kordoc 4.18.3 → 4.18.4

package/dist/index.js

@@ -1,7 +1,7 @@
 import {
   blocksToPages,
   inlineImagesIntoMarkdown
-} from "./chunk-BGOR5GCF.js";
+} from "./chunk-274ZPMHK.js";
 import {
   parsePageRange
 } from "./chunk-V4BH5SRQ.js";
@@ -31,7 +31,7 @@ import {
   tidyScriptTags,
   unframeLayoutTables,
   wrapScript
-} from "./chunk-PSJLB3SZ.js";
+} from "./chunk-XG4SYOZL.js";
 import {
   KordocError,
   OPTIONAL_DEP_INSTALL_HINT,
@@ -49,7 +49,7 @@ import {
   toArrayBuffer,
   unescapeHtml,
   unzipLimitBytes
-} from "./chunk-CAN5P6N6.js";
+} from "./chunk-7GR2GCJF.js";
 
 // src/plain-markdown.ts
 function toPlainMarkdown(md2) {
@@ -3469,7 +3469,7 @@ function resolveImageBlocks(binDataMap, blocks, warnings, sweepUnreferenced) {
   resolveImageSentinels(blocks, renamed);
   return images;
 }
-function collectHwp5BinData(fileIndex) {
+function collectHwp5BinData(fileIndex, readStream) {
   const binDataMap = /* @__PURE__ */ new Map();
   if (fileIndex) {
     for (const entry of fileIndex) {
@@ -3477,35 +3477,37 @@ function collectHwp5BinData(fileIndex) {
       const match = entry.name.match(BIN_ENTRY_RE);
       if (!match) continue;
       const idx = parseInt(match[1], 16);
-      const data = normalizeBinPayload(Buffer.from(entry.content));
+      const raw = readStream ? readStream(`/BinData/${entry.name}`) : Buffer.from(entry.content);
+      if (!raw) continue;
+      const data = normalizeBinPayload(raw);
       binDataMap.set(idx, { data, name: entry.name });
     }
   }
   return binDataMap;
 }
-function collectHwp5BinDataLenient(lcfb) {
+function collectHwp5BinDataLenient(lcfb, readStream = lcfb.findStream) {
   const binDataMap = /* @__PURE__ */ new Map();
   const binRe = /^BIN([0-9A-Fa-f]{4,8})(?:\.|$)/;
   for (const e of lcfb.entries()) {
     const match = e.name.match(binRe);
     if (!match) continue;
     const idx = parseInt(match[1], 16);
-    const raw = lcfb.findStream(e.name);
+    const raw = readStream(e.name);
     if (!raw) continue;
     binDataMap.set(idx, { data: normalizeBinPayload(raw), name: e.name });
   }
   return binDataMap;
 }
-function extractHwp5Images(fileIndex, blocks, warnings, sweepUnreferenced) {
-  const binDataMap = collectHwp5BinData(fileIndex);
+function extractHwp5Images(fileIndex, blocks, warnings, sweepUnreferenced, readStream) {
+  const binDataMap = collectHwp5BinData(fileIndex, readStream);
   if (binDataMap.size === 0) {
     resolveImageSentinels(blocks, /* @__PURE__ */ new Map());
     return [];
   }
   return resolveImageBlocks(binDataMap, blocks, warnings, sweepUnreferenced);
 }
-function extractHwp5ImagesLenient(lcfb, blocks, warnings, sweepUnreferenced) {
-  const binDataMap = collectHwp5BinDataLenient(lcfb);
+function extractHwp5ImagesLenient(lcfb, blocks, warnings, sweepUnreferenced, readStream) {
+  const binDataMap = collectHwp5BinDataLenient(lcfb, readStream);
   if (binDataMap.size === 0) {
     resolveImageSentinels(blocks, /* @__PURE__ */ new Map());
     return [];
@@ -3762,8 +3764,11 @@ async function resolveSectionPaths(zip) {
 function parseSectionPathsFromManifest(xml) {
   const parser = createXmlParser();
   const doc = parser.parseFromString(stripDtd(xml), "text/xml");
-  const items = doc.getElementsByTagName("opf:item");
-  const spine = doc.getElementsByTagName("opf:itemref");
+  const opf = "http://www.idpf.org/2007/opf/";
+  const namespacedItems = doc.getElementsByTagNameNS(opf, "item");
+  const namespacedSpine = doc.getElementsByTagNameNS(opf, "itemref");
+  const items = namespacedItems.length ? namespacedItems : doc.getElementsByTagName("opf:item");
+  const spine = namespacedSpine.length ? namespacedSpine : doc.getElementsByTagName("opf:itemref");
   const idToHref = /* @__PURE__ */ new Map();
   for (let i = 0; i < items.length; i++) {
     const item = items[i];
@@ -5993,7 +5998,7 @@ function readHwp5DocInfoRecords(c) {
   }
 }
 function readHwp5SectionStreams(c) {
-  return c.distribution ? c.cfb ? findViewTextSections(c.cfb, c.compressed) : findViewTextSectionsLenient(c.lenientCfb, c.compressed) : c.encrypted ? findSectionsVia(c.findStream) : c.cfb ? findSections(c.cfb) : findSectionsLenient(c.lenientCfb, c.compressed);
+  return c.distribution ? c.cfb ? findViewTextSections(c.cfb, c.compressed) : findViewTextSectionsLenient(c.lenientCfb, c.compressed) : c.encrypted ? findSectionsVia(c.findStream) : c.cfb ? findSections(c.cfb) : findSectionsLenient(c.lenientCfb);
 }
 function readHwp5SectionRecords(c, sections, warnings) {
   const out = [];
@@ -6013,7 +6018,8 @@ function readHwp5SectionRecords(c, sections, warnings) {
   return out;
 }
 function readHwp5BinData(c) {
-  return c.cfb ? collectHwp5BinData(c.cfb.FileIndex) : collectHwp5BinDataLenient(c.lenientCfb);
+  const readStream = c.encrypted ? c.findStream : void 0;
+  return c.cfb ? collectHwp5BinData(c.cfb.FileIndex, readStream) : collectHwp5BinDataLenient(c.lenientCfb, readStream);
 }
 function parseHwp5Document(buffer, options) {
   const c = openHwp5Container(buffer, options);
@@ -6078,7 +6084,7 @@ function parseHwp5Document(buffer, options) {
     bodyBlocks = bodyBlocks.filter((b) => b.pageNumber != null && pageFilter.has(b.pageNumber));
   }
   const blocks = [...doc.headerBlocks, ...bodyBlocks, ...doc.footerBlocks];
-  const images = cfb ? extractHwp5Images(cfb.FileIndex, blocks, warnings, !pageFilter) : extractHwp5ImagesLenient(lenientCfb, blocks, warnings, !pageFilter);
+  const images = cfb ? extractHwp5Images(cfb.FileIndex, blocks, warnings, !pageFilter, encrypted ? findStream : void 0) : extractHwp5ImagesLenient(lenientCfb, blocks, warnings, !pageFilter, encrypted ? findStream : void 0);
   let flatBlocks = flattenLayoutTables(blocks);
   if (options?.layoutTables !== "keep") flatBlocks = unframeLayoutTables(flatBlocks, !!options?.keepTrailingEmptyCols);
   if (options?.dedupeRunningHeaders) {
@@ -6285,16 +6291,15 @@ function findSections(cfb) {
   }
   return sections.sort((a, b) => a.idx - b.idx).map((s) => s.content);
 }
-function findSectionsLenient(lcfb, compressed) {
+function findSectionsLenient(lcfb) {
   const sections = [];
-  let totalDecompressed = 0;
+  let totalBytes = 0;
   for (let i = 0; i < MAX_SECTIONS; i++) {
     const raw = lcfb.findStream(`/BodyText/Section${i}`) ?? lcfb.findStream(`Section${i}`);
     if (!raw) break;
-    const content = compressed ? decompressStream(raw) : raw;
-    totalDecompressed += content.length;
-    if (totalDecompressed > MAX_TOTAL_DECOMPRESS) throw new KordocError("\uCD1D \uC555\uCD95 \uD574\uC81C \uD06C\uAE30 \uCD08\uACFC (decompression bomb \uC758\uC2EC)");
-    sections.push({ idx: i, content });
+    totalBytes += raw.length;
+    if (totalBytes > MAX_TOTAL_DECOMPRESS) throw new KordocError("\uCD1D \uC555\uCD95 \uD574\uC81C \uD06C\uAE30 \uCD08\uACFC (decompression bomb \uC758\uC2EC)");
+    sections.push({ idx: i, content: raw });
   }
   if (sections.length === 0) {
     for (const e of lcfb.entries()) {
@@ -6303,10 +6308,9 @@ function findSectionsLenient(lcfb, compressed) {
         const idx = parseInt(e.name.replace("Section", ""), 10) || 0;
         const raw = lcfb.findStream(e.name);
         if (raw) {
-          const content = compressed ? decompressStream(raw) : raw;
-          totalDecompressed += content.length;
-          if (totalDecompressed > MAX_TOTAL_DECOMPRESS) throw new KordocError("\uCD1D \uC555\uCD95 \uD574\uC81C \uD06C\uAE30 \uCD08\uACFC (decompression bomb \uC758\uC2EC)");
-          sections.push({ idx, content });
+          totalBytes += raw.length;
+          if (totalBytes > MAX_TOTAL_DECOMPRESS) throw new KordocError("\uCD1D \uC555\uCD95 \uD574\uC81C \uD06C\uAE30 \uCD08\uACFC (decompression bomb \uC758\uC2EC)");
+          sections.push({ idx, content: raw });
         }
       }
     }
@@ -29308,6 +29312,7 @@ var ImageRegistry = class {
   supplied;
   byUrl = /* @__PURE__ */ new Map();
   ids = /* @__PURE__ */ new Set();
+  names = /* @__PURE__ */ new Set();
   parts = [];
   picSeq = 0;
   binSeq = 0;
@@ -29328,8 +29333,12 @@ var ImageRegistry = class {
       let n = 1;
       while (this.ids.has(itemId)) itemId = `${m[1].replace(/[^A-Za-z0-9_]/g, "_")}_${n++}`;
       this.ids.add(itemId);
+      let name = `BinData/${url}`;
+      let suffix = 1;
+      while (this.names.has(name)) name = `BinData/${m[1]}_${suffix++}.${m[2]}`;
+      this.names.add(name);
       part = {
-        name: `BinData/${url}`,
+        name,
         itemId,
         mime,
         data: m[2].toLowerCase() === "bmp" ? PLACEHOLDER_BMP : PLACEHOLDER_PNG
@@ -29358,8 +29367,9 @@ var ImageRegistry = class {
     const base = safe && !url.includes("..") ? safe[1] : `image${++this.binSeq}`;
     let itemId = base.replace(/[^A-Za-z0-9_]/g, "_");
     let n = 1;
-    while (this.ids.has(itemId)) itemId = `${base.replace(/[^A-Za-z0-9_]/g, "_")}_${n++}`;
+    while (this.ids.has(itemId) || this.names.has(`BinData/${itemId}.${sniffed.ext}`)) itemId = `${base.replace(/[^A-Za-z0-9_]/g, "_")}_${n++}`;
     this.ids.add(itemId);
+    this.names.add(`BinData/${itemId}.${sniffed.ext}`);
     const dim = probeImageSize(bytes);
     let wHU, hHU;
     if (dim && dim.w > 0 && dim.h > 0) {
@@ -34335,30 +34345,11 @@ async function resolveSectionEntryNames(zip) {
     const f = zip.file(mp);
     if (!f) continue;
     const xml = await f.async("text");
-    const paths = sectionPathsFromManifest(xml).filter((p) => zip.file(p) !== null);
+    const paths = parseSectionPathsFromManifest(xml).filter((p) => zip.file(p) !== null);
     if (paths.length > 0) return paths;
   }
   return Object.keys(zip.files).filter((n) => /[Ss]ection\d+\.xml$/.test(n)).sort(compareSectionPaths);
 }
-function sectionPathsFromManifest(xml) {
-  const attr = (tag, name) => {
-    const m = tag.match(new RegExp(`(?:^|\\s)${name}\\s*=\\s*(?:"([^"]*)"|'([^']*)')`));
-    return m ? m[1] ?? m[2] : "";
-  };
-  const idToHref = /* @__PURE__ */ new Map();
-  for (const m of xml.matchAll(/<opf:item(\s(?:"[^"]*"|'[^']*'|[^>"'])*?)\/?>/g)) {
-    const id = attr(m[1], "id");
-    const href = normalizeSectionHref(attr(m[1], "href"));
-    if (id && href) idToHref.set(id, href);
-  }
-  const ordered = [];
-  for (const m of xml.matchAll(/<opf:itemref(\s(?:"[^"]*"|'[^']*'|[^>"'])*?)\/?>/g)) {
-    const href = idToHref.get(attr(m[1], "idref"));
-    if (href) ordered.push(href);
-  }
-  if (ordered.length > 0) return ordered;
-  return Array.from(idToHref.values()).sort(compareSectionPaths);
-}
 
 // src/roundtrip/hwp5-patch.ts
 import { deflateRawSync as deflateRawSync2 } from "zlib";
@@ -41489,7 +41480,7 @@ async function parse(input, options) {
       buffer = toArrayBuffer(buf);
     } catch (err) {
       const msg2 = err instanceof Error && "code" in err && err.code === "ENOENT" ? `\uD30C\uC77C\uC744 \uCC3E\uC744 \uC218 \uC5C6\uC2B5\uB2C8\uB2E4: ${input}` : `\uD30C\uC77C \uC77D\uAE30 \uC2E4\uD328: ${input}`;
-      return { success: false, fileType: "unknown", error: msg2, code: "PARSE_ERROR" };
+      return { success: false, fileType: "unknown", error: msg2, code: classifyError(err) };
     }
   } else if (Buffer.isBuffer(input)) {
     buffer = toArrayBuffer(input);
@@ -41564,7 +41555,7 @@ function scriptsOff(r, off) {
 }
 async function parseImage(buffer, options) {
   try {
-    const { parseImageDocument } = await import("./image-ocr-POCME5I4.js");
+    const { parseImageDocument } = await import("./image-ocr-4BGX2EJJ.js");
     const { blocks, warnings } = await parseImageDocument(buffer, options);
     stripScriptTags({ blocks });
     return {
@@ -41624,7 +41615,7 @@ async function parseHwp(buffer, options) {
 async function parsePdf(buffer, options) {
   let parsePdfDocument;
   try {
-    const mod = await import("./parser-SCXRLOW7.js");
+    const mod = await import("./parser-GFDGAVLV.js");
     parsePdfDocument = mod.parsePdfDocument;
   } catch {
     return {