kordoc 4.21.21 → 4.21.23
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.
- package/README.md +6 -4
- package/dist/{-ZYF6HX4M.js → -FLEBKE6V.js} +25 -25
- package/dist/{chunk-QHNVH4TC.js → chunk-2S37CQUJ.js} +21 -21
- package/dist/{chunk-YM3K3LCD.js → chunk-4F6OVO2A.js} +122 -46
- package/dist/chunk-4F6OVO2A.js.map +1 -0
- package/dist/{chunk-ZZIQWSGC.js → chunk-75J4Y5ZQ.js} +2 -2
- package/dist/{chunk-7YAJCXCD.js → chunk-7MTKOSAF.js} +2 -2
- package/dist/{chunk-KBV4WFZ6.js → chunk-C3RR6Z3M.js} +4 -4
- package/dist/{chunk-KBV4WFZ6.js.map → chunk-C3RR6Z3M.js.map} +1 -1
- package/dist/{chunk-TRO4CQHJ.js → chunk-CA742RMF.js} +4 -4
- package/dist/{chunk-CY4QM2KQ.js → chunk-DACTHSPR.js} +2 -2
- package/dist/{chunk-DTZIZ5PP.cjs → chunk-EGYLKSA4.cjs} +13 -13
- package/dist/{chunk-DTZIZ5PP.cjs.map → chunk-EGYLKSA4.cjs.map} +1 -1
- package/dist/{chunk-T7PN5D4D.cjs → chunk-EVPP2BSO.cjs} +6 -6
- package/dist/{chunk-T7PN5D4D.cjs.map → chunk-EVPP2BSO.cjs.map} +1 -1
- package/dist/{chunk-TMUUDMDH.cjs → chunk-GELBKBGQ.cjs} +23 -23
- package/dist/{chunk-TMUUDMDH.cjs.map → chunk-GELBKBGQ.cjs.map} +1 -1
- package/dist/{chunk-FBJVDBMO.js → chunk-GNIOEZDG.js} +2 -2
- package/dist/{chunk-RBRHO7WL.js → chunk-GPIMKPEI.js} +2 -2
- package/dist/{chunk-MCUX2Y2C.cjs → chunk-GUU7OHIH.cjs} +2 -2
- package/dist/{chunk-MCUX2Y2C.cjs.map → chunk-GUU7OHIH.cjs.map} +1 -1
- package/dist/{chunk-5HTFWUJE.js → chunk-J25WF6DP.js} +2 -2
- package/dist/{chunk-HWIW3WK2.js → chunk-JA4ZWCYL.js} +3 -3
- package/dist/{chunk-2ZGP7A5T.js → chunk-JLG7KK4M.js} +2 -2
- package/dist/{chunk-3TCF5HPP.js → chunk-JUA7IACQ.js} +2 -2
- package/dist/{chunk-YQZOPT3A.js → chunk-L4NFUEWJ.js} +5 -5
- package/dist/{chunk-YQZOPT3A.js.map → chunk-L4NFUEWJ.js.map} +1 -1
- package/dist/{chunk-T3DTTK23.js → chunk-QRIGJUOQ.js} +26 -25
- package/dist/chunk-QRIGJUOQ.js.map +1 -0
- package/dist/{chunk-J5PBI3ZC.js → chunk-RDV2RHVH.js} +3 -3
- package/dist/{chunk-YHPJTLCA.js → chunk-RKWJMA2B.js} +2 -2
- package/dist/{chunk-FZMGF6XU.js → chunk-TAU45SJK.js} +2 -2
- package/dist/{chunk-BG6LUCQL.js → chunk-U6D7LUCR.js} +2 -2
- package/dist/{chunk-BC54LBST.js → chunk-UKLNKY7Q.js} +3 -3
- package/dist/{chunk-H2MS4VG3.cjs → chunk-VA43X5N4.cjs} +3 -3
- package/dist/{chunk-H2MS4VG3.cjs.map → chunk-VA43X5N4.cjs.map} +1 -1
- package/dist/{chunk-7FCNOKOR.js → chunk-VKVDMH7X.js} +2 -1
- package/dist/{chunk-KTLHADU4.js → chunk-XEUBMLAS.js} +2 -2
- package/dist/{chunk-57VNFMMS.cjs → chunk-Y7XHHV55.cjs} +28 -27
- package/dist/chunk-Y7XHHV55.cjs.map +1 -0
- package/dist/{chunk-PTF4NCFM.js → chunk-YA6UGARL.js} +2 -2
- package/dist/{chunk-WGMYVRSA.js → chunk-ZE77KLKN.js} +2 -2
- package/dist/chunks-PR2GRNLX.js +10 -0
- package/dist/cli.js +34 -34
- package/dist/{formula-P7M7IH2L.js → formula-2MTBKZKO.js} +4 -4
- package/dist/{formula-A3I5VAP7.cjs → formula-CY2HF27M.cjs} +9 -9
- package/dist/{formula-A3I5VAP7.cjs.map → formula-CY2HF27M.cjs.map} +1 -1
- package/dist/{formula-6XNIXZ5L.js → formula-LGLSIYDI.js} +3 -3
- package/dist/{image-ocr-ITNY454I.js → image-ocr-BKRQVMW6.js} +9 -9
- package/dist/{image-ocr-UJRSQ2T3.cjs → image-ocr-HJM5IJKU.cjs} +15 -15
- package/dist/{image-ocr-UJRSQ2T3.cjs.map → image-ocr-HJM5IJKU.cjs.map} +1 -1
- package/dist/{image-ocr-3C6W3KZO.js → image-ocr-HP65O2CC.js} +6 -6
- package/dist/index.cjs +653 -577
- package/dist/index.cjs.map +1 -1
- package/dist/index.js +83 -7
- package/dist/index.js.map +1 -1
- package/dist/mcp.js +28 -28
- package/dist/{model-bundle-LAWQH75O.js → model-bundle-M7YMF4L4.js} +5 -5
- package/dist/{models-6QTD2CFM.js → models-AMPRHBYB.js} +5 -5
- package/dist/{parse-worker-v2-FEPJG3ER.js → parse-worker-v2-NMYEMVYX.js} +14 -14
- package/dist/{parser-VKWUMUML.cjs → parser-J6M4WG3I.cjs} +136 -136
- package/dist/{parser-VKWUMUML.cjs.map → parser-J6M4WG3I.cjs.map} +1 -1
- package/dist/{parser-56I364YM.js → parser-LX5MOE6Q.js} +8 -8
- package/dist/{parser-KFCY3NIN.js → parser-M45SFXWF.js} +22 -22
- package/dist/pdf-ocr-4YVV5PT4.cjs +19 -0
- package/dist/{pdf-ocr-OKAHNEJA.cjs.map → pdf-ocr-4YVV5PT4.cjs.map} +1 -1
- package/dist/pdf-ocr-D73YYQJ5.js +18 -0
- package/dist/pdf-ocr-SLWZO4D2.js +22 -0
- package/dist/{profile-io-OCITHUQF.js → profile-io-CUMSC3TD.js} +3 -3
- package/dist/{rasterize-SFCG4YOI.js → rasterize-W6NY4ADH.js} +3 -3
- package/dist/{redact-doc-5SMPXIGG.js → redact-doc-7HVCZT3C.js} +25 -25
- package/dist/{render-FGPG5A2N.js → render-FPMJX4CU.js} +9 -9
- package/dist/seal-YTRVR373.js +10 -0
- package/dist/{setup-JQVYZO7I.js → setup-UKIYOUT6.js} +3 -3
- package/dist/visual-DG5EVBZD.js +20 -0
- package/dist/{watch-C77BUSP7.js → watch-YHKMAVJ6.js} +16 -16
- package/package.json +1 -1
- package/dist/chunk-57VNFMMS.cjs.map +0 -1
- package/dist/chunk-T3DTTK23.js.map +0 -1
- package/dist/chunk-YM3K3LCD.js.map +0 -1
- package/dist/chunks-MDMJRCJ7.js +0 -10
- package/dist/pdf-ocr-MME37BL4.js +0 -18
- package/dist/pdf-ocr-OKAHNEJA.cjs +0 -19
- package/dist/pdf-ocr-ZZ6AJL6Z.js +0 -22
- package/dist/seal-GA7NZDTG.js +0 -10
- package/dist/visual-YMG644YW.js +0 -20
- /package/dist/{-ZYF6HX4M.js.map → -FLEBKE6V.js.map} +0 -0
- /package/dist/{chunk-QHNVH4TC.js.map → chunk-2S37CQUJ.js.map} +0 -0
- /package/dist/{chunk-ZZIQWSGC.js.map → chunk-75J4Y5ZQ.js.map} +0 -0
- /package/dist/{chunk-7YAJCXCD.js.map → chunk-7MTKOSAF.js.map} +0 -0
- /package/dist/{chunk-TRO4CQHJ.js.map → chunk-CA742RMF.js.map} +0 -0
- /package/dist/{chunk-CY4QM2KQ.js.map → chunk-DACTHSPR.js.map} +0 -0
- /package/dist/{chunk-FBJVDBMO.js.map → chunk-GNIOEZDG.js.map} +0 -0
- /package/dist/{chunk-RBRHO7WL.js.map → chunk-GPIMKPEI.js.map} +0 -0
- /package/dist/{chunk-5HTFWUJE.js.map → chunk-J25WF6DP.js.map} +0 -0
- /package/dist/{chunk-HWIW3WK2.js.map → chunk-JA4ZWCYL.js.map} +0 -0
- /package/dist/{chunk-2ZGP7A5T.js.map → chunk-JLG7KK4M.js.map} +0 -0
- /package/dist/{chunk-3TCF5HPP.js.map → chunk-JUA7IACQ.js.map} +0 -0
- /package/dist/{chunk-J5PBI3ZC.js.map → chunk-RDV2RHVH.js.map} +0 -0
- /package/dist/{chunk-YHPJTLCA.js.map → chunk-RKWJMA2B.js.map} +0 -0
- /package/dist/{chunk-FZMGF6XU.js.map → chunk-TAU45SJK.js.map} +0 -0
- /package/dist/{chunk-BG6LUCQL.js.map → chunk-U6D7LUCR.js.map} +0 -0
- /package/dist/{chunk-BC54LBST.js.map → chunk-UKLNKY7Q.js.map} +0 -0
- /package/dist/{chunk-7FCNOKOR.js.map → chunk-VKVDMH7X.js.map} +0 -0
- /package/dist/{chunk-KTLHADU4.js.map → chunk-XEUBMLAS.js.map} +0 -0
- /package/dist/{chunk-PTF4NCFM.js.map → chunk-YA6UGARL.js.map} +0 -0
- /package/dist/{chunk-WGMYVRSA.js.map → chunk-ZE77KLKN.js.map} +0 -0
- /package/dist/{chunks-MDMJRCJ7.js.map → chunks-PR2GRNLX.js.map} +0 -0
- /package/dist/{formula-P7M7IH2L.js.map → formula-2MTBKZKO.js.map} +0 -0
- /package/dist/{formula-6XNIXZ5L.js.map → formula-LGLSIYDI.js.map} +0 -0
- /package/dist/{image-ocr-ITNY454I.js.map → image-ocr-BKRQVMW6.js.map} +0 -0
- /package/dist/{image-ocr-3C6W3KZO.js.map → image-ocr-HP65O2CC.js.map} +0 -0
- /package/dist/{model-bundle-LAWQH75O.js.map → model-bundle-M7YMF4L4.js.map} +0 -0
- /package/dist/{models-6QTD2CFM.js.map → models-AMPRHBYB.js.map} +0 -0
- /package/dist/{parse-worker-v2-FEPJG3ER.js.map → parse-worker-v2-NMYEMVYX.js.map} +0 -0
- /package/dist/{parser-56I364YM.js.map → parser-LX5MOE6Q.js.map} +0 -0
- /package/dist/{parser-KFCY3NIN.js.map → parser-M45SFXWF.js.map} +0 -0
- /package/dist/{pdf-ocr-MME37BL4.js.map → pdf-ocr-D73YYQJ5.js.map} +0 -0
- /package/dist/{pdf-ocr-ZZ6AJL6Z.js.map → pdf-ocr-SLWZO4D2.js.map} +0 -0
- /package/dist/{profile-io-OCITHUQF.js.map → profile-io-CUMSC3TD.js.map} +0 -0
- /package/dist/{rasterize-SFCG4YOI.js.map → rasterize-W6NY4ADH.js.map} +0 -0
- /package/dist/{redact-doc-5SMPXIGG.js.map → redact-doc-7HVCZT3C.js.map} +0 -0
- /package/dist/{render-FGPG5A2N.js.map → render-FPMJX4CU.js.map} +0 -0
- /package/dist/{seal-GA7NZDTG.js.map → seal-YTRVR373.js.map} +0 -0
- /package/dist/{setup-JQVYZO7I.js.map → setup-UKIYOUT6.js.map} +0 -0
- /package/dist/{visual-YMG644YW.js.map → visual-DG5EVBZD.js.map} +0 -0
- /package/dist/{watch-C77BUSP7.js.map → watch-YHKMAVJ6.js.map} +0 -0
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
import {
|
|
2
2
|
blocksToPages,
|
|
3
3
|
encodePng
|
|
4
|
-
} from "./chunk-
|
|
4
|
+
} from "./chunk-QRIGJUOQ.js";
|
|
5
5
|
import {
|
|
6
6
|
hasRequestedPagesAfter,
|
|
7
7
|
parsePageRange
|
|
@@ -59,7 +59,7 @@ import {
|
|
|
59
59
|
stripAbsorbedRunningLines,
|
|
60
60
|
trimTrailingEmptyTableCols,
|
|
61
61
|
wrapJoiner
|
|
62
|
-
} from "./chunk-
|
|
62
|
+
} from "./chunk-L4NFUEWJ.js";
|
|
63
63
|
import {
|
|
64
64
|
CELL_BASELINES,
|
|
65
65
|
CELL_EDGES,
|
|
@@ -72,12 +72,12 @@ import {
|
|
|
72
72
|
escapeLiteralTags,
|
|
73
73
|
stripScriptTags,
|
|
74
74
|
unframeLayoutTables
|
|
75
|
-
} from "./chunk-
|
|
76
|
-
import "./chunk-
|
|
75
|
+
} from "./chunk-XEUBMLAS.js";
|
|
76
|
+
import "./chunk-GNIOEZDG.js";
|
|
77
77
|
import {
|
|
78
78
|
KordocError,
|
|
79
79
|
sanitizeHref
|
|
80
|
-
} from "./chunk-
|
|
80
|
+
} from "./chunk-75J4Y5ZQ.js";
|
|
81
81
|
|
|
82
82
|
// src/pdf/image-extract.ts
|
|
83
83
|
import { createHash } from "crypto";
|
|
@@ -3214,7 +3214,7 @@ function mergeLinkRuns(markdown) {
|
|
|
3214
3214
|
|
|
3215
3215
|
// src/pdf/formula-ocr.ts
|
|
3216
3216
|
async function applyFormulaOcr(buffer, blocks, pageFilter, effectivePageCount, warnings, _onProgress) {
|
|
3217
|
-
const formulaMod = await import("./formula-
|
|
3217
|
+
const formulaMod = await import("./formula-LGLSIYDI.js");
|
|
3218
3218
|
const { FormulaPipeline, ensureFormulaModels } = formulaMod;
|
|
3219
3219
|
await ensureFormulaModels((p) => {
|
|
3220
3220
|
if (p.phase === "download" && p.total) {
|
|
@@ -3637,7 +3637,7 @@ async function parsePdfDocument(buffer, options) {
|
|
|
3637
3637
|
}
|
|
3638
3638
|
if (targets.size > 0) {
|
|
3639
3639
|
try {
|
|
3640
|
-
const { runPdfOcr } = await import("./pdf-ocr-
|
|
3640
|
+
const { runPdfOcr } = await import("./pdf-ocr-D73YYQJ5.js");
|
|
3641
3641
|
const mode = typeof options?.ocr === "function" ? options.ocr : "builtin";
|
|
3642
3642
|
const regionPages = new Map([...ocrRegions].filter(([p]) => options?.ocr !== "force" && !isImageBased && !pageQuality.find((q) => q.page === p)?.needsOcr));
|
|
3643
3643
|
const ocrPageBlocks = await runPdfOcr(ocrBuffer, targets, mode, warnings, options?.onProgress, options?.tables !== false, vectorPageOps, regionPages);
|
|
@@ -3883,4 +3883,4 @@ export {
|
|
|
3883
3883
|
parsePdfDocument,
|
|
3884
3884
|
removeHeaderFooterBlocks
|
|
3885
3885
|
};
|
|
3886
|
-
//# sourceMappingURL=parser-
|
|
3886
|
+
//# sourceMappingURL=parser-LX5MOE6Q.js.map
|
|
@@ -1,4 +1,14 @@
|
|
|
1
1
|
#!/usr/bin/env node
|
|
2
|
+
import {
|
|
3
|
+
blocksToPages
|
|
4
|
+
} from "./chunk-JLG7KK4M.js";
|
|
5
|
+
import {
|
|
6
|
+
encodePng
|
|
7
|
+
} from "./chunk-VKVDMH7X.js";
|
|
8
|
+
import {
|
|
9
|
+
hasRequestedPagesAfter,
|
|
10
|
+
parsePageRange
|
|
11
|
+
} from "./chunk-VQFA55IC.js";
|
|
2
12
|
import {
|
|
3
13
|
CELL_LINES,
|
|
4
14
|
CLIP_TABLES,
|
|
@@ -51,18 +61,7 @@ import {
|
|
|
51
61
|
stripAbsorbedRunningLines,
|
|
52
62
|
trimTrailingEmptyTableCols,
|
|
53
63
|
wrapJoiner
|
|
54
|
-
} from "./chunk-
|
|
55
|
-
import {
|
|
56
|
-
ocrModelsCached
|
|
57
|
-
} from "./chunk-WGMYVRSA.js";
|
|
58
|
-
import "./chunk-3TCF5HPP.js";
|
|
59
|
-
import "./chunk-CY4QM2KQ.js";
|
|
60
|
-
import {
|
|
61
|
-
blocksToPages
|
|
62
|
-
} from "./chunk-2ZGP7A5T.js";
|
|
63
|
-
import {
|
|
64
|
-
encodePng
|
|
65
|
-
} from "./chunk-7FCNOKOR.js";
|
|
64
|
+
} from "./chunk-C3RR6Z3M.js";
|
|
66
65
|
import {
|
|
67
66
|
CELL_BASELINES,
|
|
68
67
|
CELL_EDGES,
|
|
@@ -71,21 +70,22 @@ import {
|
|
|
71
70
|
escapeLiteralTags,
|
|
72
71
|
stripScriptTags,
|
|
73
72
|
unframeLayoutTables
|
|
74
|
-
} from "./chunk-
|
|
75
|
-
import {
|
|
76
|
-
hasRequestedPagesAfter,
|
|
77
|
-
parsePageRange
|
|
78
|
-
} from "./chunk-VQFA55IC.js";
|
|
73
|
+
} from "./chunk-YA6UGARL.js";
|
|
79
74
|
import {
|
|
80
75
|
WINGDINGS,
|
|
81
76
|
assignTableIds,
|
|
82
77
|
blocksToMarkdown,
|
|
83
78
|
escapeLiteralDollar
|
|
84
|
-
} from "./chunk-
|
|
79
|
+
} from "./chunk-GPIMKPEI.js";
|
|
80
|
+
import {
|
|
81
|
+
ocrModelsCached
|
|
82
|
+
} from "./chunk-ZE77KLKN.js";
|
|
83
|
+
import "./chunk-JUA7IACQ.js";
|
|
84
|
+
import "./chunk-DACTHSPR.js";
|
|
85
85
|
import {
|
|
86
86
|
KordocError,
|
|
87
87
|
sanitizeHref
|
|
88
|
-
} from "./chunk-
|
|
88
|
+
} from "./chunk-TAU45SJK.js";
|
|
89
89
|
|
|
90
90
|
// src/pdf/image-extract.ts
|
|
91
91
|
import { createHash } from "crypto";
|
|
@@ -3222,7 +3222,7 @@ function mergeLinkRuns(markdown) {
|
|
|
3222
3222
|
|
|
3223
3223
|
// src/pdf/formula-ocr.ts
|
|
3224
3224
|
async function applyFormulaOcr(buffer, blocks, pageFilter, effectivePageCount, warnings, _onProgress) {
|
|
3225
|
-
const formulaMod = await import("./formula-
|
|
3225
|
+
const formulaMod = await import("./formula-2MTBKZKO.js");
|
|
3226
3226
|
const { FormulaPipeline, ensureFormulaModels } = formulaMod;
|
|
3227
3227
|
await ensureFormulaModels((p) => {
|
|
3228
3228
|
if (p.phase === "download" && p.total) {
|
|
@@ -3645,7 +3645,7 @@ async function parsePdfDocument(buffer, options) {
|
|
|
3645
3645
|
}
|
|
3646
3646
|
if (targets.size > 0) {
|
|
3647
3647
|
try {
|
|
3648
|
-
const { runPdfOcr } = await import("./pdf-ocr-
|
|
3648
|
+
const { runPdfOcr } = await import("./pdf-ocr-SLWZO4D2.js");
|
|
3649
3649
|
const mode = typeof options?.ocr === "function" ? options.ocr : "builtin";
|
|
3650
3650
|
const regionPages = new Map([...ocrRegions].filter(([p]) => options?.ocr !== "force" && !isImageBased && !pageQuality.find((q) => q.page === p)?.needsOcr));
|
|
3651
3651
|
const ocrPageBlocks = await runPdfOcr(ocrBuffer, targets, mode, warnings, options?.onProgress, options?.tables !== false, vectorPageOps, regionPages);
|
|
@@ -3891,4 +3891,4 @@ export {
|
|
|
3891
3891
|
parsePdfDocument,
|
|
3892
3892
|
removeHeaderFooterBlocks
|
|
3893
3893
|
};
|
|
3894
|
-
//# sourceMappingURL=parser-
|
|
3894
|
+
//# sourceMappingURL=parser-M45SFXWF.js.map
|
|
@@ -0,0 +1,19 @@
|
|
|
1
|
+
"use strict";Object.defineProperty(exports, "__esModule", {value: true});
|
|
2
|
+
|
|
3
|
+
|
|
4
|
+
|
|
5
|
+
|
|
6
|
+
var _chunkEGYLKSA4cjs = require('./chunk-EGYLKSA4.cjs');
|
|
7
|
+
require('./chunk-TK3FDAEE.cjs');
|
|
8
|
+
require('./chunk-GELBKBGQ.cjs');
|
|
9
|
+
require('./chunk-EVPP2BSO.cjs');
|
|
10
|
+
require('./chunk-VA43X5N4.cjs');
|
|
11
|
+
require('./chunk-GUU7OHIH.cjs');
|
|
12
|
+
require('./chunk-GS7T56RP.cjs');
|
|
13
|
+
|
|
14
|
+
|
|
15
|
+
|
|
16
|
+
|
|
17
|
+
|
|
18
|
+
exports.MAX_OCR_PIXELS = _chunkEGYLKSA4cjs.MAX_OCR_PIXELS; exports.groupOcrLines = _chunkEGYLKSA4cjs.groupOcrLines; exports.ocrItemsToBlocks = _chunkEGYLKSA4cjs.ocrItemsToBlocks; exports.runPdfOcr = _chunkEGYLKSA4cjs.runPdfOcr;
|
|
19
|
+
//# sourceMappingURL=pdf-ocr-4YVV5PT4.cjs.map
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"sources":["/Users/mong-e/workspace/kordoc/dist/pdf-ocr-
|
|
1
|
+
{"version":3,"sources":["/Users/mong-e/workspace/kordoc/dist/pdf-ocr-4YVV5PT4.cjs"],"names":[],"mappings":"AAAA;AACE;AACA;AACA;AACA;AACF,wDAA6B;AAC7B,gCAA6B;AAC7B,gCAA6B;AAC7B,gCAA6B;AAC7B,gCAA6B;AAC7B,gCAA6B;AAC7B,gCAA6B;AAC7B;AACE;AACA;AACA;AACA;AACF,mOAAC","file":"/Users/mong-e/workspace/kordoc/dist/pdf-ocr-4YVV5PT4.cjs"}
|
|
@@ -0,0 +1,18 @@
|
|
|
1
|
+
import {
|
|
2
|
+
MAX_OCR_PIXELS,
|
|
3
|
+
groupOcrLines,
|
|
4
|
+
ocrItemsToBlocks,
|
|
5
|
+
runPdfOcr
|
|
6
|
+
} from "./chunk-RDV2RHVH.js";
|
|
7
|
+
import "./chunk-DC6CTQTK.js";
|
|
8
|
+
import "./chunk-L4NFUEWJ.js";
|
|
9
|
+
import "./chunk-XEUBMLAS.js";
|
|
10
|
+
import "./chunk-GNIOEZDG.js";
|
|
11
|
+
import "./chunk-75J4Y5ZQ.js";
|
|
12
|
+
export {
|
|
13
|
+
MAX_OCR_PIXELS,
|
|
14
|
+
groupOcrLines,
|
|
15
|
+
ocrItemsToBlocks,
|
|
16
|
+
runPdfOcr
|
|
17
|
+
};
|
|
18
|
+
//# sourceMappingURL=pdf-ocr-D73YYQJ5.js.map
|
|
@@ -0,0 +1,22 @@
|
|
|
1
|
+
#!/usr/bin/env node
|
|
2
|
+
import {
|
|
3
|
+
MAX_OCR_PIXELS,
|
|
4
|
+
groupOcrLines,
|
|
5
|
+
ocrItemsToBlocks,
|
|
6
|
+
runPdfOcr
|
|
7
|
+
} from "./chunk-CA742RMF.js";
|
|
8
|
+
import "./chunk-6H25SXDI.js";
|
|
9
|
+
import "./chunk-C3RR6Z3M.js";
|
|
10
|
+
import "./chunk-YA6UGARL.js";
|
|
11
|
+
import "./chunk-GPIMKPEI.js";
|
|
12
|
+
import "./chunk-ZE77KLKN.js";
|
|
13
|
+
import "./chunk-JUA7IACQ.js";
|
|
14
|
+
import "./chunk-DACTHSPR.js";
|
|
15
|
+
import "./chunk-TAU45SJK.js";
|
|
16
|
+
export {
|
|
17
|
+
MAX_OCR_PIXELS,
|
|
18
|
+
groupOcrLines,
|
|
19
|
+
ocrItemsToBlocks,
|
|
20
|
+
runPdfOcr
|
|
21
|
+
};
|
|
22
|
+
//# sourceMappingURL=pdf-ocr-SLWZO4D2.js.map
|
|
@@ -2,10 +2,10 @@
|
|
|
2
2
|
import {
|
|
3
3
|
formatProfileSchema,
|
|
4
4
|
parseFormatProfileJson
|
|
5
|
-
} from "./chunk-
|
|
6
|
-
import "./chunk-
|
|
5
|
+
} from "./chunk-J25WF6DP.js";
|
|
6
|
+
import "./chunk-TAU45SJK.js";
|
|
7
7
|
export {
|
|
8
8
|
formatProfileSchema,
|
|
9
9
|
parseFormatProfileJson
|
|
10
10
|
};
|
|
11
|
-
//# sourceMappingURL=profile-io-
|
|
11
|
+
//# sourceMappingURL=profile-io-CUMSC3TD.js.map
|
|
@@ -3,11 +3,11 @@ import {
|
|
|
3
3
|
loadSharp,
|
|
4
4
|
rasterizePageSvg,
|
|
5
5
|
rasterizeSvg
|
|
6
|
-
} from "./chunk-
|
|
7
|
-
import "./chunk-
|
|
6
|
+
} from "./chunk-RKWJMA2B.js";
|
|
7
|
+
import "./chunk-TAU45SJK.js";
|
|
8
8
|
export {
|
|
9
9
|
loadSharp,
|
|
10
10
|
rasterizePageSvg,
|
|
11
11
|
rasterizeSvg
|
|
12
12
|
};
|
|
13
|
-
//# sourceMappingURL=rasterize-
|
|
13
|
+
//# sourceMappingURL=rasterize-W6NY4ADH.js.map
|
|
@@ -4,23 +4,8 @@ import {
|
|
|
4
4
|
parse,
|
|
5
5
|
replaceOleStream,
|
|
6
6
|
wipeFreeSectors
|
|
7
|
-
} from "./chunk-
|
|
8
|
-
import "./chunk-
|
|
9
|
-
import {
|
|
10
|
-
allLinesegRemovalSplices,
|
|
11
|
-
findElementEnd,
|
|
12
|
-
patchZipEntries,
|
|
13
|
-
scanSectionXml
|
|
14
|
-
} from "./chunk-HWIW3WK2.js";
|
|
15
|
-
import {
|
|
16
|
-
DEFAULT_REDACT_RULES,
|
|
17
|
-
mdMarkerRanges,
|
|
18
|
-
nameParticleAt,
|
|
19
|
-
normalizeForDetect,
|
|
20
|
-
redactMarkdown,
|
|
21
|
-
redactText
|
|
22
|
-
} from "./chunk-3WSNLH4B.js";
|
|
23
|
-
import "./chunk-7YAJCXCD.js";
|
|
7
|
+
} from "./chunk-4F6OVO2A.js";
|
|
8
|
+
import "./chunk-7MTKOSAF.js";
|
|
24
9
|
import {
|
|
25
10
|
FLAG_COMPRESSED,
|
|
26
11
|
FLAG_DISTRIBUTION,
|
|
@@ -36,14 +21,29 @@ import {
|
|
|
36
21
|
isExtendedOnlyCtrlChar,
|
|
37
22
|
parseFileHeader,
|
|
38
23
|
readRecords
|
|
39
|
-
} from "./chunk-
|
|
40
|
-
import "./chunk-
|
|
41
|
-
import "./chunk-
|
|
24
|
+
} from "./chunk-2S37CQUJ.js";
|
|
25
|
+
import "./chunk-RKWJMA2B.js";
|
|
26
|
+
import "./chunk-JLG7KK4M.js";
|
|
27
|
+
import "./chunk-VKVDMH7X.js";
|
|
42
28
|
import "./chunk-VQFA55IC.js";
|
|
43
|
-
import "./chunk-
|
|
44
|
-
import "./chunk-
|
|
45
|
-
import
|
|
46
|
-
|
|
29
|
+
import "./chunk-YA6UGARL.js";
|
|
30
|
+
import "./chunk-GPIMKPEI.js";
|
|
31
|
+
import {
|
|
32
|
+
allLinesegRemovalSplices,
|
|
33
|
+
findElementEnd,
|
|
34
|
+
patchZipEntries,
|
|
35
|
+
scanSectionXml
|
|
36
|
+
} from "./chunk-JA4ZWCYL.js";
|
|
37
|
+
import "./chunk-U6D7LUCR.js";
|
|
38
|
+
import "./chunk-TAU45SJK.js";
|
|
39
|
+
import {
|
|
40
|
+
DEFAULT_REDACT_RULES,
|
|
41
|
+
mdMarkerRanges,
|
|
42
|
+
nameParticleAt,
|
|
43
|
+
normalizeForDetect,
|
|
44
|
+
redactMarkdown,
|
|
45
|
+
redactText
|
|
46
|
+
} from "./chunk-3WSNLH4B.js";
|
|
47
47
|
|
|
48
48
|
// src/redact-scrub.ts
|
|
49
49
|
import { createRequire } from "module";
|
|
@@ -1282,4 +1282,4 @@ function markdownLeftovers(markdown, ctx) {
|
|
|
1282
1282
|
export {
|
|
1283
1283
|
redactDocument
|
|
1284
1284
|
};
|
|
1285
|
-
//# sourceMappingURL=redact-doc-
|
|
1285
|
+
//# sourceMappingURL=redact-doc-7HVCZT3C.js.map
|
|
@@ -15,17 +15,17 @@ import {
|
|
|
15
15
|
renderHwpxToSvg,
|
|
16
16
|
renderSceneToHtml,
|
|
17
17
|
unionBBox
|
|
18
|
-
} from "./chunk-
|
|
19
|
-
import "./chunk-7FCNOKOR.js";
|
|
20
|
-
import "./chunk-PTF4NCFM.js";
|
|
21
|
-
import "./chunk-VQFA55IC.js";
|
|
22
|
-
import "./chunk-BG6LUCQL.js";
|
|
23
|
-
import "./chunk-RBRHO7WL.js";
|
|
18
|
+
} from "./chunk-2S37CQUJ.js";
|
|
24
19
|
import {
|
|
25
20
|
rasterizePageSvg,
|
|
26
21
|
rasterizeSvg
|
|
27
|
-
} from "./chunk-
|
|
28
|
-
import "./chunk-
|
|
22
|
+
} from "./chunk-RKWJMA2B.js";
|
|
23
|
+
import "./chunk-VKVDMH7X.js";
|
|
24
|
+
import "./chunk-VQFA55IC.js";
|
|
25
|
+
import "./chunk-YA6UGARL.js";
|
|
26
|
+
import "./chunk-GPIMKPEI.js";
|
|
27
|
+
import "./chunk-U6D7LUCR.js";
|
|
28
|
+
import "./chunk-TAU45SJK.js";
|
|
29
29
|
export {
|
|
30
30
|
RegionCollector,
|
|
31
31
|
buildHwp5RenderStyles,
|
|
@@ -45,4 +45,4 @@ export {
|
|
|
45
45
|
renderSceneToHtml,
|
|
46
46
|
unionBBox
|
|
47
47
|
};
|
|
48
|
-
//# sourceMappingURL=render-
|
|
48
|
+
//# sourceMappingURL=render-FPMJX4CU.js.map
|
|
@@ -2,8 +2,8 @@
|
|
|
2
2
|
import {
|
|
3
3
|
getAccessRoot,
|
|
4
4
|
isOfflineMode
|
|
5
|
-
} from "./chunk-
|
|
6
|
-
import "./chunk-
|
|
5
|
+
} from "./chunk-DACTHSPR.js";
|
|
6
|
+
import "./chunk-TAU45SJK.js";
|
|
7
7
|
|
|
8
8
|
// src/setup.ts
|
|
9
9
|
import { createInterface } from "readline/promises";
|
|
@@ -243,4 +243,4 @@ function printManualConfig() {
|
|
|
243
243
|
export {
|
|
244
244
|
runSetup
|
|
245
245
|
};
|
|
246
|
-
//# sourceMappingURL=setup-
|
|
246
|
+
//# sourceMappingURL=setup-UKIYOUT6.js.map
|
|
@@ -0,0 +1,20 @@
|
|
|
1
|
+
#!/usr/bin/env node
|
|
2
|
+
import {
|
|
3
|
+
extractTables
|
|
4
|
+
} from "./chunk-4F6OVO2A.js";
|
|
5
|
+
import "./chunk-7MTKOSAF.js";
|
|
6
|
+
import "./chunk-2S37CQUJ.js";
|
|
7
|
+
import "./chunk-RKWJMA2B.js";
|
|
8
|
+
import "./chunk-JLG7KK4M.js";
|
|
9
|
+
import "./chunk-VKVDMH7X.js";
|
|
10
|
+
import "./chunk-VQFA55IC.js";
|
|
11
|
+
import "./chunk-YA6UGARL.js";
|
|
12
|
+
import "./chunk-GPIMKPEI.js";
|
|
13
|
+
import "./chunk-JA4ZWCYL.js";
|
|
14
|
+
import "./chunk-U6D7LUCR.js";
|
|
15
|
+
import "./chunk-TAU45SJK.js";
|
|
16
|
+
import "./chunk-3WSNLH4B.js";
|
|
17
|
+
export {
|
|
18
|
+
extractTables
|
|
19
|
+
};
|
|
20
|
+
//# sourceMappingURL=visual-DG5EVBZD.js.map
|
|
@@ -1,26 +1,26 @@
|
|
|
1
1
|
#!/usr/bin/env node
|
|
2
|
-
import {
|
|
3
|
-
assertNetworkAllowed
|
|
4
|
-
} from "./chunk-CY4QM2KQ.js";
|
|
5
2
|
import {
|
|
6
3
|
parse
|
|
7
|
-
} from "./chunk-
|
|
8
|
-
import "./chunk-
|
|
9
|
-
import "./chunk-HWIW3WK2.js";
|
|
10
|
-
import "./chunk-3WSNLH4B.js";
|
|
11
|
-
import "./chunk-7YAJCXCD.js";
|
|
4
|
+
} from "./chunk-4F6OVO2A.js";
|
|
5
|
+
import "./chunk-7MTKOSAF.js";
|
|
12
6
|
import {
|
|
13
7
|
detectFormat
|
|
14
|
-
} from "./chunk-
|
|
15
|
-
import "./chunk-
|
|
16
|
-
import "./chunk-
|
|
8
|
+
} from "./chunk-2S37CQUJ.js";
|
|
9
|
+
import "./chunk-RKWJMA2B.js";
|
|
10
|
+
import "./chunk-JLG7KK4M.js";
|
|
11
|
+
import "./chunk-VKVDMH7X.js";
|
|
17
12
|
import "./chunk-VQFA55IC.js";
|
|
18
|
-
import "./chunk-
|
|
19
|
-
import "./chunk-
|
|
20
|
-
import
|
|
13
|
+
import "./chunk-YA6UGARL.js";
|
|
14
|
+
import "./chunk-GPIMKPEI.js";
|
|
15
|
+
import {
|
|
16
|
+
assertNetworkAllowed
|
|
17
|
+
} from "./chunk-DACTHSPR.js";
|
|
18
|
+
import "./chunk-JA4ZWCYL.js";
|
|
19
|
+
import "./chunk-U6D7LUCR.js";
|
|
21
20
|
import {
|
|
22
21
|
toArrayBuffer
|
|
23
|
-
} from "./chunk-
|
|
22
|
+
} from "./chunk-TAU45SJK.js";
|
|
23
|
+
import "./chunk-3WSNLH4B.js";
|
|
24
24
|
|
|
25
25
|
// src/watch.ts
|
|
26
26
|
import { watch, readFileSync, writeFileSync, mkdirSync, statSync, existsSync, realpathSync } from "fs";
|
|
@@ -256,4 +256,4 @@ export {
|
|
|
256
256
|
postWebhookToAddress,
|
|
257
257
|
watchDirectory
|
|
258
258
|
};
|
|
259
|
-
//# sourceMappingURL=watch-
|
|
259
|
+
//# sourceMappingURL=watch-YHKMAVJ6.js.map
|
package/package.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "kordoc",
|
|
3
3
|
"mcpName": "io.github.chrisryugj/kordoc",
|
|
4
|
-
"version": "4.21.
|
|
4
|
+
"version": "4.21.23",
|
|
5
5
|
"description": "Parse Korean documents (HWP3-5, HWPX, HWPML, PDF, XLS(X), DOCX) to Markdown — CLI + MCP server with lossless patch/roundtrip, form-filler, document diff, print & layout-preserving SVG renderer",
|
|
6
6
|
"type": "module",
|
|
7
7
|
"exports": {
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
{"version":3,"sources":["/Users/mong-e/workspace/kordoc/dist/chunk-57VNFMMS.cjs","../src/page-markdown.ts","../src/image/transcode.ts"],"names":[],"mappings":"AAAA;AACE;AACF,wDAA6B;AAC7B;AACA;ACsBO,SAAS,aAAA,CAAc,MAAA,EAAmB,OAAA,EAAwC,kCAAA,EAA8C;AACrI,EAAA,MAAM,cAAA,kBAAgB,MAAA,mBAAO,IAAA,mBAAK,CAAA,CAAA,EAAA,GAAK,OAAO,CAAA,CAAE,WAAA,IAAe,QAAQ,CAAA,6BAAG,YAAA;AAC1E,EAAA,GAAA,CAAI,cAAA,IAAkB,KAAA,CAAA,EAAW,OAAO,KAAA,CAAA;AAExC,EAAA,MAAM,OAAA,kBAAS,IAAI,GAAA,CAAuB,CAAA;AAC1C,EAAA,IAAI,QAAA,EAAU,aAAA;AACd,EAAA,IAAA,CAAA,MAAW,MAAA,GAAS,MAAA,EAAQ;AAC1B,IAAA,GAAA,CAAI,OAAO,KAAA,CAAM,WAAA,IAAe,QAAA,EAAU,QAAA,EAAU,KAAA,CAAM,UAAA;AAC1D,IAAA,MAAM,OAAA,EAAS,MAAA,CAAO,GAAA,CAAI,OAAO,CAAA;AACjC,IAAA,GAAA,CAAI,MAAA,EAAQ,MAAA,CAAO,IAAA,CAAK,KAAK,CAAA;AAAA,IAAA,KACxB,MAAA,CAAO,GAAA,CAAI,OAAA,EAAS,CAAC,KAAK,CAAC,CAAA;AAAA,EAClC;AAEA,EAAA,MAAM,QAAA,EAAU,CAAC,GAAG,MAAA,CAAO,IAAA,CAAK,CAAC,CAAA;AACjC,EAAA,MAAM,IAAA,EAAM,IAAA,CAAK,GAAA,CAAI,GAAG,OAAO,CAAA;AAC/B,EAAA,MAAM,IAAA,EAAM,IAAA,CAAK,GAAA,CAAI,GAAG,OAAO,CAAA;AAE/B,EAAA,MAAM,MAAA,EAAwB,CAAC,CAAA;AAC/B,EAAA,IAAA,CAAA,IAAS,WAAA,EAAa,GAAA,EAAK,WAAA,GAAc,GAAA,EAAK,UAAA,EAAA,EAAc;AAC1D,IAAA,MAAM,WAAA,EAAa,MAAA,CAAO,GAAA,CAAI,UAAU,CAAA;AACxC,IAAA,KAAA,CAAM,IAAA,CAAK,EAAE,UAAA,EAAY,QAAA,EAAU,WAAA,EAAa,MAAA,CAAO,UAAU,EAAA,EAAI,GAAG,CAAC,CAAA;AAAA,EAC3E;AACA,EAAA,OAAO,KAAA;AACT;ADvBA;AACA;AEjBA,4BAA4B;AAI5B,IAAM,UAAA,EAAA,CAAa,CAAA,EAAA,GAAM;AACvB,EAAA,MAAM,MAAA,EAAQ,IAAI,WAAA,CAAY,GAAG,CAAA;AACjC,EAAA,IAAA,CAAA,IAAS,EAAA,EAAI,CAAA,EAAG,EAAA,EAAI,GAAA,EAAK,CAAA,EAAA,EAAK;AAC5B,IAAA,IAAI,EAAA,EAAI,CAAA;AACR,IAAA,IAAA,CAAA,IAAS,EAAA,EAAI,CAAA,EAAG,EAAA,EAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,MAAA,EAAA,EAAK,EAAA,EAAI,EAAA,EAAM,WAAA,EAAc,EAAA,IAAM,EAAA,EAAO,EAAA,IAAM,CAAA;AAAA,IAClD;AACA,IAAA,KAAA,CAAM,CAAC,EAAA,EAAI,EAAA,IAAM,CAAA;AAAA,EACnB;AACA,EAAA,OAAO,KAAA;AACT,CAAA,CAAA,CAAG,CAAA;AAGH,SAAS,KAAA,CAAM,KAAA,EAA2B;AACxC,EAAA,IAAI,EAAA,EAAI,UAAA;AACR,EAAA,IAAA,CAAA,IAAS,EAAA,EAAI,CAAA,EAAG,EAAA,EAAI,KAAA,CAAM,MAAA,EAAQ,CAAA,EAAA,EAAK;AACrC,IAAA,EAAA,EAAI,SAAA,CAAA,CAAW,EAAA,EAAI,KAAA,CAAM,CAAC,CAAA,EAAA,EAAK,GAAI,EAAA,EAAK,EAAA,IAAM,CAAA;AAAA,EAChD;AACA,EAAA,OAAA,CAAQ,EAAA,EAAI,UAAA,EAAA,IAAgB,CAAA;AAC9B;AAIA,IAAM,OAAA,EAAS,CAAA;AAEf,IAAM,QAAA,EAAU,KAAA;AAOhB,IAAM,WAAA,EAAa,IAAA;AAUZ,SAAS,QAAA,CAAS,GAAA,EAAoC;AAE3D,EAAA,GAAA,CAAI,GAAA,CAAI,OAAA,EAAS,EAAA,EAAI,OAAO,IAAA;AAC5B,EAAA,GAAA,CAAI,GAAA,CAAI,CAAC,EAAA,IAAM,GAAA,GAAQ,GAAA,CAAI,CAAC,EAAA,IAAM,EAAA,EAAM,OAAO,IAAA;AAE/C,EAAA,MAAM,GAAA,EAAK,IAAI,QAAA,CAAS,GAAA,CAAI,MAAA,EAAQ,GAAA,CAAI,UAAA,EAAY,GAAA,CAAI,UAAU,CAAA;AAClE,EAAA,MAAM,WAAA,EAAa,EAAA,CAAG,SAAA,CAAU,EAAA,EAAI,IAAI,CAAA;AACxC,EAAA,MAAM,WAAA,EAAa,EAAA,CAAG,SAAA,CAAU,EAAA,EAAI,IAAI,CAAA;AACxC,EAAA,GAAA,CAAI,WAAA,EAAa,EAAA,EAAI,OAAO,IAAA;AAE5B,EAAA,MAAM,MAAA,EAAQ,EAAA,CAAG,QAAA,CAAS,EAAA,EAAI,IAAI,CAAA;AAClC,EAAA,MAAM,UAAA,EAAY,EAAA,CAAG,QAAA,CAAS,EAAA,EAAI,IAAI,CAAA;AACtC,EAAA,MAAM,SAAA,EAAW,EAAA,CAAG,SAAA,CAAU,EAAA,EAAI,IAAI,CAAA;AACtC,EAAA,MAAM,YAAA,EAAc,EAAA,CAAG,SAAA,CAAU,EAAA,EAAI,IAAI,CAAA;AAEzC,EAAA,GAAA,CAAI,YAAA,IAAgB,MAAA,EAAQ,OAAO,IAAA;AACnC,EAAA,GAAA,CAAI,SAAA,IAAa,GAAA,GAAM,SAAA,IAAa,EAAA,EAAI,OAAO,IAAA;AAC/C,EAAA,GAAA,CAAI,MAAA,GAAS,EAAA,GAAK,UAAA,IAAc,CAAA,EAAG,OAAO,IAAA;AAC1C,EAAA,GAAA,CAAI,MAAA,EAAQ,QAAA,GAAW,IAAA,CAAK,GAAA,CAAI,SAAS,EAAA,EAAI,OAAA,EAAS,OAAO,IAAA;AAE7D,EAAA,MAAM,QAAA,EAAU,UAAA,EAAY,CAAA;AAC5B,EAAA,MAAM,OAAA,EAAS,IAAA,CAAK,GAAA,CAAI,SAAS,CAAA;AAEjC,EAAA,GAAA,CAAI,MAAA,EAAQ,OAAA,EAAS,UAAA,EAAY,OAAO,IAAA;AACxC,EAAA,MAAM,cAAA,EAAgB,SAAA,GAAY,CAAA;AAElC,EAAA,MAAM,UAAA,EAAa,MAAA,EAAQ,cAAA,EAAgB,EAAA,EAAK,CAAC,CAAA;AACjD,EAAA,GAAA,CAAI,WAAA,EAAa,UAAA,EAAY,OAAA,EAAS,GAAA,CAAI,MAAA,EAAQ,OAAO,IAAA;AAEzD,EAAA,MAAM,KAAA,EAAO,IAAI,UAAA,CAAW,MAAA,EAAQ,OAAA,EAAS,CAAC,CAAA;AAC9C,EAAA,IAAI,SAAA,EAAW,CAAA;AAEf,EAAA,IAAA,CAAA,IAAS,EAAA,EAAI,CAAA,EAAG,EAAA,EAAI,MAAA,EAAQ,CAAA,EAAA,EAAK;AAE/B,IAAA,MAAM,OAAA,EAAS,QAAA,EAAU,EAAA,EAAI,OAAA,EAAS,EAAA,EAAI,CAAA;AAC1C,IAAA,IAAI,IAAA,EAAM,WAAA,EAAa,OAAA,EAAS,SAAA;AAChC,IAAA,IAAI,IAAA,EAAM,EAAA,EAAI,MAAA,EAAQ,CAAA;AACtB,IAAA,IAAA,CAAA,IAAS,EAAA,EAAI,CAAA,EAAG,EAAA,EAAI,KAAA,EAAO,CAAA,EAAA,EAAK;AAE9B,MAAA,IAAA,CAAK,GAAG,EAAA,EAAI,GAAA,CAAI,IAAA,EAAM,CAAC,CAAA;AACvB,MAAA,IAAA,CAAK,IAAA,EAAM,CAAC,EAAA,EAAI,GAAA,CAAI,IAAA,EAAM,CAAC,CAAA;AAC3B,MAAA,IAAA,CAAK,IAAA,EAAM,CAAC,EAAA,EAAI,GAAA,CAAI,GAAG,CAAA;AACvB,MAAA,MAAM,EAAA,EAAI,SAAA,IAAa,GAAA,EAAK,GAAA,CAAI,IAAA,EAAM,CAAC,EAAA,EAAI,GAAA;AAC3C,MAAA,IAAA,CAAK,IAAA,EAAM,CAAC,EAAA,EAAI,CAAA;AAChB,MAAA,SAAA,GAAY,CAAA;AACZ,MAAA,IAAA,GAAO,aAAA;AACP,MAAA,IAAA,GAAO,CAAA;AAAA,IACT;AAAA,EACF;AAIA,EAAA,GAAA,CAAI,SAAA,IAAa,GAAA,GAAM,SAAA,IAAa,CAAA,EAAG;AACrC,IAAA,IAAA,CAAA,IAAS,EAAA,EAAI,CAAA,EAAG,EAAA,EAAI,IAAA,CAAK,MAAA,EAAQ,EAAA,GAAK,CAAA,EAAG,IAAA,CAAK,CAAC,EAAA,EAAI,GAAA;AAAA,EACrD;AAEA,EAAA,OAAO,SAAA,CAAU,KAAA,EAAO,MAAA,EAAQ,IAAI,CAAA;AACtC;AAIA,IAAM,cAAA,EAAgB,UAAA,CAAW,EAAA,CAAG,GAAA,EAAM,EAAA,EAAM,EAAA,EAAM,EAAA,EAAM,EAAA,EAAM,EAAA,EAAM,EAAA,EAAM,EAAI,CAAA;AAGlF,SAAS,KAAA,CAAM,IAAA,EAAc,IAAA,EAA8B;AACzD,EAAA,MAAM,KAAA,EAAO,IAAI,UAAA,CAAW,EAAA,EAAI,IAAA,CAAK,MAAM,CAAA;AAC3C,EAAA,IAAA,CAAK,CAAC,EAAA,EAAI,IAAA,CAAK,UAAA,CAAW,CAAC,CAAA;AAC3B,EAAA,IAAA,CAAK,CAAC,EAAA,EAAI,IAAA,CAAK,UAAA,CAAW,CAAC,CAAA;AAC3B,EAAA,IAAA,CAAK,CAAC,EAAA,EAAI,IAAA,CAAK,UAAA,CAAW,CAAC,CAAA;AAC3B,EAAA,IAAA,CAAK,CAAC,EAAA,EAAI,IAAA,CAAK,UAAA,CAAW,CAAC,CAAA;AAC3B,EAAA,IAAA,CAAK,GAAA,CAAI,IAAA,EAAM,CAAC,CAAA;AAEhB,EAAA,MAAM,IAAA,EAAM,IAAI,UAAA,CAAW,EAAA,EAAI,IAAA,CAAK,OAAA,EAAS,CAAC,CAAA;AAC9C,EAAA,MAAM,GAAA,EAAK,IAAI,QAAA,CAAS,GAAA,CAAI,MAAM,CAAA;AAClC,EAAA,EAAA,CAAG,SAAA,CAAU,CAAA,EAAG,IAAA,CAAK,MAAA,EAAQ,KAAK,CAAA;AAClC,EAAA,GAAA,CAAI,GAAA,CAAI,IAAA,EAAM,CAAC,CAAA;AACf,EAAA,EAAA,CAAG,SAAA,CAAU,EAAA,EAAI,IAAA,CAAK,MAAA,EAAQ,KAAA,CAAM,IAAI,CAAA,EAAG,KAAK,CAAA;AAChD,EAAA,OAAO,GAAA;AACT;AAGO,SAAS,SAAA,CAAU,KAAA,EAAe,MAAA,EAAgB,IAAA,EAA8B;AAErF,EAAA,MAAM,KAAA,EAAO,IAAI,UAAA,CAAW,EAAE,CAAA;AAC9B,EAAA,MAAM,GAAA,EAAK,IAAI,QAAA,CAAS,IAAA,CAAK,MAAM,CAAA;AACnC,EAAA,EAAA,CAAG,SAAA,CAAU,CAAA,EAAG,KAAA,EAAO,KAAK,CAAA;AAC5B,EAAA,EAAA,CAAG,SAAA,CAAU,CAAA,EAAG,MAAA,EAAQ,KAAK,CAAA;AAC7B,EAAA,IAAA,CAAK,CAAC,EAAA,EAAI,CAAA;AACV,EAAA,IAAA,CAAK,CAAC,EAAA,EAAI,CAAA;AAGV,EAAA,MAAM,OAAA,EAAS,MAAA,EAAQ,CAAA;AACvB,EAAA,MAAM,IAAA,EAAM,IAAI,UAAA,CAAA,CAAY,OAAA,EAAS,CAAA,EAAA,EAAK,MAAM,CAAA;AAChD,EAAA,IAAA,CAAA,IAAS,EAAA,EAAI,CAAA,EAAG,EAAA,EAAI,MAAA,EAAQ,CAAA,EAAA,EAAK;AAC/B,IAAA,MAAM,SAAA,EAAW,EAAA,EAAA,CAAK,OAAA,EAAS,CAAA,CAAA;AAC/B,IAAA,GAAA,CAAI,QAAQ,EAAA,EAAI,CAAA;AAChB,IAAA,GAAA,CAAI,GAAA,CAAI,IAAA,CAAK,QAAA,CAAS,EAAA,EAAI,MAAA,EAAQ,EAAA,EAAI,OAAA,EAAS,MAAM,CAAA,EAAG,SAAA,EAAW,CAAC,CAAA;AAAA,EACtE;AACA,EAAA,MAAM,KAAA,EAAO,+BAAA,GAAe,CAAA;AAE5B,EAAA,MAAM,UAAA,EAAY,KAAA,CAAM,MAAA,EAAQ,IAAI,CAAA;AACpC,EAAA,MAAM,UAAA,EAAY,KAAA,CAAM,MAAA,EAAQ,IAAI,CAAA;AACpC,EAAA,MAAM,UAAA,EAAY,KAAA,CAAM,MAAA,EAAQ,IAAI,UAAA,CAAW,CAAC,CAAC,CAAA;AAEjD,EAAA,MAAM,IAAA,EAAM,IAAI,UAAA,CAAW,aAAA,CAAc,OAAA,EAAS,SAAA,CAAU,OAAA,EAAS,SAAA,CAAU,OAAA,EAAS,SAAA,CAAU,MAAM,CAAA;AACxG,EAAA,IAAI,EAAA,EAAI,CAAA;AACR,EAAA,GAAA,CAAI,GAAA,CAAI,aAAA,EAAe,CAAC,CAAA;AAAG,EAAA,EAAA,GAAK,aAAA,CAAc,MAAA;AAC9C,EAAA,GAAA,CAAI,GAAA,CAAI,SAAA,EAAW,CAAC,CAAA;AAAG,EAAA,EAAA,GAAK,SAAA,CAAU,MAAA;AACtC,EAAA,GAAA,CAAI,GAAA,CAAI,SAAA,EAAW,CAAC,CAAA;AAAG,EAAA,EAAA,GAAK,SAAA,CAAU,MAAA;AACtC,EAAA,GAAA,CAAI,GAAA,CAAI,SAAA,EAAW,CAAC,CAAA;AACpB,EAAA,OAAO,GAAA;AACT;AAqBO,SAAS,wBAAA,CACd,QAAA,EACA,MAAA,EACA,IAAA,EACQ;AACR,EAAA,MAAM,SAAA,kBAAW,IAAA,6BAAM,WAAA,IAAa,KAAA;AAGpC,EAAA,MAAM,KAAA,kBAAO,IAAI,GAAA,CAAoB,CAAA;AACrC,EAAA,IAAA,CAAA,MAAW,IAAA,GAAO,MAAA,EAAQ;AACxB,IAAA,IAAI,MAAA,EAAQ,GAAA,CAAI,IAAA;AAChB,IAAA,IAAI,KAAA,EAAO,GAAA,CAAI,QAAA;AACf,IAAA,GAAA,CAAI,SAAA,GAAY,KAAA,IAAS,WAAA,EAAa;AACpC,MAAA,MAAM,IAAA,EAAM,QAAA,CAAS,GAAA,CAAI,IAAI,CAAA;AAC7B,MAAA,GAAA,CAAI,GAAA,EAAK;AACP,QAAA,MAAA,EAAQ,GAAA;AACR,QAAA,KAAA,EAAO,WAAA;AAAA,MACT;AAAA,IACF;AACA,IAAA,IAAA,CAAK,GAAA,CAAI,GAAA,CAAI,QAAA,EAAU,CAAA,KAAA,EAAQ,IAAI,CAAA,QAAA,EAAW,MAAA,CAAO,IAAA,CAAK,KAAK,CAAA,CAAE,QAAA,CAAS,QAAQ,CAAC,CAAA,CAAA;AACrF,EAAA;AAC4B,EAAA;AAE8D,EAAA;AAC/D,IAAA;AACuB,IAAA;AACjD,EAAA;AACgG,EAAA;AACtE,IAAA;AAC0B,IAAA;AACpD,EAAA;AACM,EAAA;AACT;AFzDwF;AACA;AACA;AACA;AACA;AACA","file":"/Users/mong-e/workspace/kordoc/dist/chunk-57VNFMMS.cjs","sourcesContent":[null,"/**\n * 페이지별 마크다운 사영 (#68)\n *\n * 문서를 다시 파싱하지 않고 이미 만들어진 `IRBlock.pageNumber` 로만 블록을 갈라\n * 페이지마다 `blocksToMarkdown()` 을 돌린다. 페이지 번호의 신뢰도는 파서가\n * 정하는 것이고(`metadata.pageMode` — \"layout\" 은 실제 페이지, \"section\" 은\n * 섹션 근사), 여기서는 그 값을 그대로 존중한다.\n */\n\nimport type { IRBlock, PageMarkdown } from \"./types.js\"\nimport { blocksToMarkdown } from \"./table/builder.js\"\n\n/**\n * 블록을 페이지 번호로 묶어 페이지별 마크다운을 만든다.\n *\n * - 어떤 블록에도 페이지 번호가 없으면 `undefined` — 페이지 개념이 없는 포맷에\n * \"전부 1페이지\" 같은 없는 사실을 만들어 내지 않는다.\n * - 번호가 없는 블록은 **직전 블록의 페이지로 이어붙인다.** 블록은 읽기 순서라\n * 이게 원문 위치에 가장 가깝고, 무엇보다 어느 페이지에도 안 실려 조용히\n * 사라지는 블록이 생기지 않는다. 선두의 번호 없는 블록은 첫 실번호 페이지로.\n * - 관측된 최소~최대 페이지 사이의 빈 페이지도 항목으로 낸다. 여러 페이지에\n * 걸친 표는 시작 페이지 한 블록이라 중간 페이지가 실제로 비어 있고, 이때\n * 항목을 빼면 소비자가 배열 길이로 페이지 수를 셀 수 없다.\n * - `render` 는 쪽 블록 → 마크다운. 문서 마크다운에 포맷 고유 마무리가 붙는 파서(PDF: 쪽번호 제거·균등배분 정리)는\n * 같은 마무리를 넘겨 쪽 마크다운이 문서 마크다운과 어긋나지 않게 한다.\n */\nexport function blocksToPages(blocks: IRBlock[], render: (blocks: IRBlock[]) => string = blocksToMarkdown): PageMarkdown[] | undefined {\n const firstNumbered = blocks.find(b => typeof b.pageNumber === \"number\")?.pageNumber\n if (firstNumbered === undefined) return undefined\n\n const byPage = new Map<number, IRBlock[]>()\n let current = firstNumbered\n for (const block of blocks) {\n if (typeof block.pageNumber === \"number\") current = block.pageNumber\n const bucket = byPage.get(current)\n if (bucket) bucket.push(block)\n else byPage.set(current, [block])\n }\n\n const numbers = [...byPage.keys()]\n const min = Math.min(...numbers)\n const max = Math.max(...numbers)\n\n const pages: PageMarkdown[] = []\n for (let pageNumber = min; pageNumber <= max; pageNumber++) {\n const pageBlocks = byPage.get(pageNumber)\n pages.push({ pageNumber, markdown: pageBlocks ? render(pageBlocks) : \"\" })\n }\n return pages\n}\n","/**\n * 순수 JS BMP → PNG 트랜스코더 + 마크다운 이미지 인라이너.\n *\n * HWP5 BinData 임베드 이미지는 비압축 BMP(수 MB)인 경우가 많다. AI 에이전트가\n * 자체 완결형 마크다운을 받도록, BMP 를 PNG(무손실 압축)로 변환한 뒤 base64\n * data URI 로 마크다운에 인라인한다.\n *\n * 의존성 최소화 원칙 — Node 내장(node:zlib, Buffer)만 사용한다.\n */\n\nimport { deflateSync } from \"node:zlib\"\n\n// ─── CRC32 (PNG 청크용) ──────────────────────────────\n\nconst CRC_TABLE = (() => {\n const table = new Uint32Array(256)\n for (let n = 0; n < 256; n++) {\n let c = n\n for (let k = 0; k < 8; k++) {\n c = (c & 1) ? (0xedb88320 ^ (c >>> 1)) : (c >>> 1)\n }\n table[n] = c >>> 0\n }\n return table\n})()\n\n/** 표준 PNG CRC32 (다항식 0xEDB88320) */\nfunction crc32(bytes: Uint8Array): number {\n let c = 0xffffffff\n for (let i = 0; i < bytes.length; i++) {\n c = CRC_TABLE[(c ^ bytes[i]) & 0xff] ^ (c >>> 8)\n }\n return (c ^ 0xffffffff) >>> 0\n}\n\n// ─── BMP 디코드 ──────────────────────────────────────\n\nconst BI_RGB = 0\n/** 폭·높이 상한 — 비정상 헤더에 의한 과대 할당 방지 */\nconst MAX_DIM = 0x7fff\n/**\n * 총 픽셀(W×H) 상한 — MAX_DIM 은 각 변만 제한하므로 곱의 폭주(과대 할당 + 긴 deflateSync\n * 동기 블로킹)를 별도로 차단한다. raw 저장 BMP 는 상위 inflate 100MB 캡을 우회하므로 필요.\n * 36MP = A4 전면 600dpi 스캔(≈35MP)까지 허용 — 그 이상은 rgba+raw 로 300MB 를 넘겨\n * MCP 서버 이벤트루프를 수 초 세우므로 원본 바이트 폴백으로 돌린다.\n */\nconst MAX_PIXELS = 36_000_000\n\n/**\n * 비압축 BMP(24/32bpp, BI_RGB) → PNG(8bit RGBA).\n *\n * 지원: BITMAPFILEHEADER + BITMAPINFOHEADER(biSize>=40), biBitCount 24/32,\n * biCompression BI_RGB, bottom-up(양수 biHeight)/top-down(음수), 24bpp 4바이트 행 패딩.\n * 미지원 변형(팔레트/8bpp·RLE·BITFIELDS·OS/2 헤더 등)은 null 을 반환하여\n * 호출부가 원본 바이트로 폴백하게 한다.\n */\nexport function bmpToPng(bmp: Uint8Array): Uint8Array | null {\n // BITMAPFILEHEADER(14) + BITMAPINFOHEADER(40)\n if (bmp.length < 54) return null\n if (bmp[0] !== 0x42 || bmp[1] !== 0x4d) return null // \"BM\"\n\n const dv = new DataView(bmp.buffer, bmp.byteOffset, bmp.byteLength)\n const dataOffset = dv.getUint32(10, true)\n const headerSize = dv.getUint32(14, true)\n if (headerSize < 40) return null // OS/2 BITMAPCOREHEADER 등 미지원\n\n const width = dv.getInt32(18, true)\n const rawHeight = dv.getInt32(22, true)\n const bitCount = dv.getUint16(28, true)\n const compression = dv.getUint32(30, true)\n\n if (compression !== BI_RGB) return null // RLE / BITFIELDS 미지원\n if (bitCount !== 24 && bitCount !== 32) return null // 팔레트 / 8bpp 등 미지원\n if (width <= 0 || rawHeight === 0) return null\n if (width > MAX_DIM || Math.abs(rawHeight) > MAX_DIM) return null\n\n const topDown = rawHeight < 0\n const height = Math.abs(rawHeight)\n // 총 픽셀 상한 검사 — rgba/raw 할당·deflate 전에 조기 차단 (각 변은 MAX_DIM 이하여도 곱은 폭주 가능)\n if (width * height > MAX_PIXELS) return null\n const bytesPerPixel = bitCount >> 3\n // 행 스트라이드 — 4바이트 경계 패딩 (24bpp 에서 유효)\n const rowStride = (width * bytesPerPixel + 3) & ~3\n if (dataOffset + rowStride * height > bmp.length) return null // 잘린 파일\n\n const rgba = new Uint8Array(width * height * 4)\n let anyAlpha = 0\n\n for (let y = 0; y < height; y++) {\n // bottom-up 은 첫 행이 이미지 맨 아래 → 뒤집어 읽는다\n const srcRow = topDown ? y : height - 1 - y\n let src = dataOffset + srcRow * rowStride\n let dst = y * width * 4\n for (let x = 0; x < width; x++) {\n // BMP 픽셀은 BGR(A) 순\n rgba[dst] = bmp[src + 2] // R\n rgba[dst + 1] = bmp[src + 1] // G\n rgba[dst + 2] = bmp[src] // B\n const a = bitCount === 32 ? bmp[src + 3] : 255\n rgba[dst + 3] = a\n anyAlpha |= a\n src += bytesPerPixel\n dst += 4\n }\n }\n\n // 32bpp BI_RGB 의 4번째 바이트는 \"미정의\" — 전부 0 이면 전면 투명으로 오해되므로\n // 불투명(255)으로 승격한다. 실제 알파가 있으면(anyAlpha!=0) 그대로 보존.\n if (bitCount === 32 && anyAlpha === 0) {\n for (let i = 3; i < rgba.length; i += 4) rgba[i] = 255\n }\n\n return encodePng(width, height, rgba)\n}\n\n// ─── PNG 인코드 (8bit RGBA, color type 6) ────────────\n\nconst PNG_SIGNATURE = Uint8Array.of(0x89, 0x50, 0x4e, 0x47, 0x0d, 0x0a, 0x1a, 0x0a)\n\n/** [4B 길이][4B 타입][데이터][4B CRC] 청크 조립. CRC 는 (타입+데이터) 기준. */\nfunction chunk(type: string, data: Uint8Array): Uint8Array {\n const body = new Uint8Array(4 + data.length)\n body[0] = type.charCodeAt(0)\n body[1] = type.charCodeAt(1)\n body[2] = type.charCodeAt(2)\n body[3] = type.charCodeAt(3)\n body.set(data, 4)\n\n const out = new Uint8Array(8 + data.length + 4)\n const dv = new DataView(out.buffer)\n dv.setUint32(0, data.length, false) // 길이(데이터만, big-endian)\n out.set(body, 4)\n dv.setUint32(8 + data.length, crc32(body), false)\n return out\n}\n\n/** 8bit RGBA 픽셀 버퍼 → PNG (PDF 이미지 XObject 추출에서도 사용) */\nexport function encodePng(width: number, height: number, rgba: Uint8Array): Uint8Array {\n // IHDR: width, height, bit depth 8, color type 6(RGBA), compression 0, filter 0, interlace 0\n const ihdr = new Uint8Array(13)\n const iv = new DataView(ihdr.buffer)\n iv.setUint32(0, width, false)\n iv.setUint32(4, height, false)\n ihdr[8] = 8\n ihdr[9] = 6\n\n // IDAT: 각 스캔라인 앞에 필터 바이트 0(None) 을 붙인 raw 데이터를 zlib deflate\n const stride = width * 4\n const raw = new Uint8Array((stride + 1) * height)\n for (let y = 0; y < height; y++) {\n const rowStart = y * (stride + 1)\n raw[rowStart] = 0 // filter: None\n raw.set(rgba.subarray(y * stride, y * stride + stride), rowStart + 1)\n }\n const idat = deflateSync(raw)\n\n const ihdrChunk = chunk(\"IHDR\", ihdr)\n const idatChunk = chunk(\"IDAT\", idat)\n const iendChunk = chunk(\"IEND\", new Uint8Array(0))\n\n const out = new Uint8Array(PNG_SIGNATURE.length + ihdrChunk.length + idatChunk.length + iendChunk.length)\n let o = 0\n out.set(PNG_SIGNATURE, o); o += PNG_SIGNATURE.length\n out.set(ihdrChunk, o); o += ihdrChunk.length\n out.set(idatChunk, o); o += idatChunk.length\n out.set(iendChunk, o)\n return out\n}\n\n// ─── 마크다운 이미지 인라이너 ────────────────────────\n\ninterface InlineImage {\n filename: string\n data: Uint8Array\n mimeType: string\n}\n\n/**\n * 마크다운의 이미지 참조를 base64 data URI 로 치환해 자체 완결형 마크다운을 만든다.\n * 두 방출 형태를 모두 커버한다:\n * - `` / `` — 일반 문단 (CLI 접두사 포함)\n * - `<img src=\"FILENAME\">` — 병합/중첩 표 셀 (table/builder.ts HTML 표 경로).\n * 이걸 빼면 CLI `--inline-images` 가 이미지 저장을 생략하면서 표 셀 이미지만\n * dangling 참조로 유실된다.\n *\n * @param opts.compress 기본 true. `image/bmp` 는 PNG 로 트랜스코딩 후 인라인하며,\n * 트랜스코딩 실패 시 원본 바이트+원본 MIME 으로 폴백한다.\n */\nexport function inlineImagesIntoMarkdown(\n markdown: string,\n images: ReadonlyArray<InlineImage>,\n opts?: { compress?: boolean },\n): string {\n const compress = opts?.compress !== false\n // 파일명 → data URI 사전 구축 후 단일 패스 치환 — 이미지 k개마다 전문 재스캔하던\n // O(k²) 제거. 맵에 없는 파일명 참조는 원문 그대로 둔다 (기존 계약 동일).\n const uris = new Map<string, string>()\n for (const img of images) {\n let bytes = img.data\n let mime = img.mimeType\n if (compress && mime === \"image/bmp\") {\n const png = bmpToPng(img.data)\n if (png) {\n bytes = png\n mime = \"image/png\"\n }\n }\n uris.set(img.filename, `data:${mime};base64,${Buffer.from(bytes).toString(\"base64\")}`)\n }\n if (uris.size === 0) return markdown\n // FILENAME 과 images/FILENAME 접두사를 모두 정확한 파일명 기준으로 치환\n let out = markdown.replace(/!\\[image\\]\\((?:images\\/)?([^()\\s]+)\\)/g, (m, name: string) => {\n const uri = uris.get(name)\n return uri !== undefined ? `` : m\n })\n out = out.replace(/(<img\\b[^>]*\\bsrc=\")(?:images\\/)?([^\"]*)(\")/g, (m, pre: string, name: string, post: string) => {\n const uri = uris.get(name)\n return uri !== undefined ? `${pre}${uri}${post}` : m\n })\n return out\n}\n"]}
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
{"version":3,"sources":["../src/page-markdown.ts","../src/image/transcode.ts"],"sourcesContent":["/**\n * 페이지별 마크다운 사영 (#68)\n *\n * 문서를 다시 파싱하지 않고 이미 만들어진 `IRBlock.pageNumber` 로만 블록을 갈라\n * 페이지마다 `blocksToMarkdown()` 을 돌린다. 페이지 번호의 신뢰도는 파서가\n * 정하는 것이고(`metadata.pageMode` — \"layout\" 은 실제 페이지, \"section\" 은\n * 섹션 근사), 여기서는 그 값을 그대로 존중한다.\n */\n\nimport type { IRBlock, PageMarkdown } from \"./types.js\"\nimport { blocksToMarkdown } from \"./table/builder.js\"\n\n/**\n * 블록을 페이지 번호로 묶어 페이지별 마크다운을 만든다.\n *\n * - 어떤 블록에도 페이지 번호가 없으면 `undefined` — 페이지 개념이 없는 포맷에\n * \"전부 1페이지\" 같은 없는 사실을 만들어 내지 않는다.\n * - 번호가 없는 블록은 **직전 블록의 페이지로 이어붙인다.** 블록은 읽기 순서라\n * 이게 원문 위치에 가장 가깝고, 무엇보다 어느 페이지에도 안 실려 조용히\n * 사라지는 블록이 생기지 않는다. 선두의 번호 없는 블록은 첫 실번호 페이지로.\n * - 관측된 최소~최대 페이지 사이의 빈 페이지도 항목으로 낸다. 여러 페이지에\n * 걸친 표는 시작 페이지 한 블록이라 중간 페이지가 실제로 비어 있고, 이때\n * 항목을 빼면 소비자가 배열 길이로 페이지 수를 셀 수 없다.\n * - `render` 는 쪽 블록 → 마크다운. 문서 마크다운에 포맷 고유 마무리가 붙는 파서(PDF: 쪽번호 제거·균등배분 정리)는\n * 같은 마무리를 넘겨 쪽 마크다운이 문서 마크다운과 어긋나지 않게 한다.\n */\nexport function blocksToPages(blocks: IRBlock[], render: (blocks: IRBlock[]) => string = blocksToMarkdown): PageMarkdown[] | undefined {\n const firstNumbered = blocks.find(b => typeof b.pageNumber === \"number\")?.pageNumber\n if (firstNumbered === undefined) return undefined\n\n const byPage = new Map<number, IRBlock[]>()\n let current = firstNumbered\n for (const block of blocks) {\n if (typeof block.pageNumber === \"number\") current = block.pageNumber\n const bucket = byPage.get(current)\n if (bucket) bucket.push(block)\n else byPage.set(current, [block])\n }\n\n const numbers = [...byPage.keys()]\n const min = Math.min(...numbers)\n const max = Math.max(...numbers)\n\n const pages: PageMarkdown[] = []\n for (let pageNumber = min; pageNumber <= max; pageNumber++) {\n const pageBlocks = byPage.get(pageNumber)\n pages.push({ pageNumber, markdown: pageBlocks ? render(pageBlocks) : \"\" })\n }\n return pages\n}\n","/**\n * 순수 JS BMP → PNG 트랜스코더 + 마크다운 이미지 인라이너.\n *\n * HWP5 BinData 임베드 이미지는 비압축 BMP(수 MB)인 경우가 많다. AI 에이전트가\n * 자체 완결형 마크다운을 받도록, BMP 를 PNG(무손실 압축)로 변환한 뒤 base64\n * data URI 로 마크다운에 인라인한다.\n *\n * 의존성 최소화 원칙 — Node 내장(node:zlib, Buffer)만 사용한다.\n */\n\nimport { deflateSync } from \"node:zlib\"\n\n// ─── CRC32 (PNG 청크용) ──────────────────────────────\n\nconst CRC_TABLE = (() => {\n const table = new Uint32Array(256)\n for (let n = 0; n < 256; n++) {\n let c = n\n for (let k = 0; k < 8; k++) {\n c = (c & 1) ? (0xedb88320 ^ (c >>> 1)) : (c >>> 1)\n }\n table[n] = c >>> 0\n }\n return table\n})()\n\n/** 표준 PNG CRC32 (다항식 0xEDB88320) */\nfunction crc32(bytes: Uint8Array): number {\n let c = 0xffffffff\n for (let i = 0; i < bytes.length; i++) {\n c = CRC_TABLE[(c ^ bytes[i]) & 0xff] ^ (c >>> 8)\n }\n return (c ^ 0xffffffff) >>> 0\n}\n\n// ─── BMP 디코드 ──────────────────────────────────────\n\nconst BI_RGB = 0\n/** 폭·높이 상한 — 비정상 헤더에 의한 과대 할당 방지 */\nconst MAX_DIM = 0x7fff\n/**\n * 총 픽셀(W×H) 상한 — MAX_DIM 은 각 변만 제한하므로 곱의 폭주(과대 할당 + 긴 deflateSync\n * 동기 블로킹)를 별도로 차단한다. raw 저장 BMP 는 상위 inflate 100MB 캡을 우회하므로 필요.\n * 36MP = A4 전면 600dpi 스캔(≈35MP)까지 허용 — 그 이상은 rgba+raw 로 300MB 를 넘겨\n * MCP 서버 이벤트루프를 수 초 세우므로 원본 바이트 폴백으로 돌린다.\n */\nconst MAX_PIXELS = 36_000_000\n\n/**\n * 비압축 BMP(24/32bpp, BI_RGB) → PNG(8bit RGBA).\n *\n * 지원: BITMAPFILEHEADER + BITMAPINFOHEADER(biSize>=40), biBitCount 24/32,\n * biCompression BI_RGB, bottom-up(양수 biHeight)/top-down(음수), 24bpp 4바이트 행 패딩.\n * 미지원 변형(팔레트/8bpp·RLE·BITFIELDS·OS/2 헤더 등)은 null 을 반환하여\n * 호출부가 원본 바이트로 폴백하게 한다.\n */\nexport function bmpToPng(bmp: Uint8Array): Uint8Array | null {\n // BITMAPFILEHEADER(14) + BITMAPINFOHEADER(40)\n if (bmp.length < 54) return null\n if (bmp[0] !== 0x42 || bmp[1] !== 0x4d) return null // \"BM\"\n\n const dv = new DataView(bmp.buffer, bmp.byteOffset, bmp.byteLength)\n const dataOffset = dv.getUint32(10, true)\n const headerSize = dv.getUint32(14, true)\n if (headerSize < 40) return null // OS/2 BITMAPCOREHEADER 등 미지원\n\n const width = dv.getInt32(18, true)\n const rawHeight = dv.getInt32(22, true)\n const bitCount = dv.getUint16(28, true)\n const compression = dv.getUint32(30, true)\n\n if (compression !== BI_RGB) return null // RLE / BITFIELDS 미지원\n if (bitCount !== 24 && bitCount !== 32) return null // 팔레트 / 8bpp 등 미지원\n if (width <= 0 || rawHeight === 0) return null\n if (width > MAX_DIM || Math.abs(rawHeight) > MAX_DIM) return null\n\n const topDown = rawHeight < 0\n const height = Math.abs(rawHeight)\n // 총 픽셀 상한 검사 — rgba/raw 할당·deflate 전에 조기 차단 (각 변은 MAX_DIM 이하여도 곱은 폭주 가능)\n if (width * height > MAX_PIXELS) return null\n const bytesPerPixel = bitCount >> 3\n // 행 스트라이드 — 4바이트 경계 패딩 (24bpp 에서 유효)\n const rowStride = (width * bytesPerPixel + 3) & ~3\n if (dataOffset + rowStride * height > bmp.length) return null // 잘린 파일\n\n const rgba = new Uint8Array(width * height * 4)\n let anyAlpha = 0\n\n for (let y = 0; y < height; y++) {\n // bottom-up 은 첫 행이 이미지 맨 아래 → 뒤집어 읽는다\n const srcRow = topDown ? y : height - 1 - y\n let src = dataOffset + srcRow * rowStride\n let dst = y * width * 4\n for (let x = 0; x < width; x++) {\n // BMP 픽셀은 BGR(A) 순\n rgba[dst] = bmp[src + 2] // R\n rgba[dst + 1] = bmp[src + 1] // G\n rgba[dst + 2] = bmp[src] // B\n const a = bitCount === 32 ? bmp[src + 3] : 255\n rgba[dst + 3] = a\n anyAlpha |= a\n src += bytesPerPixel\n dst += 4\n }\n }\n\n // 32bpp BI_RGB 의 4번째 바이트는 \"미정의\" — 전부 0 이면 전면 투명으로 오해되므로\n // 불투명(255)으로 승격한다. 실제 알파가 있으면(anyAlpha!=0) 그대로 보존.\n if (bitCount === 32 && anyAlpha === 0) {\n for (let i = 3; i < rgba.length; i += 4) rgba[i] = 255\n }\n\n return encodePng(width, height, rgba)\n}\n\n// ─── PNG 인코드 (8bit RGBA, color type 6) ────────────\n\nconst PNG_SIGNATURE = Uint8Array.of(0x89, 0x50, 0x4e, 0x47, 0x0d, 0x0a, 0x1a, 0x0a)\n\n/** [4B 길이][4B 타입][데이터][4B CRC] 청크 조립. CRC 는 (타입+데이터) 기준. */\nfunction chunk(type: string, data: Uint8Array): Uint8Array {\n const body = new Uint8Array(4 + data.length)\n body[0] = type.charCodeAt(0)\n body[1] = type.charCodeAt(1)\n body[2] = type.charCodeAt(2)\n body[3] = type.charCodeAt(3)\n body.set(data, 4)\n\n const out = new Uint8Array(8 + data.length + 4)\n const dv = new DataView(out.buffer)\n dv.setUint32(0, data.length, false) // 길이(데이터만, big-endian)\n out.set(body, 4)\n dv.setUint32(8 + data.length, crc32(body), false)\n return out\n}\n\n/** 8bit RGBA 픽셀 버퍼 → PNG (PDF 이미지 XObject 추출에서도 사용) */\nexport function encodePng(width: number, height: number, rgba: Uint8Array): Uint8Array {\n // IHDR: width, height, bit depth 8, color type 6(RGBA), compression 0, filter 0, interlace 0\n const ihdr = new Uint8Array(13)\n const iv = new DataView(ihdr.buffer)\n iv.setUint32(0, width, false)\n iv.setUint32(4, height, false)\n ihdr[8] = 8\n ihdr[9] = 6\n\n // IDAT: 각 스캔라인 앞에 필터 바이트 0(None) 을 붙인 raw 데이터를 zlib deflate\n const stride = width * 4\n const raw = new Uint8Array((stride + 1) * height)\n for (let y = 0; y < height; y++) {\n const rowStart = y * (stride + 1)\n raw[rowStart] = 0 // filter: None\n raw.set(rgba.subarray(y * stride, y * stride + stride), rowStart + 1)\n }\n const idat = deflateSync(raw)\n\n const ihdrChunk = chunk(\"IHDR\", ihdr)\n const idatChunk = chunk(\"IDAT\", idat)\n const iendChunk = chunk(\"IEND\", new Uint8Array(0))\n\n const out = new Uint8Array(PNG_SIGNATURE.length + ihdrChunk.length + idatChunk.length + iendChunk.length)\n let o = 0\n out.set(PNG_SIGNATURE, o); o += PNG_SIGNATURE.length\n out.set(ihdrChunk, o); o += ihdrChunk.length\n out.set(idatChunk, o); o += idatChunk.length\n out.set(iendChunk, o)\n return out\n}\n\n// ─── 마크다운 이미지 인라이너 ────────────────────────\n\ninterface InlineImage {\n filename: string\n data: Uint8Array\n mimeType: string\n}\n\n/**\n * 마크다운의 이미지 참조를 base64 data URI 로 치환해 자체 완결형 마크다운을 만든다.\n * 두 방출 형태를 모두 커버한다:\n * - `` / `` — 일반 문단 (CLI 접두사 포함)\n * - `<img src=\"FILENAME\">` — 병합/중첩 표 셀 (table/builder.ts HTML 표 경로).\n * 이걸 빼면 CLI `--inline-images` 가 이미지 저장을 생략하면서 표 셀 이미지만\n * dangling 참조로 유실된다.\n *\n * @param opts.compress 기본 true. `image/bmp` 는 PNG 로 트랜스코딩 후 인라인하며,\n * 트랜스코딩 실패 시 원본 바이트+원본 MIME 으로 폴백한다.\n */\nexport function inlineImagesIntoMarkdown(\n markdown: string,\n images: ReadonlyArray<InlineImage>,\n opts?: { compress?: boolean },\n): string {\n const compress = opts?.compress !== false\n // 파일명 → data URI 사전 구축 후 단일 패스 치환 — 이미지 k개마다 전문 재스캔하던\n // O(k²) 제거. 맵에 없는 파일명 참조는 원문 그대로 둔다 (기존 계약 동일).\n const uris = new Map<string, string>()\n for (const img of images) {\n let bytes = img.data\n let mime = img.mimeType\n if (compress && mime === \"image/bmp\") {\n const png = bmpToPng(img.data)\n if (png) {\n bytes = png\n mime = \"image/png\"\n }\n }\n uris.set(img.filename, `data:${mime};base64,${Buffer.from(bytes).toString(\"base64\")}`)\n }\n if (uris.size === 0) return markdown\n // FILENAME 과 images/FILENAME 접두사를 모두 정확한 파일명 기준으로 치환\n let out = markdown.replace(/!\\[image\\]\\((?:images\\/)?([^()\\s]+)\\)/g, (m, name: string) => {\n const uri = uris.get(name)\n return uri !== undefined ? `` : m\n })\n out = out.replace(/(<img\\b[^>]*\\bsrc=\")(?:images\\/)?([^\"]*)(\")/g, (m, pre: string, name: string, post: string) => {\n const uri = uris.get(name)\n return uri !== undefined ? `${pre}${uri}${post}` : m\n })\n return out\n}\n"],"mappings":";;;;;AA0BO,SAAS,cAAc,QAAmB,SAAwC,kBAA8C;AACrI,QAAM,gBAAgB,OAAO,KAAK,OAAK,OAAO,EAAE,eAAe,QAAQ,GAAG;AAC1E,MAAI,kBAAkB,OAAW,QAAO;AAExC,QAAM,SAAS,oBAAI,IAAuB;AAC1C,MAAI,UAAU;AACd,aAAW,SAAS,QAAQ;AAC1B,QAAI,OAAO,MAAM,eAAe,SAAU,WAAU,MAAM;AAC1D,UAAM,SAAS,OAAO,IAAI,OAAO;AACjC,QAAI,OAAQ,QAAO,KAAK,KAAK;AAAA,QACxB,QAAO,IAAI,SAAS,CAAC,KAAK,CAAC;AAAA,EAClC;AAEA,QAAM,UAAU,CAAC,GAAG,OAAO,KAAK,CAAC;AACjC,QAAM,MAAM,KAAK,IAAI,GAAG,OAAO;AAC/B,QAAM,MAAM,KAAK,IAAI,GAAG,OAAO;AAE/B,QAAM,QAAwB,CAAC;AAC/B,WAAS,aAAa,KAAK,cAAc,KAAK,cAAc;AAC1D,UAAM,aAAa,OAAO,IAAI,UAAU;AACxC,UAAM,KAAK,EAAE,YAAY,UAAU,aAAa,OAAO,UAAU,IAAI,GAAG,CAAC;AAAA,EAC3E;AACA,SAAO;AACT;;;ACvCA,SAAS,mBAAmB;AAI5B,IAAM,aAAa,MAAM;AACvB,QAAM,QAAQ,IAAI,YAAY,GAAG;AACjC,WAAS,IAAI,GAAG,IAAI,KAAK,KAAK;AAC5B,QAAI,IAAI;AACR,aAAS,IAAI,GAAG,IAAI,GAAG,KAAK;AAC1B,UAAK,IAAI,IAAM,aAAc,MAAM,IAAO,MAAM;AAAA,IAClD;AACA,UAAM,CAAC,IAAI,MAAM;AAAA,EACnB;AACA,SAAO;AACT,GAAG;AAGH,SAAS,MAAM,OAA2B;AACxC,MAAI,IAAI;AACR,WAAS,IAAI,GAAG,IAAI,MAAM,QAAQ,KAAK;AACrC,QAAI,WAAW,IAAI,MAAM,CAAC,KAAK,GAAI,IAAK,MAAM;AAAA,EAChD;AACA,UAAQ,IAAI,gBAAgB;AAC9B;AAIA,IAAM,SAAS;AAEf,IAAM,UAAU;AAOhB,IAAM,aAAa;AAUZ,SAAS,SAAS,KAAoC;AAE3D,MAAI,IAAI,SAAS,GAAI,QAAO;AAC5B,MAAI,IAAI,CAAC,MAAM,MAAQ,IAAI,CAAC,MAAM,GAAM,QAAO;AAE/C,QAAM,KAAK,IAAI,SAAS,IAAI,QAAQ,IAAI,YAAY,IAAI,UAAU;AAClE,QAAM,aAAa,GAAG,UAAU,IAAI,IAAI;AACxC,QAAM,aAAa,GAAG,UAAU,IAAI,IAAI;AACxC,MAAI,aAAa,GAAI,QAAO;AAE5B,QAAM,QAAQ,GAAG,SAAS,IAAI,IAAI;AAClC,QAAM,YAAY,GAAG,SAAS,IAAI,IAAI;AACtC,QAAM,WAAW,GAAG,UAAU,IAAI,IAAI;AACtC,QAAM,cAAc,GAAG,UAAU,IAAI,IAAI;AAEzC,MAAI,gBAAgB,OAAQ,QAAO;AACnC,MAAI,aAAa,MAAM,aAAa,GAAI,QAAO;AAC/C,MAAI,SAAS,KAAK,cAAc,EAAG,QAAO;AAC1C,MAAI,QAAQ,WAAW,KAAK,IAAI,SAAS,IAAI,QAAS,QAAO;AAE7D,QAAM,UAAU,YAAY;AAC5B,QAAM,SAAS,KAAK,IAAI,SAAS;AAEjC,MAAI,QAAQ,SAAS,WAAY,QAAO;AACxC,QAAM,gBAAgB,YAAY;AAElC,QAAM,YAAa,QAAQ,gBAAgB,IAAK,CAAC;AACjD,MAAI,aAAa,YAAY,SAAS,IAAI,OAAQ,QAAO;AAEzD,QAAM,OAAO,IAAI,WAAW,QAAQ,SAAS,CAAC;AAC9C,MAAI,WAAW;AAEf,WAAS,IAAI,GAAG,IAAI,QAAQ,KAAK;AAE/B,UAAM,SAAS,UAAU,IAAI,SAAS,IAAI;AAC1C,QAAI,MAAM,aAAa,SAAS;AAChC,QAAI,MAAM,IAAI,QAAQ;AACtB,aAAS,IAAI,GAAG,IAAI,OAAO,KAAK;AAE9B,WAAK,GAAG,IAAI,IAAI,MAAM,CAAC;AACvB,WAAK,MAAM,CAAC,IAAI,IAAI,MAAM,CAAC;AAC3B,WAAK,MAAM,CAAC,IAAI,IAAI,GAAG;AACvB,YAAM,IAAI,aAAa,KAAK,IAAI,MAAM,CAAC,IAAI;AAC3C,WAAK,MAAM,CAAC,IAAI;AAChB,kBAAY;AACZ,aAAO;AACP,aAAO;AAAA,IACT;AAAA,EACF;AAIA,MAAI,aAAa,MAAM,aAAa,GAAG;AACrC,aAAS,IAAI,GAAG,IAAI,KAAK,QAAQ,KAAK,EAAG,MAAK,CAAC,IAAI;AAAA,EACrD;AAEA,SAAO,UAAU,OAAO,QAAQ,IAAI;AACtC;AAIA,IAAM,gBAAgB,WAAW,GAAG,KAAM,IAAM,IAAM,IAAM,IAAM,IAAM,IAAM,EAAI;AAGlF,SAAS,MAAM,MAAc,MAA8B;AACzD,QAAM,OAAO,IAAI,WAAW,IAAI,KAAK,MAAM;AAC3C,OAAK,CAAC,IAAI,KAAK,WAAW,CAAC;AAC3B,OAAK,CAAC,IAAI,KAAK,WAAW,CAAC;AAC3B,OAAK,CAAC,IAAI,KAAK,WAAW,CAAC;AAC3B,OAAK,CAAC,IAAI,KAAK,WAAW,CAAC;AAC3B,OAAK,IAAI,MAAM,CAAC;AAEhB,QAAM,MAAM,IAAI,WAAW,IAAI,KAAK,SAAS,CAAC;AAC9C,QAAM,KAAK,IAAI,SAAS,IAAI,MAAM;AAClC,KAAG,UAAU,GAAG,KAAK,QAAQ,KAAK;AAClC,MAAI,IAAI,MAAM,CAAC;AACf,KAAG,UAAU,IAAI,KAAK,QAAQ,MAAM,IAAI,GAAG,KAAK;AAChD,SAAO;AACT;AAGO,SAAS,UAAU,OAAe,QAAgB,MAA8B;AAErF,QAAM,OAAO,IAAI,WAAW,EAAE;AAC9B,QAAM,KAAK,IAAI,SAAS,KAAK,MAAM;AACnC,KAAG,UAAU,GAAG,OAAO,KAAK;AAC5B,KAAG,UAAU,GAAG,QAAQ,KAAK;AAC7B,OAAK,CAAC,IAAI;AACV,OAAK,CAAC,IAAI;AAGV,QAAM,SAAS,QAAQ;AACvB,QAAM,MAAM,IAAI,YAAY,SAAS,KAAK,MAAM;AAChD,WAAS,IAAI,GAAG,IAAI,QAAQ,KAAK;AAC/B,UAAM,WAAW,KAAK,SAAS;AAC/B,QAAI,QAAQ,IAAI;AAChB,QAAI,IAAI,KAAK,SAAS,IAAI,QAAQ,IAAI,SAAS,MAAM,GAAG,WAAW,CAAC;AAAA,EACtE;AACA,QAAM,OAAO,YAAY,GAAG;AAE5B,QAAM,YAAY,MAAM,QAAQ,IAAI;AACpC,QAAM,YAAY,MAAM,QAAQ,IAAI;AACpC,QAAM,YAAY,MAAM,QAAQ,IAAI,WAAW,CAAC,CAAC;AAEjD,QAAM,MAAM,IAAI,WAAW,cAAc,SAAS,UAAU,SAAS,UAAU,SAAS,UAAU,MAAM;AACxG,MAAI,IAAI;AACR,MAAI,IAAI,eAAe,CAAC;AAAG,OAAK,cAAc;AAC9C,MAAI,IAAI,WAAW,CAAC;AAAG,OAAK,UAAU;AACtC,MAAI,IAAI,WAAW,CAAC;AAAG,OAAK,UAAU;AACtC,MAAI,IAAI,WAAW,CAAC;AACpB,SAAO;AACT;AAqBO,SAAS,yBACd,UACA,QACA,MACQ;AACR,QAAM,WAAW,MAAM,aAAa;AAGpC,QAAM,OAAO,oBAAI,IAAoB;AACrC,aAAW,OAAO,QAAQ;AACxB,QAAI,QAAQ,IAAI;AAChB,QAAI,OAAO,IAAI;AACf,QAAI,YAAY,SAAS,aAAa;AACpC,YAAM,MAAM,SAAS,IAAI,IAAI;AAC7B,UAAI,KAAK;AACP,gBAAQ;AACR,eAAO;AAAA,MACT;AAAA,IACF;AACA,SAAK,IAAI,IAAI,UAAU,QAAQ,IAAI,WAAW,OAAO,KAAK,KAAK,EAAE,SAAS,QAAQ,CAAC,EAAE;AAAA,EACvF;AACA,MAAI,KAAK,SAAS,EAAG,QAAO;AAE5B,MAAI,MAAM,SAAS,QAAQ,0CAA0C,CAAC,GAAG,SAAiB;AACxF,UAAM,MAAM,KAAK,IAAI,IAAI;AACzB,WAAO,QAAQ,SAAY,YAAY,GAAG,MAAM;AAAA,EAClD,CAAC;AACD,QAAM,IAAI,QAAQ,gDAAgD,CAAC,GAAG,KAAa,MAAc,SAAiB;AAChH,UAAM,MAAM,KAAK,IAAI,IAAI;AACzB,WAAO,QAAQ,SAAY,GAAG,GAAG,GAAG,GAAG,GAAG,IAAI,KAAK;AAAA,EACrD,CAAC;AACD,SAAO;AACT;","names":[]}
|