]ection(\\d+)\\.xml$/)?.[1] ?? Number.MAX_SAFE_INTEGER)\r\n const bi = Number(b.match(/[Ss]ection(\\d+)\\.xml$/)?.[1] ?? Number.MAX_SAFE_INTEGER)\r\n return ai === bi ? a.localeCompare(b) : ai - bi\r\n}\r\n\r\n// ─── ZIP 안전 로딩 (ZIP bomb 방지) ────────────────────\r\n\r\n/**\r\n * ZIP bomb 사전 검사 — Central Directory에서 비압축 합계와 엔트리 수 확인.\r\n * HWPX/XLSX/DOCX 등 모든 ZIP 기반 포맷에서 공통 사용.\r\n */\r\nexport function precheckZipSize(\r\n buffer: ArrayBuffer,\r\n maxUncompressedSize = 256 * 1024 * 1024, // parser-shared MAX_DECOMPRESS_SIZE 와 동기\r\n maxEntries = 500,\r\n): { totalUncompressed: number; entryCount: number } {\r\n try {\r\n const data = new DataView(buffer)\r\n const len = buffer.byteLength\r\n // EOCD 시그니처 역방향 스캔\r\n let eocdOffset = -1\r\n for (let i = len - 22; i >= Math.max(0, len - 65557); i--) {\r\n if (data.getUint32(i, true) === 0x06054b50) { eocdOffset = i; break }\r\n }\r\n if (eocdOffset < 0) return { totalUncompressed: 0, entryCount: 0 }\r\n\r\n const entryCount = data.getUint16(eocdOffset + 10, true)\r\n if (entryCount > maxEntries) {\r\n throw new KordocError(`ZIP 엔트리 수 초과: ${entryCount} (최대 ${maxEntries})`)\r\n }\r\n\r\n const cdSize = data.getUint32(eocdOffset + 12, true)\r\n const cdOffset = data.getUint32(eocdOffset + 16, true)\r\n if (cdOffset + cdSize > len) return { totalUncompressed: 0, entryCount }\r\n\r\n let totalUncompressed = 0\r\n let pos = cdOffset\r\n for (let i = 0; i < entryCount && pos + 46 <= cdOffset + cdSize; i++) {\r\n if (data.getUint32(pos, true) !== 0x02014b50) break\r\n totalUncompressed += data.getUint32(pos + 24, true)\r\n const nameLen = data.getUint16(pos + 28, true)\r\n const extraLen = data.getUint16(pos + 30, true)\r\n const commentLen = data.getUint16(pos + 32, true)\r\n pos += 46 + nameLen + extraLen + commentLen\r\n }\r\n\r\n if (totalUncompressed > maxUncompressedSize) {\r\n throw new KordocError(`ZIP 비압축 크기 초과: ${(totalUncompressed / 1024 / 1024).toFixed(1)}MB (최대 ${maxUncompressedSize / 1024 / 1024}MB)`)\r\n }\r\n\r\n return { totalUncompressed, entryCount }\r\n } catch (err) {\r\n if (err instanceof KordocError) throw err\r\n return { totalUncompressed: 0, entryCount: 0 }\r\n }\r\n}\r\n\r\n/**\r\n * XML 파싱 전 프롤로그 정리 — 선두 BOM 제거 + DOCTYPE 제거(XXE/Billion Laughs 방지).\r\n *\r\n * BOM: 일부 OpenXML 라이터가 `[Content_Types].xml`·`.rels` 앞에 EF BB BF를 붙인다.\r\n * XML 스펙상 적법(엔티티는 BOM으로 시작할 수 있음)이고 엑셀·리브레오피스는 여는데,\r\n * xmldom은 \"선언이 문서 시작이 아님\" fatalError로 파일 전체를 거부했다 (#63).\r\n */\r\nexport function stripDtd(xml: string): string {\r\n return xml.replace(/^\\uFEFF/, \"\").replace(/<!DOCTYPE\\s[^[>]*(\\[[\\s\\S]*?\\])?\\s*>/gi, \"\")\r\n}\r\n\r\n/** 하이퍼링크 URL 살균 — javascript: 등 XSS 위험 스킴 차단 */\r\nconst SAFE_HREF_RE = /^(?:https?:|mailto:|tel:|#)/i\r\nexport function sanitizeHref(href: string): string | null {\r\n const trimmed = href.trim()\r\n if (!trimmed || !SAFE_HREF_RE.test(trimmed)) return null\r\n // 괄호 percent-encoding — 마크다운 링크 목적지에서 불균형 ')'는 링크를 조기 종료시켜\r\n // 문법이 깨진다 (원문 하이퍼링크에 ')'가 박힌 실문서 존재). %28/%29는 의미 동일.\r\n return trimmed.replace(/\\(/g, \"%28\").replace(/\\)/g, \"%29\")\r\n}\r\n\r\n// ─── 안전한 min/max (스택 오버플로 방지) ─────────────\r\n\r\n/** Math.min(...arr) 대체 — 대형 배열에서 스택 오버플로 방지 */\r\nexport function safeMin(arr: number[]): number {\r\n let min = Infinity\r\n for (let i = 0; i < arr.length; i++) if (arr[i] < min) min = arr[i]\r\n return min\r\n}\r\n\r\n/** Math.max(...arr) 대체 — 대형 배열에서 스택 오버플로 방지 */\r\nexport function safeMax(arr: number[]): number {\r\n let max = -Infinity\r\n for (let i = 0; i < arr.length; i++) if (arr[i] > max) max = arr[i]\r\n return max\r\n}\r\n\r\n// ─── 에러 분류 ──────────────────────────────────────\r\n\r\nimport type { ErrorCode } from \"./types.js\"\r\n\r\n/** 에러를 구조화된 ErrorCode로 분류 — KordocError 메시지 패턴 매칭 */\r\nexport function classifyError(err: unknown): ErrorCode {\r\n if (!(err instanceof Error)) return \"PARSE_ERROR\"\r\n const msg = err.message\r\n // DRM 을 암호화보다 먼저 — \"DRM 암호화된 HWPX…\" 류 메시지가 ENCRYPTED 로 떨어지지 않도록\r\n if (msg.includes(\"DRM\")) return \"DRM_PROTECTED\"\r\n if (msg.includes(\"암호화\") || msg.includes(\"암호로 보호\")) return \"ENCRYPTED\"\r\n if (msg.includes(\"optional dependency\")) return \"MISSING_DEPENDENCY\"\r\n // V8 문자열 한계 초과 — 이미지 base64 총량이 큰 문서의 JSON 직렬화 (#65)\r\n if (msg.includes(\"Invalid string length\") || msg.includes(\"Cannot create a string longer\")) return \"OUTPUT_TOO_LARGE\"\r\n if (msg.includes(\"ZIP bomb\") || msg.includes(\"ZIP 비압축 크기 초과\") || msg.includes(\"ZIP 엔트리 수 초과\")) return \"ZIP_BOMB\"\r\n if (msg.includes(\"bomb\") || msg.includes(\"크기 초과\") || msg.includes(\"압축 해제\")) return \"DECOMPRESSION_BOMB\"\r\n if (msg.includes(\"이미지 기반\")) return \"IMAGE_BASED_PDF\"\r\n if (msg.includes(\"섹션\") && (msg.includes(\"찾을 수 없\") || msg.includes(\"없음\"))) return \"NO_SECTIONS\"\r\n if (msg.includes(\"시그니처\") || msg.includes(\"복구할 수 없\")) return \"CORRUPTED\"\r\n return \"PARSE_ERROR\"\r\n}\r\n"],"mappings":";;;AAIO,IAAM,UAAkB,OAA4C,UAAqB;AAOzF,SAAS,cAAc,KAA0B;AACtD,MAAI,IAAI,eAAe,KAAK,IAAI,eAAe,IAAI,OAAO,YAAY;AACpE,WAAO,IAAI;AAAA,EACb;AACA,SAAO,IAAI,OAAO,MAAM,IAAI,YAAY,IAAI,aAAa,IAAI,UAAU;AACzE;AAMO,IAAM,cAAN,cAA0B,MAAM;AAAA,EACrC,YAAY,SAAiB;AAC3B,UAAM,OAAO;AACb,SAAK,OAAO;AAAA,EACd;AACF;AAMO,SAAS,cAAc,KAAsB;AAClD,MAAI,eAAe,YAAa,QAAO,IAAI;AAC3C,SAAO;AACT;AAMO,SAAS,gBAAgB,MAAuB;AACrD,MAAI,KAAK,SAAS,IAAM,EAAG,QAAO;AAClC,QAAM,aAAa,KAAK,QAAQ,OAAO,GAAG;AAC1C,QAAM,WAAW,WAAW,MAAM,GAAG;AACrC,SAAO,SAAS,KAAK,OAAK,MAAM,IAAI,KAAK,WAAW,WAAW,GAAG,KAAK,aAAa,KAAK,UAAU;AACrG;AASO,SAAS,qBAAqB,MAA6B;AAChE,MAAI,CAAC,KAAM,QAAO;AAClB,MAAI,aAAa,KAAK,QAAQ,OAAO,GAAG,EAAE,QAAQ,QAAQ,EAAE;AAC5D,MAAI,gBAAgB,UAAU,EAAG,QAAO;AACxC,MAAI,uBAAuB,KAAK,UAAU,EAAG,cAAa,cAAc;AACxE,SAAO,8BAA8B,KAAK,UAAU,IAAI,aAAa;AACvE;AAGO,SAAS,oBAAoB,GAAW,GAAmB;AAChE,QAAM,KAAK,OAAO,EAAE,MAAM,uBAAuB,IAAI,CAAC,KAAK,OAAO,gBAAgB;AAClF,QAAM,KAAK,OAAO,EAAE,MAAM,uBAAuB,IAAI,CAAC,KAAK,OAAO,gBAAgB;AAClF,SAAO,OAAO,KAAK,EAAE,cAAc,CAAC,IAAI,KAAK;AAC/C;AAQO,SAAS,gBACd,QACA,sBAAsB,MAAM,OAAO,MACnC,aAAa,KACsC;AACnD,MAAI;AACF,UAAM,OAAO,IAAI,SAAS,MAAM;AAChC,UAAM,MAAM,OAAO;AAEnB,QAAI,aAAa;AACjB,aAAS,IAAI,MAAM,IAAI,KAAK,KAAK,IAAI,GAAG,MAAM,KAAK,GAAG,KAAK;AACzD,UAAI,KAAK,UAAU,GAAG,IAAI,MAAM,WAAY;AAAE,qBAAa;AAAG;AAAA,MAAM;AAAA,IACtE;AACA,QAAI,aAAa,EAAG,QAAO,EAAE,mBAAmB,GAAG,YAAY,EAAE;AAEjE,UAAM,aAAa,KAAK,UAAU,aAAa,IAAI,IAAI;AACvD,QAAI,aAAa,YAAY;AAC3B,YAAM,IAAI,YAAY,+CAAiB,UAAU,kBAAQ,UAAU,GAAG;AAAA,IACxE;AAEA,UAAM,SAAS,KAAK,UAAU,aAAa,IAAI,IAAI;AACnD,UAAM,WAAW,KAAK,UAAU,aAAa,IAAI,IAAI;AACrD,QAAI,WAAW,SAAS,IAAK,QAAO,EAAE,mBAAmB,GAAG,WAAW;AAEvE,QAAI,oBAAoB;AACxB,QAAI,MAAM;AACV,aAAS,IAAI,GAAG,IAAI,cAAc,MAAM,MAAM,WAAW,QAAQ,KAAK;AACpE,UAAI,KAAK,UAAU,KAAK,IAAI,MAAM,SAAY;AAC9C,2BAAqB,KAAK,UAAU,MAAM,IAAI,IAAI;AAClD,YAAM,UAAU,KAAK,UAAU,MAAM,IAAI,IAAI;AAC7C,YAAM,WAAW,KAAK,UAAU,MAAM,IAAI,IAAI;AAC9C,YAAM,aAAa,KAAK,UAAU,MAAM,IAAI,IAAI;AAChD,aAAO,KAAK,UAAU,WAAW;AAAA,IACnC;AAEA,QAAI,oBAAoB,qBAAqB;AAC3C,YAAM,IAAI,YAAY,sDAAmB,oBAAoB,OAAO,MAAM,QAAQ,CAAC,CAAC,oBAAU,sBAAsB,OAAO,IAAI,KAAK;AAAA,IACtI;AAEA,WAAO,EAAE,mBAAmB,WAAW;AAAA,EACzC,SAAS,KAAK;AACZ,QAAI,eAAe,YAAa,OAAM;AACtC,WAAO,EAAE,mBAAmB,GAAG,YAAY,EAAE;AAAA,EAC/C;AACF;AASO,SAAS,SAAS,KAAqB;AAC5C,SAAO,IAAI,QAAQ,WAAW,EAAE,EAAE,QAAQ,0CAA0C,EAAE;AACxF;AAGA,IAAM,eAAe;AACd,SAAS,aAAa,MAA6B;AACxD,QAAM,UAAU,KAAK,KAAK;AAC1B,MAAI,CAAC,WAAW,CAAC,aAAa,KAAK,OAAO,EAAG,QAAO;AAGpD,SAAO,QAAQ,QAAQ,OAAO,KAAK,EAAE,QAAQ,OAAO,KAAK;AAC3D;AAKO,SAAS,QAAQ,KAAuB;AAC7C,MAAI,MAAM;AACV,WAAS,IAAI,GAAG,IAAI,IAAI,QAAQ,IAAK,KAAI,IAAI,CAAC,IAAI,IAAK,OAAM,IAAI,CAAC;AAClE,SAAO;AACT;AAGO,SAAS,QAAQ,KAAuB;AAC7C,MAAI,MAAM;AACV,WAAS,IAAI,GAAG,IAAI,IAAI,QAAQ,IAAK,KAAI,IAAI,CAAC,IAAI,IAAK,OAAM,IAAI,CAAC;AAClE,SAAO;AACT;AAOO,SAAS,cAAc,KAAyB;AACrD,MAAI,EAAE,eAAe,OAAQ,QAAO;AACpC,QAAM,MAAM,IAAI;AAEhB,MAAI,IAAI,SAAS,KAAK,EAAG,QAAO;AAChC,MAAI,IAAI,SAAS,oBAAK,KAAK,IAAI,SAAS,iCAAQ,EAAG,QAAO;AAC1D,MAAI,IAAI,SAAS,qBAAqB,EAAG,QAAO;AAEhD,MAAI,IAAI,SAAS,uBAAuB,KAAK,IAAI,SAAS,+BAA+B,EAAG,QAAO;AACnG,MAAI,IAAI,SAAS,UAAU,KAAK,IAAI,SAAS,kDAAe,KAAK,IAAI,SAAS,4CAAc,EAAG,QAAO;AACtG,MAAI,IAAI,SAAS,MAAM,KAAK,IAAI,SAAS,2BAAO,KAAK,IAAI,SAAS,2BAAO,EAAG,QAAO;AACnF,MAAI,IAAI,SAAS,iCAAQ,EAAG,QAAO;AACnC,MAAI,IAAI,SAAS,cAAI,MAAM,IAAI,SAAS,4BAAQ,KAAK,IAAI,SAAS,cAAI,GAAI,QAAO;AACjF,MAAI,IAAI,SAAS,0BAAM,KAAK,IAAI,SAAS,kCAAS,EAAG,QAAO;AAC5D,SAAO;AACT;","names":[]}
@@ -0,0 +1,10 @@
1
+ #!/usr/bin/env node
2
+ import {
3
+ blocksToChunks
4
+ } from "./chunk-VNEV6VKD.js";
5
+ import "./chunk-IBVMJIHR.js";
6
+ import "./chunk-ZYQ5X6PG.js";
7
+ export {
8
+ blocksToChunks
9
+ };
10
+ //# sourceMappingURL=chunks-Y2VE2WCJ.js.map
package/dist/cli.js CHANGED
@@ -1,13 +1,13 @@
1
1
  #!/usr/bin/env node
2
2
  import {
3
3
  parseFormatProfileJson
4
- } from "./chunk-GCHH4WHT.js";
4
+ } from "./chunk-KQOTL4WI.js";
5
5
  import {
6
6
  BODY_FONTS,
7
7
  BULLET2_CHARS,
8
8
  H2_MARKERS,
9
9
  buildGongmunOptions
10
- } from "./chunk-5RPYBC2Q.js";
10
+ } from "./chunk-6GBLFQMA.js";
11
11
  import {
12
12
  BUILTIN_TEMPLATES,
13
13
  PRESET_ALIAS,
@@ -27,30 +27,31 @@ import {
27
27
  readBuiltinTemplate,
28
28
  resolveBuiltinTemplate,
29
29
  unknownFontWarnings
30
- } from "./chunk-45EUJ2P6.js";
31
- import "./chunk-7FCNOKOR.js";
32
- import "./chunk-UEWW3DH2.js";
33
- import "./chunk-MOL7MDBG.js";
34
- import "./chunk-CBUC73JM.js";
35
- import "./chunk-DZIXKL7E.js";
36
- import "./chunk-7ZQKECQT.js";
30
+ } from "./chunk-7BTO44Q7.js";
31
+ import "./chunk-W34QSJFF.js";
32
+ import "./chunk-PSAHY5XW.js";
33
+ import "./chunk-3TBUDJDE.js";
34
+ import "./chunk-OPL22SYF.js";
35
+ import "./chunk-O5EK52TG.js";
36
+ import "./chunk-VNEV6VKD.js";
37
37
  import {
38
38
  blocksToMarkdown
39
- } from "./chunk-ZHNUH2IW.js";
40
- import "./chunk-WDIPTY3Q.js";
41
- import "./chunk-TP4P7M4I.js";
39
+ } from "./chunk-IBVMJIHR.js";
40
+ import "./chunk-UQ4NSK66.js";
41
+ import "./chunk-KC63W72C.js";
42
42
  import {
43
43
  VERSION,
44
+ classifyError,
44
45
  sanitizeError,
45
46
  toArrayBuffer
46
- } from "./chunk-TTEHX4FN.js";
47
+ } from "./chunk-ZYQ5X6PG.js";
47
48
 
48
49
  // src/cli.ts
49
50
  import { readFileSync, writeFileSync, mkdirSync, statSync } from "fs";
50
51
  import { basename, dirname, resolve, extname } from "path";
51
52
  import { Command } from "commander";
52
53
  var program = new Command();
53
- program.name("kordoc").description("\uBAA8\uB450 \uD30C\uC2F1\uD574\uBC84\uB9AC\uACA0\uB2E4 \u2014 HWP, HWPX, PDF, XLSX, DOCX, \uC774\uBBF8\uC9C0(PNG/JPG/WebP) \u2192 Markdown").version(VERSION).argument("<files...>", "\uBCC0\uD658\uD560 \uD30C\uC77C \uACBD\uB85C (HWP, HWPX, PDF, XLSX, DOCX, PNG/JPG/WebP \u2014 \uC774\uBBF8\uC9C0\uB294 OCR \uC790\uB3D9 \uC801\uC6A9)").option("-o, --output <path>", "\uCD9C\uB825 \uD30C\uC77C \uACBD\uB85C (\uB2E8\uC77C \uD30C\uC77C \uC2DC)").option("-d, --out-dir <dir>", "\uCD9C\uB825 \uB514\uB809\uD1A0\uB9AC (\uB2E4\uC911 \uD30C\uC77C \uC2DC)").option("-p, --pages <range>", "\uD398\uC774\uC9C0/\uC139\uC158 \uBC94\uC704 (\uC608: 1-3, 1,3,5)").option("--format <type>", "\uCD9C\uB825 \uD615\uC2DD: markdown (\uAE30\uBCF8), json, chunks (RAG\uC6A9 \uAD6C\uC870 \uCCAD\uD06C JSON \u2014 \uD5E4\uB529\xB7\uAC1C\uC870\uC2DD \uC704\uACC4 breadcrumb + \uD45C \uB3C5\uB9BD \uCCAD\uD06C)", "markdown").option("--no-header-footer", "PDF \uBA38\uB9AC\uAE00/\uBC14\uB2E5\uAE00 \uC790\uB3D9 \uC81C\uAC70 \uB044\uAE30 (\uAE30\uBCF8: \uC81C\uAC70\uD568)").option("--formula-ocr", "PDF \uC218\uC2DD OCR \uD65C\uC131\uD654 (MFD+MFR ONNX, \uCCAB \uC0AC\uC6A9 \uC2DC \uBAA8\uB378 ~155MB \uC790\uB3D9 \uB2E4\uC6B4\uB85C\uB4DC)").option("--ocr", "\uC2A4\uCE94/\uC774\uBBF8\uC9C0 PDF \uD14D\uC2A4\uD2B8 OCR (\uB0B4\uC7A5 PP-OCRv5 korean, \uCCAB \uC0AC\uC6A9 \uC2DC \uBAA8\uB378 ~18MB \uC790\uB3D9 \uB2E4\uC6B4\uB85C\uB4DC \u2014 OCR \uD544\uC694 \uD398\uC774\uC9C0\uB9CC \uC778\uC2DD)").option("--ocr-force", "\uC804 \uD398\uC774\uC9C0 \uAC15\uC81C OCR (\uD14D\uC2A4\uD2B8\uCE35\uC774 \uC788\uC5B4\uB3C4 \uBB34\uC2DC\uD558\uACE0 \uC7AC\uC778\uC2DD)").option("--dedupe-headers", "HWP5 \uB808\uC774\uC544\uC6C3 \uD45C \uD398\uC774\uC9C0 \uBC18\uBCF5 \uB7EC\uB2DD \uD5E4\uB354 \uC911\uBCF5 \uC81C\uAC70 (\uAE30\uBCF8 off \u2014 \uBD99\uC784\uBCC4 \uC7AC\uBC88\uD638 \uC624\uC0AD\uC81C \uC8FC\uC758)").option("--keep-empty-cols", "\uD45C \uC624\uB978\uCABD \uB05D \uBE48 \uC5F4(\uC11C\uC2DD \uC785\uB825\uB780) \uBCF4\uC874 (#47, \uAE30\uBCF8 off: \uD6C4\uD589 \uBE48 \uC5F4 \uD2B8\uB9BC)").option("--keep-empty-paragraphs", "\uBE48 \uBB38\uB2E8 \uBCF4\uC874 \u2014 \uBCF8\uBB38\uC740 \uBE48 paragraph \uBE14\uB85D, \uD45C \uC140\uC740 \uBE48 \uC904\uB85C (#57, \uAE30\uBCF8 off: \uBE48 \uBB38\uB2E8 \uC81C\uAC70)").option("--inline-images", "\uC774\uBBF8\uC9C0\uB97C base64 data URI \uB85C \uB9C8\uD06C\uB2E4\uC6B4\uC5D0 \uC778\uB77C\uC778 (BMP\u2192PNG \uC555\uCD95, HWP5 \uC804\uC6A9 \u2014 \uC778\uB77C\uC778\uB41C \uACBD\uC6B0\uB9CC \uD30C\uC77C \uBBF8\uC800\uC7A5, \uADF8 \uC678 \uD3EC\uB9F7\uC740 \uC800\uC7A5 \uC720\uC9C0)").option("--password <pw>", "\uC554\uD638\uB85C \uBCF4\uD638\uB41C \uBB38\uC11C\uC758 \uC5F4\uAE30 \uC554\uD638 (#59, HWPX\xB7HWP3\xB7HWP5. \uD55C\uCEF4 DRM \uBB38\uC11C\uB294 \uD574\uB2F9 \uC5C6\uC74C)").option("--silent", "\uC9C4\uD589 \uBA54\uC2DC\uC9C0 \uC228\uAE30\uAE30").action(async (files, opts) => {
54
+ program.name("kordoc").description("\uBAA8\uB450 \uD30C\uC2F1\uD574\uBC84\uB9AC\uACA0\uB2E4 \u2014 HWP, HWPX, PDF, XLSX, DOCX, \uC774\uBBF8\uC9C0(PNG/JPG/WebP) \u2192 Markdown").version(VERSION).argument("<files...>", "\uBCC0\uD658\uD560 \uD30C\uC77C \uACBD\uB85C (HWP, HWPX, PDF, XLSX, DOCX, PNG/JPG/WebP \u2014 \uC774\uBBF8\uC9C0\uB294 OCR \uC790\uB3D9 \uC801\uC6A9)").option("-o, --output <path>", "\uCD9C\uB825 \uD30C\uC77C \uACBD\uB85C (\uB2E8\uC77C \uD30C\uC77C \uC2DC)").option("-d, --out-dir <dir>", "\uCD9C\uB825 \uB514\uB809\uD1A0\uB9AC (\uB2E4\uC911 \uD30C\uC77C \uC2DC)").option("-p, --pages <range>", "\uD398\uC774\uC9C0/\uC139\uC158 \uBC94\uC704 (\uC608: 1-3, 1,3,5)").option("--format <type>", "\uCD9C\uB825 \uD615\uC2DD: markdown (\uAE30\uBCF8), json, chunks (RAG\uC6A9 \uAD6C\uC870 \uCCAD\uD06C JSON \u2014 \uD5E4\uB529\xB7\uAC1C\uC870\uC2DD \uC704\uACC4 breadcrumb + \uD45C \uB3C5\uB9BD \uCCAD\uD06C)", "markdown").option("--no-header-footer", "PDF \uBA38\uB9AC\uAE00/\uBC14\uB2E5\uAE00 \uC790\uB3D9 \uC81C\uAC70 \uB044\uAE30 (\uAE30\uBCF8: \uC81C\uAC70\uD568)").option("--no-tables", "PDF \uD45C \uAC10\uC9C0 \uB044\uAE30 \u2014 \uD14C\uB450\uB9AC \uBC15\uC2A4\uB97C \uD45C\uB85C \uC624\uC778\uD574 \uC77D\uAE30 \uC21C\uC11C\uAC00 \uB4A4\uC9D1\uD788\uB294 \uBB38\uC11C(2\uB2E8 \uC2DC\uD5D8\uC9C0 \uB4F1)\uC5D0\uC11C \uC790\uC5F0 \uC77D\uAE30\uC21C \uD14D\uC2A4\uD2B8\uB9CC \uBF51\uB294\uB2E4 (#64)").option("--formula-ocr", "PDF \uC218\uC2DD OCR \uD65C\uC131\uD654 (MFD+MFR ONNX, \uCCAB \uC0AC\uC6A9 \uC2DC \uBAA8\uB378 ~155MB \uC790\uB3D9 \uB2E4\uC6B4\uB85C\uB4DC)").option("--ocr", "\uC2A4\uCE94/\uC774\uBBF8\uC9C0 PDF \uD14D\uC2A4\uD2B8 OCR (\uB0B4\uC7A5 PP-OCRv5 korean, \uCCAB \uC0AC\uC6A9 \uC2DC \uBAA8\uB378 ~18MB \uC790\uB3D9 \uB2E4\uC6B4\uB85C\uB4DC \u2014 OCR \uD544\uC694 \uD398\uC774\uC9C0\uB9CC \uC778\uC2DD)").option("--ocr-force", "\uC804 \uD398\uC774\uC9C0 \uAC15\uC81C OCR (\uD14D\uC2A4\uD2B8\uCE35\uC774 \uC788\uC5B4\uB3C4 \uBB34\uC2DC\uD558\uACE0 \uC7AC\uC778\uC2DD)").option("--dedupe-headers", "HWP5 \uB808\uC774\uC544\uC6C3 \uD45C \uD398\uC774\uC9C0 \uBC18\uBCF5 \uB7EC\uB2DD \uD5E4\uB354 \uC911\uBCF5 \uC81C\uAC70 (\uAE30\uBCF8 off \u2014 \uBD99\uC784\uBCC4 \uC7AC\uBC88\uD638 \uC624\uC0AD\uC81C \uC8FC\uC758)").option("--keep-empty-cols", "\uD45C \uC624\uB978\uCABD \uB05D \uBE48 \uC5F4(\uC11C\uC2DD \uC785\uB825\uB780) \uBCF4\uC874 (#47, \uAE30\uBCF8 off: \uD6C4\uD589 \uBE48 \uC5F4 \uD2B8\uB9BC)").option("--keep-empty-paragraphs", "\uBE48 \uBB38\uB2E8 \uBCF4\uC874 \u2014 \uBCF8\uBB38\uC740 \uBE48 paragraph \uBE14\uB85D, \uD45C \uC140\uC740 \uBE48 \uC904\uB85C (#57, \uAE30\uBCF8 off: \uBE48 \uBB38\uB2E8 \uC81C\uAC70)").option("--inline-images", "\uC774\uBBF8\uC9C0\uB97C base64 data URI \uB85C \uB9C8\uD06C\uB2E4\uC6B4\uC5D0 \uC778\uB77C\uC778 (BMP\u2192PNG \uC555\uCD95, HWP5 \uC804\uC6A9 \u2014 \uC778\uB77C\uC778\uB41C \uACBD\uC6B0\uB9CC \uD30C\uC77C \uBBF8\uC800\uC7A5, \uADF8 \uC678 \uD3EC\uB9F7\uC740 \uC800\uC7A5 \uC720\uC9C0)").option("--image-refs", "--format json \uC5D0\uC11C \uC774\uBBF8\uC9C0 \uBC14\uC774\uD2B8\uB97C \uC778\uB77C\uC778\uD558\uC9C0 \uC54A\uACE0 \uD30C\uC77C \uCC38\uC870(images/<\uD30C\uC77C\uBA85>)\uB9CC \uB0A8\uAE40 (#65 \u2014 \uC774\uBBF8\uC9C0\uAC00 \uC218\uBC31 \uC7A5\uC778 \uBB38\uC11C\uC758 \uC9C1\uB82C\uD654 \uD55C\uACC4 \uD68C\uD53C, -o/-d \uC640 \uD568\uAED8 \uC0AC\uC6A9)").option("--password <pw>", "\uC554\uD638\uB85C \uBCF4\uD638\uB41C \uBB38\uC11C\uC758 \uC5F4\uAE30 \uC554\uD638 (#59, HWPX\xB7HWP3\xB7HWP5. \uD55C\uCEF4 DRM \uBB38\uC11C\uB294 \uD574\uB2F9 \uC5C6\uC74C)").option("--silent", "\uC9C4\uD589 \uBA54\uC2DC\uC9C0 \uC228\uAE30\uAE30").action(async (files, opts) => {
54
55
  const validFormats = ["markdown", "json", "chunks"];
55
56
  if (!validFormats.includes(opts.format)) {
56
57
  process.stderr.write(`[kordoc] \uC9C0\uC6D0\uD558\uC9C0 \uC54A\uB294 \uD615\uC2DD: ${opts.format} (markdown, json, chunks)
@@ -66,6 +67,7 @@ program.name("kordoc").description("\uBAA8\uB450 \uD30C\uC2F1\uD574\uBC84\uB9AC\
66
67
  const absPath = resolve(filePath);
67
68
  const fileName = basename(absPath);
68
69
  const filePrefix = files.length > 1 ? `[${fi + 1}/${files.length}] ` : "";
70
+ let detectedFormat = "unknown";
69
71
  try {
70
72
  const fileSize = statSync(absPath).size;
71
73
  if (fileSize > 500 * 1024 * 1024) {
@@ -78,12 +80,14 @@ program.name("kordoc").description("\uBAA8\uB450 \uD30C\uC2F1\uD574\uBC84\uB9AC\
78
80
  const buffer = readFileSync(absPath);
79
81
  const arrayBuffer = toArrayBuffer(buffer);
80
82
  const format = detectFormat(arrayBuffer);
83
+ detectedFormat = format;
81
84
  if (!opts.silent) {
82
85
  process.stderr.write(`[kordoc] ${filePrefix}${fileName} (${format}) ...`);
83
86
  }
84
87
  const parseOptions = { filePath: absPath };
85
88
  if (opts.pages) parseOptions.pages = opts.pages;
86
89
  if (opts.headerFooter === false) parseOptions.removeHeaderFooter = false;
90
+ if (opts.tables === false) parseOptions.tables = false;
87
91
  if (opts.formulaOcr) parseOptions.formulaOcr = true;
88
92
  if (opts.ocrForce) parseOptions.ocr = "force";
89
93
  else if (opts.ocr) parseOptions.ocr = true;
@@ -98,6 +102,7 @@ program.name("kordoc").description("\uBAA8\uB450 \uD30C\uC2F1\uD574\uBC84\uB9AC\
98
102
  };
99
103
  }
100
104
  const result = await parse(arrayBuffer, parseOptions);
105
+ detectedFormat = result.fileType;
101
106
  if (!result.success) {
102
107
  process.stderr.write(` FAIL
103
108
  `);
@@ -116,15 +121,27 @@ program.name("kordoc").description("\uBAA8\uB450 \uD30C\uC2F1\uD574\uBC84\uB9AC\
116
121
  if (opts.outDir && result.images?.length && !imagesInlined) {
117
122
  markdown = markdown.replace(/!\[image\]\(image_/g, "![image](images/image_").replace(/(<img\b[^>]*\bsrc=")image_/g, "$1images/image_");
118
123
  }
119
- let output;
120
- if (opts.format === "json") {
121
- output = JSON.stringify(
122
- result,
124
+ const serializeJson = (refsOnly) => {
125
+ const payload = refsOnly && result.images?.length ? { ...result, images: result.images.map((img) => ({ filename: img.filename, mimeType: img.mimeType, path: `images/${img.filename}` })) } : result;
126
+ return JSON.stringify(
127
+ payload,
123
128
  (_key, value) => value instanceof Uint8Array ? Buffer.from(value).toString("base64") : value,
124
129
  2
125
130
  );
131
+ };
132
+ let output;
133
+ if (opts.format === "json") {
134
+ const savesImages = Boolean(opts.output && files.length === 1 || opts.outDir) && !imagesInlined;
135
+ try {
136
+ output = serializeJson(Boolean(opts.imageRefs) && savesImages);
137
+ } catch (err) {
138
+ if (!savesImages || !result.images?.length) throw err;
139
+ output = serializeJson(true);
140
+ process.stderr.write(` \u26A0\uFE0F \uC774\uBBF8\uC9C0 base64 \uC778\uB77C\uC778\uC774 \uC9C1\uB82C\uD654 \uD55C\uACC4\uB97C \uB118\uC5B4 \uD30C\uC77C \uCC38\uC870\uB85C \uB300\uCCB4\uD588\uC2B5\uB2C8\uB2E4 (${result.images.length}\uAC1C \u2192 images/)
141
+ `);
142
+ }
126
143
  } else if (opts.format === "chunks") {
127
- const { blocksToChunks } = await import("./chunks-EYDY4Y6O.js");
144
+ const { blocksToChunks } = await import("./chunks-Y2VE2WCJ.js");
128
145
  output = JSON.stringify(blocksToChunks(result.blocks), null, 2);
129
146
  } else {
130
147
  output = markdown;
@@ -159,6 +176,14 @@ program.name("kordoc").description("\uBAA8\uB450 \uD30C\uC2F1\uD574\uBC84\uB9AC\
159
176
  process.stderr.write(`
160
177
  [kordoc] ERROR: ${fileName} \u2014 ${sanitizeError(err)}
161
178
  `);
179
+ if (opts.format === "json") {
180
+ process.stdout.write(JSON.stringify({
181
+ success: false,
182
+ fileType: detectedFormat,
183
+ error: sanitizeError(err),
184
+ code: classifyError(err)
185
+ }, null, 2) + "\n");
186
+ }
162
187
  process.exitCode = 1;
163
188
  }
164
189
  }
@@ -171,7 +196,7 @@ program.command("watch <dir>").description("\uB514\uB809\uD1A0\uB9AC \uAC10\uC2D
171
196
  if (command.getOptionValueSource("format") === "default" && program.getOptionValueSource("format") === "cli") {
172
197
  opts.format = rootOpts.format;
173
198
  }
174
- const { watchDirectory } = await import("./watch-Y2VKD2MN.js");
199
+ const { watchDirectory } = await import("./watch-OIVXH2RN.js");
175
200
  await watchDirectory({
176
201
  dir,
177
202
  outDir: opts.outDir,
@@ -393,7 +418,7 @@ program.command("fill [file]").description("\uC11C\uC2DD \uBB38\uC11C\uC758 \uBE
393
418
  });
394
419
  program.command("seal <file>").description('\uB3C4\uC7A5/\uC11C\uBA85 \uC774\uBBF8\uC9C0\uB97C \uC575\uCEE4 \uBB38\uAD6C \uC704\uC5D0 \uBD80\uC720 \uBC30\uCE58 (\uD45C/\uD398\uC774\uC9C0 \uBD88\uD655\uC7A5) \u2014 kordoc seal \uC2E0\uCCAD\uC11C.hwpx --image \uB3C4\uC7A5.png --anchor "(\uC778)" -o \uACB0\uACFC.hwpx').requiredOption("--image <path>", "\uB3C4\uC7A5/\uC11C\uBA85 \uC774\uBBF8\uC9C0 (\uD22C\uBA85 \uBC30\uACBD PNG \uAD8C\uC7A5)").option("--anchor <text>", "\uC575\uCEE4 \uBB38\uAD6C", "(\uC778)").option("-n, --occurrence <num>", "\uAC19\uC740 \uC575\uCEE4\uAC00 \uC5EC\uB7FF\uC77C \uB54C 0-based \uC120\uD0DD", "0").option("--size-mm <num>", "\uB3C4\uC7A5 \uD55C \uBCC0 \uD06C\uAE30 mm (\uAE30\uBCF8: \uC904\uB192\uC774\xD71.6, 7~18 \uD074\uB7A8\uD504)").option("--mode <mode>", "overlap(\uBB38\uAD6C \uC704 \uACB9\uCE68) | right(\uC624\uB978\uCABD \uC606) | auto", "auto").option("--dx <mm>", "x \uBBF8\uC138\uC870\uC815 mm", "0").option("--dy <mm>", "y \uBBF8\uC138\uC870\uC815 mm", "0").option("-o, --output <path>", "\uCD9C\uB825 \uACBD\uB85C (\uAE30\uBCF8: <\uC785\uB825>.sealed.hwpx)").option("--silent", "\uC9C4\uD589 \uBA54\uC2DC\uC9C0 \uC228\uAE30\uAE30").action(async (file, opts) => {
395
420
  try {
396
- const { placeSealHwpx, detectFormat: detectFormat2 } = await import("./-CWYIC4ES.js");
421
+ const { placeSealHwpx, detectFormat: detectFormat2 } = await import("./-DEVLQKIT.js");
397
422
  const rootOpts = program.opts();
398
423
  const output = opts.output ?? rootOpts.output;
399
424
  const silent = opts.silent ?? rootOpts.silent;
@@ -471,7 +496,7 @@ program.command("seal <file>").description('\uB3C4\uC7A5/\uC11C\uBA85 \uC774\uBB
471
496
  });
472
497
  program.command("patch <original> <edited>").description("\uC11C\uC2DD \uBCF4\uC874 \uB77C\uC6B4\uB4DC\uD2B8\uB9BD \uD328\uCE58 \u2014 \uD3B8\uC9D1\uB41C \uB9C8\uD06C\uB2E4\uC6B4\uC744 \uC6D0\uBCF8 HWPX/HWP\uC5D0 in-place \uBC18\uC601 (kordoc patch \uC6D0\uBCF8.hwpx \uD3B8\uC9D1.md -o \uCD9C\uB825.hwpx). \uBBF8\uC801\uC6A9(skip) \uD3B8\uC9D1\uC774 \uC788\uC73C\uBA74 exit 2").option("-o, --output <path>", "\uCD9C\uB825 \uACBD\uB85C (\uAE30\uBCF8: <\uC6D0\uBCF8>.patched.hwpx|.hwp)").option("--no-verify", "\uD328\uCE58 \uD6C4 \uC7AC\uD30C\uC2F1 \uC790\uB3D9 \uAC80\uC99D \uC0DD\uB7B5").option("--silent", "\uC9C4\uD589 \uBA54\uC2DC\uC9C0 \uC228\uAE30\uAE30").action(async (original, edited, opts) => {
473
498
  try {
474
- const { patchHwpx, patchHwp, detectFormat: detectFormat2 } = await import("./-CWYIC4ES.js");
499
+ const { patchHwpx, patchHwp, detectFormat: detectFormat2 } = await import("./-DEVLQKIT.js");
475
500
  const rootOpts = program.opts();
476
501
  const output = opts.output ?? rootOpts.output;
477
502
  const silent = opts.silent ?? rootOpts.silent;
@@ -512,7 +537,7 @@ program.command("patch <original> <edited>").description("\uC11C\uC2DD \uBCF4\uC
512
537
  });
513
538
  program.command("validate <file>").description("HWPX \uAD6C\uC870 \uAC80\uC99D \u2014 ZIP\xB7mimetype\xB7\uD544\uC218 \uD30C\uC77C\xB7XML \uC6F0\uD3FC\uB4DC\xB7secCnt\xB7manifest \uCC38\uC870 (\uD55C\uCEF4\uB3C5\uC2A4 \uAC70\uBD80 \uC694\uC778 \uC0AC\uC804 \uCC28\uB2E8)").option("--json", "\uACB0\uACFC\uB97C JSON\uC73C\uB85C stdout\uC5D0 \uCD9C\uB825").action(async (file, opts) => {
514
539
  try {
515
- const { validateHwpx } = await import("./-CWYIC4ES.js");
540
+ const { validateHwpx } = await import("./-DEVLQKIT.js");
516
541
  const buf = new Uint8Array(readFileSync(resolve(file)));
517
542
  const result = await validateHwpx(buf);
518
543
  if (opts.json) {
@@ -726,7 +751,7 @@ program.command("lint <file>").description("\uACF5\uBB38\uC11C \uD45C\uAE30\uBC9
726
751
  }
727
752
  });
728
753
  program.command("redact <files...>").description("\uAC1C\uC778\uC815\uBCF4 \uC11C\uC2DD \uBCF4\uC874 \uB9C8\uC2A4\uD0B9 \u2014 \uC8FC\uBBFC\uBC88\uD638\xB7\uC804\uD654\xB7\uC774\uBA54\uC77C\xB7\uCE74\uB4DC\xB7\uACC4\uC88C\uB97C \uD0D0\uC9C0\uD574 HWPX/HWP\uB294 \uC6D0\uBCF8 \uC11C\uC2DD \uADF8\uB300\uB85C patch, \uADF8 \uC678 \uD3EC\uB9F7\uC740 \uB9C8\uC2A4\uD0B9\uB41C \uB9C8\uD06C\uB2E4\uC6B4 \uCD9C\uB825. \uC790\uB3D9 \uAC80\uCD9C \uBCF4\uC870 \uB3C4\uAD6C\uC774\uBBC0\uB85C \uACB0\uACFC\uB294 \uBC18\uB4DC\uC2DC \uC0AC\uB78C\uC774 \uCD5C\uC885 \uD655\uC778\uD558\uC138\uC694 (\uC774\uBBF8\uC9C0 \uC548 \uD14D\uC2A4\uD2B8\uB294 \uD0D0\uC9C0 \uBD88\uAC00)").option("--rules <csv>", "\uC801\uC6A9 \uB8F0 (\uAE30\uBCF8: rrn,phone,email,card,account \u2014 passport,driver\uB294 opt-in)").option("--mask-char <ch>", "\uB9C8\uC2A4\uD06C \uBB38\uC790 1\uAE00\uC790 (\uAE30\uBCF8: \u25CF)").option("-o, --output <path>", "\uCD9C\uB825 \uACBD\uB85C (\uB2E8\uC77C \uD30C\uC77C \uC2DC)").option("-d, --out-dir <dir>", "\uCD9C\uB825 \uB514\uB809\uD1A0\uB9AC (\uB2E4\uC911 \uD30C\uC77C \uC2DC)").option("--dry-run", "\uD0D0\uC9C0 \uB9AC\uD3EC\uD2B8\uB9CC \uCD9C\uB825, \uD30C\uC77C \uBBF8\uC0DD\uC131").option("--json", "\uB9AC\uD3EC\uD2B8\uB97C JSON\uC73C\uB85C stdout \uCD9C\uB825").option("--silent", "\uC9C4\uD589 \uBA54\uC2DC\uC9C0 \uC228\uAE30\uAE30").action(async (files, opts) => {
729
- const { redactMarkdown, DEFAULT_REDACT_RULES, patchHwpx, patchHwp } = await import("./-CWYIC4ES.js");
754
+ const { redactMarkdown, DEFAULT_REDACT_RULES, patchHwpx, patchHwp } = await import("./-DEVLQKIT.js");
730
755
  const rootOpts = program.opts();
731
756
  const output = opts.output ?? rootOpts.output;
732
757
  const silent = opts.silent ?? rootOpts.silent;
@@ -815,7 +840,7 @@ program.command("render <file>").description("\uB808\uC774\uC544\uC6C3 \uBCF4\uC
815
840
  const rootOpts = program.opts();
816
841
  const output = opts.output ?? rootOpts.output;
817
842
  const silent = opts.silent ?? rootOpts.silent;
818
- const { renderHwpxToSvg } = await import("./render-4WHQ7JCX.js");
843
+ const { renderHwpxToSvg } = await import("./render-YFIWQYVB.js");
819
844
  const absPath = resolve(file);
820
845
  const buffer = readFileSync(absPath);
821
846
  const highlights = opts.highlight ? String(opts.highlight).split(",") : void 0;
@@ -837,7 +862,7 @@ program.command("render <file>").description("\uB808\uC774\uC544\uC6C3 \uBCF4\uC
837
862
  });
838
863
  program.command("render-worker").description("persistent \uB80C\uB354 \uC6CC\uCEE4 \u2014 stdin NDJSON \uC694\uCCAD \u2192 \uC870\uD310 SVG \uD30C\uC77C \uCD9C\uB825 (\uD504\uB85C\uC138\uC2A4 \uC720\uC9C0, \uCF5C\uB4DC\uC2A4\uD0C0\uD2B8 \uC81C\uAC70)").action(async () => {
839
864
  const { createInterface } = await import("readline");
840
- const { renderHwpxToSvg } = await import("./render-4WHQ7JCX.js");
865
+ const { renderHwpxToSvg } = await import("./render-YFIWQYVB.js");
841
866
  const rl = createInterface({ input: process.stdin });
842
867
  const write = (o) => void process.stdout.write(JSON.stringify(o) + "\n");
843
868
  write({ ready: true, version: VERSION });
@@ -882,12 +907,12 @@ program.command("mcp").description("MCP \uC11C\uBC84 \uC2E4\uD589 (Claude / Curs
882
907
  await startMcpServer();
883
908
  });
884
909
  program.command("setup").description("\uB300\uD654\uD615 \uC124\uCE58 \uB9C8\uBC95\uC0AC \u2014 AI \uD074\uB77C\uC774\uC5B8\uD2B8 \uC790\uB3D9 \uB4F1\uB85D (Mac/Win/Linux)").action(async () => {
885
- const { runSetup } = await import("./setup-GUQT5ELJ.js");
910
+ const { runSetup } = await import("./setup-I4I75QJU.js");
886
911
  await runSetup();
887
912
  });
888
913
  program.command("check-formula-models").description("PDF \uC218\uC2DD OCR \uBAA8\uB378(MFD + MFR + tokenizer, ~155MB) \uC0C1\uD0DC \uD655\uC778 \u2014 \uC5C6\uAC70\uB098 SHA \uBD88\uC77C\uCE58\uBA74 \uB2E4\uC6B4\uB85C\uB4DC").option("--status-only", "\uC0C1\uD0DC\uB9CC JSON \uC73C\uB85C \uCD9C\uB825 (\uB2E4\uC6B4\uB85C\uB4DC \uC548 \uD568)").action(async (opts) => {
889
914
  try {
890
- const { getFormulaModelStatus, ensureFormulaModels, getFormulaModelsDir } = await import("./formula-JLYS73QL.js");
915
+ const { getFormulaModelStatus, ensureFormulaModels, getFormulaModelsDir } = await import("./formula-ETB6MESN.js");
891
916
  const dir = getFormulaModelsDir();
892
917
  if (opts.statusOnly) {
893
918
  const status = await getFormulaModelStatus();
@@ -941,7 +966,7 @@ program.command("check-formula-models").description("PDF \uC218\uC2DD OCR \uBAA8
941
966
  });
942
967
  program.command("check-ocr-models").description("\uD14D\uC2A4\uD2B8 OCR \uBAA8\uB378(PP-OCRv5 korean det+rec, ~18MB) \uC0C1\uD0DC \uD655\uC778 \u2014 \uC5C6\uAC70\uB098 SHA \uBD88\uC77C\uCE58\uBA74 \uB2E4\uC6B4\uB85C\uB4DC").option("--status-only", "\uC0C1\uD0DC\uB9CC JSON \uC73C\uB85C \uCD9C\uB825 (\uB2E4\uC6B4\uB85C\uB4DC \uC548 \uD568)").action(async (opts) => {
943
968
  try {
944
- const { getOcrModelStatus, ensureOcrModels, getOcrModelsDir } = await import("./models-6SIKLKRH.js");
969
+ const { getOcrModelStatus, ensureOcrModels, getOcrModelsDir } = await import("./models-6VMAVZR6.js");
945
970
  const dir = getOcrModelsDir();
946
971
  if (opts.statusOnly) {
947
972
  const status = await getOcrModelStatus();