kordoc 4.18.13 → 4.19.0

package/dist/index.d.cts

@@ -116,6 +116,8 @@ interface IRTable {
     classification?: TableClassificationSummary;
     /** 원본 표 식별자(HWPX `hp:tbl id`) — 렌더 region(`RenderRegion.sourceId`)과의 조인 키 */
     sourceId?: string;
+    /** `tableFormat: "gfm"` 관계 표지의 표 id(`t1`, `t2` …) — 문서 블록 트리 전위 순회 순번. 그 옵션으로 렌더할 때만 채워진다(원본 id 아님) */
+    markdownId?: string;
     /** 렌더 인프라가 준 페이지 로컬 pt 조각(다중 페이지 표는 여럿) — extractTables 가 채운다 */
     regions?: BoundingBox[];
     /** 표 캡션 (예: "표 1. 부서별 예산") — v3.0 */
@@ -206,6 +208,9 @@ interface ParseOptions {
     /** 모든 표를 HTML 로 — 파이프 표도 HTML 표로 옮기고, 표마다 태그를 한 줄씩 들여써 낸다(BeautifulSoup prettify 모양, 첫 행 `<th>`).
      *  HTML 표만 다루는 소비자·채점기용. 기본 false(병합·중첩 없는 표는 GFM 파이프 표) */
     htmlTables?: boolean;
+    /** `"gfm"`: 모든 표를 HTML 없이 GFM 파이프 표로 — 세로 병합 칸은 덮인 행마다 같은 값, 가로 병합 칸은 시작 칸에만 값, 셀 안 표는 부모 표 뒤 독립 표로 꺼내고
+     *  부모·자식 관계를 `<!-- <table id="t2" parent_id="t1" /> -->` 표지로 남긴다(RAG 색인용). `htmlTables` 와 함께 쓸 수 없다. 기본: 병합·중첩 표는 HTML */
+    tableFormat?: "gfm";
     /** 표 오른쪽 끝의 빈 열(서식 문서의 입력란) 보존 (#47).
      *  기본 false: 마크다운 가독성을 위해 후행 빈 열을 트림.
      *  양식 인식 경로(parse_form·fill)는 내부적으로 항상 켠다. */
@@ -1568,7 +1573,13 @@ declare function applySplices(xml: string, splices: SpliceEdit[]): string;
 /** 2-pass colSpan/rowSpan 테이블 빌더 및 Markdown 변환 */
 
 declare function flattenLayoutTables(blocks: IRBlock[]): IRBlock[];
-declare function blocksToMarkdown(blocks: IRBlock[]): string;
+/** blocksToMarkdown 렌더 옵션 */
+interface MarkdownOptions {
+    /** 표 출력 형식 (`ParseOptions.tableFormat`). "gfm": 모든 표를 HTML 없이 GFM 파이프 표로 — 셀 안 표는 부모 뒤 독립 표로 꺼내고
+     *  부모·자식 관계를 HTML 주석 표지로 남긴다. 기본(미지정)은 병합·중첩 표를 HTML 로 */
+    tableFormat?: "gfm";
+}
+declare function blocksToMarkdown(blocks: IRBlock[], options?: MarkdownOptions): string;
 /**
  * 셀에 GFM 표 문법으로 담을 수 없는 구조 콘텐츠(중첩표·구분선)가 있는가 (#76 Task 4) — hasNestedTables 의 일반화.
  * 이미지는 GFM 셀에 `![image](src)` 로 인라인할 수 있어 여기 세지 않는다 — 대신 GFM 경로가 blocks 를 직접 직렬화해
@@ -1762,6 +1773,8 @@ interface ChunkOptions {
      * "block": IRBlock 1개 = 청크 1개
      */
     granularity?: "block" | "section";
+    /** 표 출력 형식 — `ParseOptions.tableFormat` 과 같다. "gfm" 이면 표 id 를 문서 전체 순번으로 붙인 뒤 청크마다 렌더한다 */
+    tableFormat?: "gfm";
 }
 declare function blocksToChunks(blocks: IRBlock[], options?: ChunkOptions): DocChunk[];
 
@@ -2340,4 +2353,4 @@ interface FillFormOutput {
  */
 declare function fillForm(input: string | ArrayBuffer | Buffer, values: Record<string, FillInput>, outputFormat?: FillOutputFormat): Promise<FillFormOutput>;
 
-export { BUILTIN_TEMPLATES, type BlockCapability, type BlockCapabilityInfo, type BlockDiff, type BlockEdit, type BlockSourceRef, type BorderDef, type BorderFillDef, type BoundingBox, type BuiltinTemplate, type CellCapability, type CellContext, type CellDiff, type CellProfile, type CharPrDef, type ChunkOptions, type ClassifyContext, type ClickHereField, DEFAULT_REDACT_RULES, type DiffChangeType, type DiffResult, type DocChunk, type DocumentMetadata, type DocumentQualitySummary, type ErrorCode, type ExtractRegionOptions, type ExtractTableVisualOptions, type ExtractedImage, type ExtractedTable, type ExtractedTableCrop, type FileType, type FillFormOutput, type FillInput, type FillOutputFormat, type FillResult, type FillValue, type FormField, type FormFieldSchema, type FormFieldType, type FormResult, type FormSchemaResult, type FormatProfile, type GongmunFont, type GongmunLintFinding, type GongmunNumbering, type GongmunOptions, type GongmunPreset, type GongmunPresetInput, type HwpxFillResult, HwpxSession, type HwpxTheme, type IRBlock, type IRBlockType, type IRCell, type IRTable, type ImageData, type InlineStyle, type MarkdownToHwpxOptions, type MeasureOptions, type MuncheLineKind, type MuncheLintFinding, type OcrProvider, type OutlineItem, PRESET_ALIAS, type PageBBox, type PageMargin, type PageMarkdown, type PageOptions, type PageQuality, type ParseFailure, type ParseOptions, type ParseResult, type ParseSuccess, type ParseWarning, type PatchOptions, type PatchResult, type PatchSkip, type PlaceSealResult, type PrintOptions, type PrintPreset, type RedactHit, type RedactOptions, type RedactRule, type RedactTextResult, type RegionAsset, type RenderAsset, type RenderDocumentOptions, type RenderDocumentResult, type RenderFormat, type RenderObjectType, type RenderRegion, type RenderScene, type RenderSourceFormat, type RenderSvgOptions, type RenderSvgResult, SPACE_EM_FIXED, SPACE_EM_FONT, type ScanCell, type ScanParaKind, type ScanParagraph, type ScanTable, type ScenePage, type SceneRenderOptions, type SceneRenderResult, type SealOp, type SealPlacement, type SectionScan, type SpliceEdit, type TRange, type TableProfile, type TableRepresentation, type TableVisualPolicy, VERSION, type ValidateIssue, type ValidateResult, ValueCursor, type WarningCode, type WatchOptions, type WrapMode, type WrapResult, applySplices, blocksToChunks, blocksToMarkdown, blocksToPages, blocksToPdf, buildParagraphSplices, buildRangeSplices, charWidthEm1000, chooseTableRepresentation, classifyTable, classifyTableTree, collectTableBlocks, compare, detectFormat, detectOle2Format, detectZipFormat, diffBlocks, extractClickHereFields, extractFormFields, extractFormSchema, extractRenderedRegions, extractTables, fillForm, fillFormFields, fillHwpx, fillWithUniqueGuard, fitRatioForFewerLines, flattenLayoutTables, formatFillValue, gongmunLintWarnings, hasStructuredCellContent, hwpxToProfile, incompatibleGongmunWarnings, inferFieldType, isHwpxFile, isKnownFont, isLabelCell, isOldHwpFile, isPdfFile, isZipFile, lintGongmunText, lintMuncheText, markdownToHwpx, markdownToPdf, measureTextWidth, muncheLintWarnings, normalizeGongmunPreset, openHwpxDocument, parse, parseDocx, parseHwp, parseHwp3, parseHwpml, parseHwpx, parseImage, parsePdf, parsePptx, parseXls, parseXlsx, patchHwp, patchHwpx, patchHwpxBlocks, placeSealHwpx, readBuiltinTemplate, readBuiltinTemplateSample, redactMarkdown, redactText, renderDocument, renderDocumentToScene, renderHtml, renderHwp5Pages, renderHwpxToSvg, renderSceneToHtml, resolveBuiltinTemplate, scanSectionXml, simulateWrap, simulateWrapKeepWord, unknownFontWarnings, usesGaejosikMunche, validateHwpx };
+export { BUILTIN_TEMPLATES, type BlockCapability, type BlockCapabilityInfo, type BlockDiff, type BlockEdit, type BlockSourceRef, type BorderDef, type BorderFillDef, type BoundingBox, type BuiltinTemplate, type CellCapability, type CellContext, type CellDiff, type CellProfile, type CharPrDef, type ChunkOptions, type ClassifyContext, type ClickHereField, DEFAULT_REDACT_RULES, type DiffChangeType, type DiffResult, type DocChunk, type DocumentMetadata, type DocumentQualitySummary, type ErrorCode, type ExtractRegionOptions, type ExtractTableVisualOptions, type ExtractedImage, type ExtractedTable, type ExtractedTableCrop, type FileType, type FillFormOutput, type FillInput, type FillOutputFormat, type FillResult, type FillValue, type FormField, type FormFieldSchema, type FormFieldType, type FormResult, type FormSchemaResult, type FormatProfile, type GongmunFont, type GongmunLintFinding, type GongmunNumbering, type GongmunOptions, type GongmunPreset, type GongmunPresetInput, type HwpxFillResult, HwpxSession, type HwpxTheme, type IRBlock, type IRBlockType, type IRCell, type IRTable, type ImageData, type InlineStyle, type MarkdownOptions, type MarkdownToHwpxOptions, type MeasureOptions, type MuncheLineKind, type MuncheLintFinding, type OcrProvider, type OutlineItem, PRESET_ALIAS, type PageBBox, type PageMargin, type PageMarkdown, type PageOptions, type PageQuality, type ParseFailure, type ParseOptions, type ParseResult, type ParseSuccess, type ParseWarning, type PatchOptions, type PatchResult, type PatchSkip, type PlaceSealResult, type PrintOptions, type PrintPreset, type RedactHit, type RedactOptions, type RedactRule, type RedactTextResult, type RegionAsset, type RenderAsset, type RenderDocumentOptions, type RenderDocumentResult, type RenderFormat, type RenderObjectType, type RenderRegion, type RenderScene, type RenderSourceFormat, type RenderSvgOptions, type RenderSvgResult, SPACE_EM_FIXED, SPACE_EM_FONT, type ScanCell, type ScanParaKind, type ScanParagraph, type ScanTable, type ScenePage, type SceneRenderOptions, type SceneRenderResult, type SealOp, type SealPlacement, type SectionScan, type SpliceEdit, type TRange, type TableProfile, type TableRepresentation, type TableVisualPolicy, VERSION, type ValidateIssue, type ValidateResult, ValueCursor, type WarningCode, type WatchOptions, type WrapMode, type WrapResult, applySplices, blocksToChunks, blocksToMarkdown, blocksToPages, blocksToPdf, buildParagraphSplices, buildRangeSplices, charWidthEm1000, chooseTableRepresentation, classifyTable, classifyTableTree, collectTableBlocks, compare, detectFormat, detectOle2Format, detectZipFormat, diffBlocks, extractClickHereFields, extractFormFields, extractFormSchema, extractRenderedRegions, extractTables, fillForm, fillFormFields, fillHwpx, fillWithUniqueGuard, fitRatioForFewerLines, flattenLayoutTables, formatFillValue, gongmunLintWarnings, hasStructuredCellContent, hwpxToProfile, incompatibleGongmunWarnings, inferFieldType, isHwpxFile, isKnownFont, isLabelCell, isOldHwpFile, isPdfFile, isZipFile, lintGongmunText, lintMuncheText, markdownToHwpx, markdownToPdf, measureTextWidth, muncheLintWarnings, normalizeGongmunPreset, openHwpxDocument, parse, parseDocx, parseHwp, parseHwp3, parseHwpml, parseHwpx, parseImage, parsePdf, parsePptx, parseXls, parseXlsx, patchHwp, patchHwpx, patchHwpxBlocks, placeSealHwpx, readBuiltinTemplate, readBuiltinTemplateSample, redactMarkdown, redactText, renderDocument, renderDocumentToScene, renderHtml, renderHwp5Pages, renderHwpxToSvg, renderSceneToHtml, resolveBuiltinTemplate, scanSectionXml, simulateWrap, simulateWrapKeepWord, unknownFontWarnings, usesGaejosikMunche, validateHwpx };

package/dist/index.d.ts

@@ -116,6 +116,8 @@ interface IRTable {
     classification?: TableClassificationSummary;
     /** 원본 표 식별자(HWPX `hp:tbl id`) — 렌더 region(`RenderRegion.sourceId`)과의 조인 키 */
     sourceId?: string;
+    /** `tableFormat: "gfm"` 관계 표지의 표 id(`t1`, `t2` …) — 문서 블록 트리 전위 순회 순번. 그 옵션으로 렌더할 때만 채워진다(원본 id 아님) */
+    markdownId?: string;
     /** 렌더 인프라가 준 페이지 로컬 pt 조각(다중 페이지 표는 여럿) — extractTables 가 채운다 */
     regions?: BoundingBox[];
     /** 표 캡션 (예: "표 1. 부서별 예산") — v3.0 */
@@ -206,6 +208,9 @@ interface ParseOptions {
     /** 모든 표를 HTML 로 — 파이프 표도 HTML 표로 옮기고, 표마다 태그를 한 줄씩 들여써 낸다(BeautifulSoup prettify 모양, 첫 행 `<th>`).
      *  HTML 표만 다루는 소비자·채점기용. 기본 false(병합·중첩 없는 표는 GFM 파이프 표) */
     htmlTables?: boolean;
+    /** `"gfm"`: 모든 표를 HTML 없이 GFM 파이프 표로 — 세로 병합 칸은 덮인 행마다 같은 값, 가로 병합 칸은 시작 칸에만 값, 셀 안 표는 부모 표 뒤 독립 표로 꺼내고
+     *  부모·자식 관계를 `<!-- <table id="t2" parent_id="t1" /> -->` 표지로 남긴다(RAG 색인용). `htmlTables` 와 함께 쓸 수 없다. 기본: 병합·중첩 표는 HTML */
+    tableFormat?: "gfm";
     /** 표 오른쪽 끝의 빈 열(서식 문서의 입력란) 보존 (#47).
      *  기본 false: 마크다운 가독성을 위해 후행 빈 열을 트림.
      *  양식 인식 경로(parse_form·fill)는 내부적으로 항상 켠다. */
@@ -1568,7 +1573,13 @@ declare function applySplices(xml: string, splices: SpliceEdit[]): string;
 /** 2-pass colSpan/rowSpan 테이블 빌더 및 Markdown 변환 */
 
 declare function flattenLayoutTables(blocks: IRBlock[]): IRBlock[];
-declare function blocksToMarkdown(blocks: IRBlock[]): string;
+/** blocksToMarkdown 렌더 옵션 */
+interface MarkdownOptions {
+    /** 표 출력 형식 (`ParseOptions.tableFormat`). "gfm": 모든 표를 HTML 없이 GFM 파이프 표로 — 셀 안 표는 부모 뒤 독립 표로 꺼내고
+     *  부모·자식 관계를 HTML 주석 표지로 남긴다. 기본(미지정)은 병합·중첩 표를 HTML 로 */
+    tableFormat?: "gfm";
+}
+declare function blocksToMarkdown(blocks: IRBlock[], options?: MarkdownOptions): string;
 /**
  * 셀에 GFM 표 문법으로 담을 수 없는 구조 콘텐츠(중첩표·구분선)가 있는가 (#76 Task 4) — hasNestedTables 의 일반화.
  * 이미지는 GFM 셀에 `![image](src)` 로 인라인할 수 있어 여기 세지 않는다 — 대신 GFM 경로가 blocks 를 직접 직렬화해
@@ -1762,6 +1773,8 @@ interface ChunkOptions {
      * "block": IRBlock 1개 = 청크 1개
      */
     granularity?: "block" | "section";
+    /** 표 출력 형식 — `ParseOptions.tableFormat` 과 같다. "gfm" 이면 표 id 를 문서 전체 순번으로 붙인 뒤 청크마다 렌더한다 */
+    tableFormat?: "gfm";
 }
 declare function blocksToChunks(blocks: IRBlock[], options?: ChunkOptions): DocChunk[];
 
@@ -2340,4 +2353,4 @@ interface FillFormOutput {
  */
 declare function fillForm(input: string | ArrayBuffer | Buffer, values: Record<string, FillInput>, outputFormat?: FillOutputFormat): Promise<FillFormOutput>;
 
-export { BUILTIN_TEMPLATES, type BlockCapability, type BlockCapabilityInfo, type BlockDiff, type BlockEdit, type BlockSourceRef, type BorderDef, type BorderFillDef, type BoundingBox, type BuiltinTemplate, type CellCapability, type CellContext, type CellDiff, type CellProfile, type CharPrDef, type ChunkOptions, type ClassifyContext, type ClickHereField, DEFAULT_REDACT_RULES, type DiffChangeType, type DiffResult, type DocChunk, type DocumentMetadata, type DocumentQualitySummary, type ErrorCode, type ExtractRegionOptions, type ExtractTableVisualOptions, type ExtractedImage, type ExtractedTable, type ExtractedTableCrop, type FileType, type FillFormOutput, type FillInput, type FillOutputFormat, type FillResult, type FillValue, type FormField, type FormFieldSchema, type FormFieldType, type FormResult, type FormSchemaResult, type FormatProfile, type GongmunFont, type GongmunLintFinding, type GongmunNumbering, type GongmunOptions, type GongmunPreset, type GongmunPresetInput, type HwpxFillResult, HwpxSession, type HwpxTheme, type IRBlock, type IRBlockType, type IRCell, type IRTable, type ImageData, type InlineStyle, type MarkdownToHwpxOptions, type MeasureOptions, type MuncheLineKind, type MuncheLintFinding, type OcrProvider, type OutlineItem, PRESET_ALIAS, type PageBBox, type PageMargin, type PageMarkdown, type PageOptions, type PageQuality, type ParseFailure, type ParseOptions, type ParseResult, type ParseSuccess, type ParseWarning, type PatchOptions, type PatchResult, type PatchSkip, type PlaceSealResult, type PrintOptions, type PrintPreset, type RedactHit, type RedactOptions, type RedactRule, type RedactTextResult, type RegionAsset, type RenderAsset, type RenderDocumentOptions, type RenderDocumentResult, type RenderFormat, type RenderObjectType, type RenderRegion, type RenderScene, type RenderSourceFormat, type RenderSvgOptions, type RenderSvgResult, SPACE_EM_FIXED, SPACE_EM_FONT, type ScanCell, type ScanParaKind, type ScanParagraph, type ScanTable, type ScenePage, type SceneRenderOptions, type SceneRenderResult, type SealOp, type SealPlacement, type SectionScan, type SpliceEdit, type TRange, type TableProfile, type TableRepresentation, type TableVisualPolicy, VERSION, type ValidateIssue, type ValidateResult, ValueCursor, type WarningCode, type WatchOptions, type WrapMode, type WrapResult, applySplices, blocksToChunks, blocksToMarkdown, blocksToPages, blocksToPdf, buildParagraphSplices, buildRangeSplices, charWidthEm1000, chooseTableRepresentation, classifyTable, classifyTableTree, collectTableBlocks, compare, detectFormat, detectOle2Format, detectZipFormat, diffBlocks, extractClickHereFields, extractFormFields, extractFormSchema, extractRenderedRegions, extractTables, fillForm, fillFormFields, fillHwpx, fillWithUniqueGuard, fitRatioForFewerLines, flattenLayoutTables, formatFillValue, gongmunLintWarnings, hasStructuredCellContent, hwpxToProfile, incompatibleGongmunWarnings, inferFieldType, isHwpxFile, isKnownFont, isLabelCell, isOldHwpFile, isPdfFile, isZipFile, lintGongmunText, lintMuncheText, markdownToHwpx, markdownToPdf, measureTextWidth, muncheLintWarnings, normalizeGongmunPreset, openHwpxDocument, parse, parseDocx, parseHwp, parseHwp3, parseHwpml, parseHwpx, parseImage, parsePdf, parsePptx, parseXls, parseXlsx, patchHwp, patchHwpx, patchHwpxBlocks, placeSealHwpx, readBuiltinTemplate, readBuiltinTemplateSample, redactMarkdown, redactText, renderDocument, renderDocumentToScene, renderHtml, renderHwp5Pages, renderHwpxToSvg, renderSceneToHtml, resolveBuiltinTemplate, scanSectionXml, simulateWrap, simulateWrapKeepWord, unknownFontWarnings, usesGaejosikMunche, validateHwpx };
+export { BUILTIN_TEMPLATES, type BlockCapability, type BlockCapabilityInfo, type BlockDiff, type BlockEdit, type BlockSourceRef, type BorderDef, type BorderFillDef, type BoundingBox, type BuiltinTemplate, type CellCapability, type CellContext, type CellDiff, type CellProfile, type CharPrDef, type ChunkOptions, type ClassifyContext, type ClickHereField, DEFAULT_REDACT_RULES, type DiffChangeType, type DiffResult, type DocChunk, type DocumentMetadata, type DocumentQualitySummary, type ErrorCode, type ExtractRegionOptions, type ExtractTableVisualOptions, type ExtractedImage, type ExtractedTable, type ExtractedTableCrop, type FileType, type FillFormOutput, type FillInput, type FillOutputFormat, type FillResult, type FillValue, type FormField, type FormFieldSchema, type FormFieldType, type FormResult, type FormSchemaResult, type FormatProfile, type GongmunFont, type GongmunLintFinding, type GongmunNumbering, type GongmunOptions, type GongmunPreset, type GongmunPresetInput, type HwpxFillResult, HwpxSession, type HwpxTheme, type IRBlock, type IRBlockType, type IRCell, type IRTable, type ImageData, type InlineStyle, type MarkdownOptions, type MarkdownToHwpxOptions, type MeasureOptions, type MuncheLineKind, type MuncheLintFinding, type OcrProvider, type OutlineItem, PRESET_ALIAS, type PageBBox, type PageMargin, type PageMarkdown, type PageOptions, type PageQuality, type ParseFailure, type ParseOptions, type ParseResult, type ParseSuccess, type ParseWarning, type PatchOptions, type PatchResult, type PatchSkip, type PlaceSealResult, type PrintOptions, type PrintPreset, type RedactHit, type RedactOptions, type RedactRule, type RedactTextResult, type RegionAsset, type RenderAsset, type RenderDocumentOptions, type RenderDocumentResult, type RenderFormat, type RenderObjectType, type RenderRegion, type RenderScene, type RenderSourceFormat, type RenderSvgOptions, type RenderSvgResult, SPACE_EM_FIXED, SPACE_EM_FONT, type ScanCell, type ScanParaKind, type ScanParagraph, type ScanTable, type ScenePage, type SceneRenderOptions, type SceneRenderResult, type SealOp, type SealPlacement, type SectionScan, type SpliceEdit, type TRange, type TableProfile, type TableRepresentation, type TableVisualPolicy, VERSION, type ValidateIssue, type ValidateResult, ValueCursor, type WarningCode, type WatchOptions, type WrapMode, type WrapResult, applySplices, blocksToChunks, blocksToMarkdown, blocksToPages, blocksToPdf, buildParagraphSplices, buildRangeSplices, charWidthEm1000, chooseTableRepresentation, classifyTable, classifyTableTree, collectTableBlocks, compare, detectFormat, detectOle2Format, detectZipFormat, diffBlocks, extractClickHereFields, extractFormFields, extractFormSchema, extractRenderedRegions, extractTables, fillForm, fillFormFields, fillHwpx, fillWithUniqueGuard, fitRatioForFewerLines, flattenLayoutTables, formatFillValue, gongmunLintWarnings, hasStructuredCellContent, hwpxToProfile, incompatibleGongmunWarnings, inferFieldType, isHwpxFile, isKnownFont, isLabelCell, isOldHwpFile, isPdfFile, isZipFile, lintGongmunText, lintMuncheText, markdownToHwpx, markdownToPdf, measureTextWidth, muncheLintWarnings, normalizeGongmunPreset, openHwpxDocument, parse, parseDocx, parseHwp, parseHwp3, parseHwpml, parseHwpx, parseImage, parsePdf, parsePptx, parseXls, parseXlsx, patchHwp, patchHwpx, patchHwpxBlocks, placeSealHwpx, readBuiltinTemplate, readBuiltinTemplateSample, redactMarkdown, redactText, renderDocument, renderDocumentToScene, renderHtml, renderHwp5Pages, renderHwpxToSvg, renderSceneToHtml, resolveBuiltinTemplate, scanSectionXml, simulateWrap, simulateWrapKeepWord, unknownFontWarnings, usesGaejosikMunche, validateHwpx };

package/dist/index.js

@@ -1,7 +1,7 @@
 import {
   blocksToPages,
   inlineImagesIntoMarkdown
-} from "./chunk-VDEENOPW.js";
+} from "./chunk-W5XX2GKO.js";
 import {
   parsePageRange
 } from "./chunk-V4BH5SRQ.js";
@@ -13,6 +13,7 @@ import {
   MAX_COLS,
   MAX_ROWS,
   MAX_TABLE_CELLS,
+  assignTableIds,
   blocksToMarkdown,
   buildTable,
   convertTableToText,
@@ -31,7 +32,7 @@ import {
   tidyScriptTags,
   unframeLayoutTables,
   wrapScript
-} from "./chunk-JPSLMPFI.js";
+} from "./chunk-DH75ZUYF.js";
 import {
   KordocError,
   OPTIONAL_DEP_INSTALL_HINT,
@@ -49,7 +50,7 @@ import {
   toArrayBuffer,
   unescapeHtml,
   unzipLimitBytes
-} from "./chunk-EBXJ45VZ.js";
+} from "./chunk-5B55S4JU.js";
 
 // src/plain-markdown.ts
 function toPlainMarkdown(md2) {
@@ -3667,7 +3668,7 @@ import JSZip2 from "jszip";
 
 // src/hwpx/zip-sections.ts
 import { inflateRawSync as inflateRawSync3 } from "zlib";
-function extractFromBrokenZip(buffer) {
+function extractFromBrokenZip(buffer, options) {
   const data = new Uint8Array(buffer);
   const view = new DataView(buffer);
   let pos = 0;
@@ -3734,7 +3735,7 @@ function extractFromBrokenZip(buffer) {
   }
   if (blocks.length === 0) throw new KordocError("\uC190\uC0C1\uB41C HWPX\uC5D0\uC11C \uC139\uC158 \uB370\uC774\uD130\uB97C \uBCF5\uAD6C\uD560 \uC218 \uC5C6\uC2B5\uB2C8\uB2E4");
   applyPageText(blocks, shared);
-  const markdown = blocksToMarkdown(blocks);
+  const markdown = blocksToMarkdown(blocks, { tableFormat: options?.tableFormat });
   return { markdown, blocks, warnings: warnings.length > 0 ? warnings : void 0 };
 }
 async function readKordocLayout(zip) {
@@ -3895,7 +3896,7 @@ async function parseHwpxDocument(buffer, options) {
   try {
     zip = await JSZip3.loadAsync(buffer);
   } catch {
-    return extractFromBrokenZip(buffer);
+    return extractFromBrokenZip(buffer, options);
   }
   const actualEntryCount = Object.keys(zip.files).length;
   if (actualEntryCount > MAX_ZIP_ENTRIES) {
@@ -3977,7 +3978,7 @@ async function parseHwpxDocument(buffer, options) {
   const shown = options?.layoutTables === "keep" ? blocks : unframeLayoutTables(blocks, !!options?.keepTrailingEmptyCols);
   detectHwpxHeadings(shown, styleMap, shared.tocTexts);
   const outline = shown.filter((b) => b.type === "heading" && b.level && b.text).map((b) => ({ level: b.level, text: b.text, pageNumber: b.pageNumber }));
-  const markdown = blocksToMarkdown(shown);
+  const markdown = blocksToMarkdown(shown, { tableFormat: options?.tableFormat });
   return { markdown, blocks: shown, metadata, outline: outline.length > 0 ? outline : void 0, warnings: warnings.length > 0 ? warnings : void 0, images: images.length > 0 ? images : void 0 };
 }
 
@@ -6097,7 +6098,7 @@ function parseHwp5Document(buffer, options) {
     detectHwp5Headings(flatBlocks, docInfo, doc.tocTexts);
   }
   const outline = flatBlocks.filter((b) => b.type === "heading" && b.level && b.text).map((b) => ({ level: b.level, text: b.text, pageNumber: b.pageNumber }));
-  let markdown = blocksToMarkdown(flatBlocks);
+  let markdown = blocksToMarkdown(flatBlocks, { tableFormat: options?.tableFormat });
   if (options?.inlineImages && options.images !== false && images.length > 0) {
     try {
       markdown = inlineImagesIntoMarkdown(markdown, images, { compress: true });
@@ -18982,7 +18983,7 @@ function parseHwp3Document(buffer, options) {
     version: "3.0"
   };
   return {
-    markdown: blocksToMarkdown(blocks),
+    markdown: blocksToMarkdown(blocks, { tableFormat: options?.tableFormat }),
     blocks,
     metadata,
     warnings: warnings.length ? warnings : void 0
@@ -19724,7 +19725,7 @@ async function parseXlsxDocument(buffer, options) {
     } catch {
     }
   }
-  const markdown = blocksToMarkdown(blocks);
+  const markdown = blocksToMarkdown(blocks, { tableFormat: options?.tableFormat });
   return { markdown, blocks, metadata, warnings: warnings.length > 0 ? warnings : void 0 };
 }
 
@@ -20332,7 +20333,7 @@ async function parseXlsDocument(buffer, options) {
     pageCount: totalSheets
   };
   return {
-    markdown: blocksToMarkdown(allBlocks),
+    markdown: blocksToMarkdown(allBlocks, { tableFormat: options?.tableFormat }),
     blocks: allBlocks,
     metadata,
     warnings: warnings.length > 0 ? warnings : void 0
@@ -21496,7 +21497,7 @@ async function parseDocxDocument(buffer, options) {
     }
   }
   const outline = blocks.filter((b) => b.type === "heading").map((b) => ({ level: b.level ?? 2, text: b.text ?? "" }));
-  const markdown = blocksToMarkdown(blocks);
+  const markdown = blocksToMarkdown(blocks, { tableFormat: options?.tableFormat });
   return {
     markdown,
     blocks,
@@ -21818,7 +21819,7 @@ async function parsePptxDocument(buffer, options) {
   }
   const outline = blocks.filter((b) => b.type === "heading").map((b) => ({ level: b.level ?? 2, text: b.text ?? "", pageNumber: b.pageNumber }));
   return {
-    markdown: blocksToMarkdown(blocks),
+    markdown: blocksToMarkdown(blocks, { tableFormat: options?.tableFormat }),
     blocks,
     metadata,
     outline: outline.length > 0 ? outline : void 0,
@@ -21878,7 +21879,7 @@ function parseHwpmlDocument(buffer, options) {
     parseSection2(el, blocks, paraShapeMap, sectionIdx, warnings, options?.keepTrailingEmptyCols ?? false);
   }
   const outline = blocks.filter((b) => b.type === "heading" && b.text).map((b) => ({ level: b.level ?? 1, text: b.text, pageNumber: b.pageNumber }));
-  const markdown = blocksToMarkdown(blocks);
+  const markdown = blocksToMarkdown(blocks, { tableFormat: options?.tableFormat });
   return {
     markdown,
     blocks,
@@ -38603,6 +38604,8 @@ function sameBreadcrumb(a, b) {
 function blocksToChunks(blocks, options) {
   const granularity = options?.granularity ?? "section";
   const includeCells = options?.includeTableCells ?? false;
+  const mdOptions = { tableFormat: options?.tableFormat };
+  if (options?.tableFormat === "gfm") assignTableIds(blocks);
   const chunks = [];
   const headingStack = [];
   const listStack = [];
@@ -38616,12 +38619,12 @@ function blocksToChunks(blocks, options) {
   let run = null;
   const flushRun = () => {
     if (!run) return;
-    push("text", run.breadcrumb, blocksToMarkdown(run.blocks), [run.start, run.end], run.page);
+    push("text", run.breadcrumb, blocksToMarkdown(run.blocks, mdOptions), [run.start, run.end], run.page);
     run = null;
   };
   for (let i = 0; i < blocks.length; i++) {
     const block = blocks[i];
-    const md2 = blocksToMarkdown([block]);
+    const md2 = blocksToMarkdown([block], mdOptions);
     if (!md2) continue;
     if (block.type === "heading") {
       flushRun();
@@ -41553,6 +41556,9 @@ async function extractTables(input, options = {}) {
 async function parse(input, options) {
   let buffer;
   const opts = typeof input === "string" && !options?.filePath ? { ...options, filePath: input } : options;
+  if (opts?.htmlTables && opts.tableFormat) {
+    return { success: false, fileType: "unknown", error: "htmlTables \uC640 tableFormat \uC740 \uD568\uAED8 \uC4F8 \uC218 \uC5C6\uC2B5\uB2C8\uB2E4" };
+  }
   if (typeof input === "string") {
     try {
       const buf = await readFile3(input);
@@ -41581,7 +41587,7 @@ async function parse(input, options) {
     out = { ...result, warnings: [...result.warnings ?? [], { code: "PARTIAL_PARSE", message: `\uC694\uCCAD\uD55C \uCABD \uBC94\uC704(${opts.pages})\uAC00 \uBB38\uC11C \uCABD(1~${result.pageCount})\uACFC \uACB9\uCE58\uC9C0 \uC54A\uC2B5\uB2C8\uB2E4` }] };
   }
   if (out.success && !out.pages) {
-    const pages = blocksToPages(out.blocks);
+    const pages = blocksToPages(out.blocks, (bs) => blocksToMarkdown(bs, { tableFormat: opts?.tableFormat }));
     if (pages) out = { ...out, pages };
   }
   if (out.success && opts?.plain) {
@@ -41634,13 +41640,13 @@ function scriptsOff(r, off) {
 }
 async function parseImage(buffer, options) {
   try {
-    const { parseImageDocument } = await import("./image-ocr-VH54SDZ7.js");
+    const { parseImageDocument } = await import("./image-ocr-B32G24XW.js");
     const { blocks, warnings } = await parseImageDocument(buffer, options);
     stripScriptTags({ blocks });
     return {
       success: true,
       fileType: "image",
-      markdown: blocksToMarkdown(blocks),
+      markdown: blocksToMarkdown(blocks, { tableFormat: options?.tableFormat }),
       blocks,
       warnings: warnings.length ? warnings : void 0,
       pageCount: 1
@@ -41694,7 +41700,7 @@ async function parseHwp(buffer, options) {
 async function parsePdf(buffer, options) {
   let parsePdfDocument;
   try {
-    const mod = await import("./parser-Z6PHB54O.js");
+    const mod = await import("./parser-YW3NMSJE.js");
     parsePdfDocument = mod.parsePdfDocument;
   } catch {
     return {