@bendyline/squisq-formats 2.8.5 → 2.8.7

package/README.md

@@ -87,7 +87,7 @@ const imported = await pdfToMarkdownDoc(pdfBuffer);
 const container = await pdfToContainer(pdfBuffer);
 ```
 
-**Fidelity:** export uses pdf-lib's standard 14 fonts (`themeId` affects colors only). Text they cannot encode — Greek, Cyrillic, Vietnamese, arrows, math, box drawing, symbols — is drawn with an embedded Unicode font: the bundled DejaVu Sans faces, loaded and subset only for a document that needs them (`unicodeFonts` supplies your own faces, or `false` turns this off). CJK and emoji are still replaced, and reported once through `onWarning`. Pages stay white for printing. Theme text, heading, and link colors are retained when they meet 4.5:1 contrast against both white paper and the shaded table header; otherwise the exporter uses its default dark ink for that role. This also applies to inherited and custom themes. Pass `images` (keyed by the markdown URL, as for DOCX) to draw PNG and JPEG images that stand alone in a paragraph; other images print as an `[Image: alt]` note. PDF has no semantic structure, so import is heuristic and best-effort: headings are detected by font size, and tables / code blocks / blockquotes / links via the `detectTables` / `detectCodeBlocks` / `detectBlockquotes` / `detectLinks` options (all default true). `pdfToMarkdownDoc` is text-only; `pdfToContainer` also extracts embedded XObject and inline raster images in browsers and Node, placed **by page** — each image is inserted after the last content block from its page (image-only pages fall back to the nearest preceding page with content, else the document end). Placement is page-level only; masks, vector artwork, and images that PDF.js cannot expose as decoded pixels are not emitted.
+**Fidelity:** export uses pdf-lib's standard 14 fonts (`themeId` affects colors only). Text they cannot encode — Greek, Cyrillic, Vietnamese, arrows, math, box drawing, symbols — is drawn with an embedded Unicode font: the bundled DejaVu Sans faces, loaded and subset only for a document that needs them (`unicodeFonts` supplies your own faces, or `false` turns this off). CJK and most emoji are still replaced, and reported once through `onWarning`. Lines break at a `<br>` (the only line break a table cell can hold), table columns are sized to their content, and a word too long for its line or cell breaks between characters. Pages stay white for printing. Theme text, heading, and link colors are retained when they meet 4.5:1 contrast against both white paper and the shaded table header; otherwise the exporter uses its default dark ink for that role. This also applies to inherited and custom themes. Pass `images` (keyed by the markdown URL, as for DOCX) to draw PNG and JPEG images that stand alone in a paragraph; other images print as an `[Image: alt]` note. PDF has no semantic structure, so import is heuristic and best-effort: headings are detected by font size, and tables / code blocks / blockquotes / links via the `detectTables` / `detectCodeBlocks` / `detectBlockquotes` / `detectLinks` options (all default true). `pdfToMarkdownDoc` is text-only; `pdfToContainer` also extracts embedded XObject and inline raster images in browsers and Node, placed **by page** — each image is inserted after the last content block from its page (image-only pages fall back to the nearest preceding page with content, else the document end). Placement is page-level only; masks, vector artwork, and images that PDF.js cannot expose as decoded pixels are not emitted.
 
 ### Diagrams as pictures
 

package/dist/{chunk-ZULJFSA4.js → chunk-4A3EHQJP.js}

@@ -1,7 +1,7 @@
 import {
   extractPlainText,
   inlineToPlainText
-} from "./chunk-AVOZAKGP.js";
+} from "./chunk-BZ5SHMOY.js";
 import {
   MAX_COL_INDEX,
   MAX_ROW_INDEX,

package/dist/{chunk-AVOZAKGP.js → chunk-BZ5SHMOY.js}

@@ -1,4 +1,5 @@
 // src/shared/text.ts
+import { isHtmlLineBreak } from "@bendyline/squisq/markdown";
 function stripHtmlTags(html) {
   return html.replace(/<[^>]*>/g, "");
 }
@@ -21,6 +22,8 @@ function inlineToPlainText(node) {
       return `{[${node.token}]}`;
     case "break":
       return " ";
+    case "htmlInline":
+      return isHtmlLineBreak(node.rawHtml) ? " " : "";
     default:
       return "";
   }

package/dist/{chunk-CJUOAGGS.js → chunk-DHAK4PII.js}

@@ -1,6 +1,6 @@
 import {
   stripHtmlTags
-} from "./chunk-AVOZAKGP.js";
+} from "./chunk-BZ5SHMOY.js";
 
 // src/ooxml/relIds.ts
 var RelIdAllocator = class {
@@ -75,6 +75,7 @@ function hasControls(value) {
 }
 
 // src/shared/inlineRuns.ts
+import { isHtmlLineBreak } from "@bendyline/squisq/markdown";
 function inlineNodesToRuns(nodes, handlers, format = {}) {
   let out = "";
   for (const node of nodes) {
@@ -101,6 +102,7 @@ function inlineNodeToRuns(node, handlers, format) {
     case "inlineMath":
       return handlers.run(node.value, { ...format, code: true });
     case "htmlInline":
+      if (isHtmlLineBreak(node.rawHtml)) return handlers.lineBreak();
       return handlers.run(stripHtmlTags(node.rawHtml), format);
     case "link":
       if (node.wikiTarget !== void 0 && !node.url)

package/dist/{chunk-AOY3UPE3.js → chunk-IYNDHW2C.js}

@@ -1,7 +1,7 @@
 import {
   convert,
   defaultRegistry
-} from "./chunk-X6WFQDZD.js";
+} from "./chunk-ALDCFHZT.js";
 import {
   ConversionError
 } from "./chunk-KXOZMWBS.js";

package/dist/{chunk-BIERSSDF.js → chunk-KROH7FNT.js}

@@ -5,7 +5,7 @@ import {
   inlineNodesToRuns,
   normalizeOoxmlHex,
   sanitizeOfficeHyperlink
-} from "./chunk-CJUOAGGS.js";
+} from "./chunk-DHAK4PII.js";
 import {
   docHasSvgIslands,
   isSvgImageEntry,
@@ -14,7 +14,7 @@ import {
 } from "./chunk-VCZUXQUL.js";
 import {
   stripHtmlTags
-} from "./chunk-AVOZAKGP.js";
+} from "./chunk-BZ5SHMOY.js";
 import {
   createPackage
 } from "./chunk-ILCJ3WFD.js";

package/dist/{chunk-QRDJOYT3.js → chunk-QU4EVC62.js}

@@ -8,7 +8,7 @@ import {
   inlineNodesToRuns,
   sanitizeOfficeHyperlink,
   toOoxmlHex
-} from "./chunk-CJUOAGGS.js";
+} from "./chunk-DHAK4PII.js";
 import {
   svgImageFromBytes,
   svgIslandImage
@@ -16,7 +16,7 @@ import {
 import {
   extractPlainText,
   stripHtmlTags
-} from "./chunk-AVOZAKGP.js";
+} from "./chunk-BZ5SHMOY.js";
 import {
   createPackage
 } from "./chunk-ILCJ3WFD.js";

package/dist/{chunk-GESOAVYN.js → chunk-SVYTUYZM.js}

@@ -8,7 +8,7 @@ import {
 } from "./chunk-VCZUXQUL.js";
 import {
   extractPlainText
-} from "./chunk-AVOZAKGP.js";
+} from "./chunk-BZ5SHMOY.js";
 import {
   escapeXml
 } from "./chunk-JU2RHXUB.js";
@@ -29,7 +29,7 @@ import {
 import JSZip from "jszip";
 import { resolveFontFamily } from "@bendyline/squisq/schemas";
 import { resolveThemeForDoc } from "@bendyline/squisq/doc";
-import { readFrontmatterThemeId, sanitizeUrl } from "@bendyline/squisq/markdown";
+import { isHtmlLineBreak, readFrontmatterThemeId, sanitizeUrl } from "@bendyline/squisq/markdown";
 async function markdownDocToEpub(sourceDoc, options = {}) {
   options.signal?.throwIfAborted();
   const themeId = options.themeId ?? readFrontmatterThemeId(sourceDoc.frontmatter);
@@ -442,6 +442,7 @@ function inlineToXhtml(node, ctx) {
     case "inlineMath":
       return `<span class="math">${escapeXml(node.value)}</span>`;
     case "htmlInline":
+      if (isHtmlLineBreak(node.rawHtml)) return "<br/>";
       return escapeXml(node.rawHtml.replace(/<[^>]+>/g, ""));
     case "inlineIcon": {
       const icon = node;

package/dist/{chunk-GLIUMDYD.js → chunk-TMOXUUTG.js}

@@ -23,7 +23,7 @@ import { PDFDocument, StandardFonts, rgb, PDFString } from "pdf-lib";
 import fontkit2 from "@pdf-lib/fontkit";
 import { contrastRatio } from "@bendyline/squisq/schemas";
 import { docToMarkdown, resolveThemeForDoc } from "@bendyline/squisq/doc";
-import { readFrontmatterThemeId, sanitizeUrl } from "@bendyline/squisq/markdown";
+import { isHtmlLineBreak, readFrontmatterThemeId, sanitizeUrl } from "@bendyline/squisq/markdown";
 
 // src/pdf/winAnsi.ts
 var CP1252_HIGH_SPECIALS = /* @__PURE__ */ new Set([
@@ -333,7 +333,7 @@ var WinAnsiTracker = class {
   warning(options = {}) {
     if (this.count === 0) return void 0;
     if (options.unicodeFont) {
-      return `PDF export replaced ${this.count} character(s) that its fonts cannot draw (${this.samples.join(", ")}). Scripts such as CJK and emoji are not available in PDF export.`;
+      return `PDF export replaced ${this.count} character(s) that its fonts cannot draw (${this.samples.join(", ")}). The PDF fonts do not cover CJK scripts or most emoji.`;
     }
     return `PDF export replaced ${this.count} character(s) that the standard PDF fonts cannot encode (${this.samples.join(", ")}). Scripts such as emoji, CJK and Cyrillic require an embedded Unicode font.`;
   }
@@ -741,16 +741,13 @@ function flattenInlines(nodes, ctx, state) {
         break;
       }
       case "break":
-        spans.push({
-          text: "\n",
-          font: ctx.fonts.regular,
-          fontSize: ctx.fontSize,
-          color: ctx.colors.text
-        });
+        spans.push(lineBreakSpan(ctx));
         break;
       case "htmlInline": {
         const html = node.rawHtml;
-        if (html) {
+        if (isHtmlLineBreak(html)) {
+          spans.push(lineBreakSpan(ctx));
+        } else if (html) {
           spans.push(
             ...textSpans(ctx, html, ctx.fonts.mono, {
               fontSize: CODE_FONT_SIZE,
@@ -799,6 +796,9 @@ function flattenInlines(nodes, ctx, state) {
   }
   return spans;
 }
+function lineBreakSpan(ctx) {
+  return { text: "\n", font: ctx.fonts.regular, fontSize: ctx.fontSize, color: ctx.colors.text };
+}
 var STANDARD_FONT_NAMES = new Set(Object.values(StandardFonts));
 function drawnWidth(font, text, size) {
   if (!STANDARD_FONT_NAMES.has(font.name)) return font.widthOfTextAtSize(text, size);
@@ -895,7 +895,7 @@ function wrapSpans(spans, maxWidth) {
       lineWidth = 0;
       continue;
     }
-    const words = splitIntoWords(span.text);
+    const words = splitIntoWords(span.text).flatMap((word) => breakLongWord(word, span, maxWidth));
     for (const word of words) {
       const wordWidth = drawnWidth(span.font, word, span.fontSize);
       if (lineWidth + wordWidth > maxWidth && currentLine.length > 0 && word.trim().length > 0) {
@@ -916,6 +916,42 @@ function wrapSpans(spans, maxWidth) {
   }
   return lines.length > 0 ? lines : [[]];
 }
+function breakLongWord(word, span, maxWidth) {
+  if (word.trim().length === 0 || drawnWidth(span.font, word, span.fontSize) <= maxWidth) {
+    return [word];
+  }
+  const pieces = [];
+  let piece = "";
+  let pieceWidth = 0;
+  for (const cluster of characterClusters(word)) {
+    const width = drawnWidth(span.font, cluster, span.fontSize);
+    if (piece && pieceWidth + width > maxWidth) {
+      pieces.push(piece);
+      piece = "";
+      pieceWidth = 0;
+    }
+    piece += cluster;
+    pieceWidth += width;
+  }
+  if (piece) pieces.push(piece);
+  return pieces;
+}
+function continuesCluster(ch) {
+  const codePoint = ch.codePointAt(0);
+  return /\p{M}/u.test(ch) || codePoint === 8205 || codePoint === 65038 || codePoint === 65039 || codePoint >= 127995 && codePoint <= 127999;
+}
+function characterClusters(text) {
+  const clusters = [];
+  for (const ch of text) {
+    const last = clusters.length - 1;
+    if (last >= 0 && (continuesCluster(ch) || clusters[last].endsWith("\u200D"))) {
+      clusters[last] += ch;
+    } else {
+      clusters.push(ch);
+    }
+  }
+  return clusters;
+}
 function splitIntoWords(text) {
   const tokens = [];
   let current = "";
@@ -1145,18 +1181,29 @@ function renderTable(node, ctx, extraIndent) {
     ...node.children.map((row) => row.children.length)
   );
   if (numCols === 0) return;
-  const colWidth = tableWidth / numCols;
-  for (let rowIdx = 0; rowIdx < node.children.length; rowIdx++) {
-    const row = node.children[rowIdx];
+  const rows = node.children.map((rowNode, rowIdx) => {
+    const row = rowNode;
     const isHeader = rowIdx === 0 && row.children.some((c) => c.isHeader);
-    const rowCellHeights = row.children.map((cell) => {
-      const cellNode = cell;
-      const spans = flattenInlines(cellNode.children, ctx, {
+    const cells = row.children.map(
+      (cell) => flattenInlines(cell.children, ctx, {
         bold: isHeader,
         italic: false,
         code: false
-      });
-      const lines = wrapSpans(spans, colWidth - 2 * TABLE_CELL_PAD_X);
+      })
+    );
+    return { isHeader, cells };
+  });
+  const colWidths = tableColumnWidths(
+    rows.map((row) => row.cells),
+    tableWidth,
+    numCols
+  );
+  const colX = colWidths.map(
+    (_, colIdx) => colWidths.slice(0, colIdx).reduce((sum, width) => sum + width, x0)
+  );
+  for (const { isHeader, cells } of rows) {
+    const rowCellHeights = cells.map((spans, colIdx) => {
+      const lines = wrapSpans(spans, colWidths[colIdx] - 2 * TABLE_CELL_PAD_X);
       return lines.length * ctx.fontSize * LINE_HEIGHT_FACTOR + 2 * TABLE_CELL_PAD_Y;
     });
     const rowHeight = Math.max(
@@ -1174,7 +1221,8 @@ function renderTable(node, ctx, extraIndent) {
       });
     }
     for (let colIdx = 0; colIdx < numCols; colIdx++) {
-      const cellX = x0 + colIdx * colWidth;
+      const cellX = colX[colIdx];
+      const colWidth = colWidths[colIdx];
       ctx.page.drawRectangle({
         x: cellX,
         y: ctx.y - rowHeight,
@@ -1183,13 +1231,8 @@ function renderTable(node, ctx, extraIndent) {
         borderColor: rgb(COLOR_TABLE_BORDER.r, COLOR_TABLE_BORDER.g, COLOR_TABLE_BORDER.b),
         borderWidth: TABLE_BORDER_WIDTH
       });
-      if (colIdx < row.children.length) {
-        const cellNode = row.children[colIdx];
-        const spans = flattenInlines(cellNode.children, ctx, {
-          bold: isHeader,
-          italic: false,
-          code: false
-        });
+      const spans = cells[colIdx];
+      if (spans) {
         const savedY = ctx.y;
         ctx.y = ctx.y - TABLE_CELL_PAD_Y;
         const cellAvailableWidth = colWidth - 2 * TABLE_CELL_PAD_X;
@@ -1215,6 +1258,44 @@ function renderTable(node, ctx, extraIndent) {
   }
   ctx.y -= PARAGRAPH_SPACING;
 }
+function tableColumnWidths(rows, tableWidth, numCols) {
+  const pad = 2 * TABLE_CELL_PAD_X;
+  const minimum = new Array(numCols).fill(pad);
+  const preferred = new Array(numCols).fill(pad);
+  for (const cells of rows) {
+    cells.forEach((spans, colIdx) => {
+      let longestLine = 0;
+      let longestWord = 0;
+      let line = 0;
+      for (const span of spans) {
+        if (span.text === "\n") {
+          line = 0;
+          continue;
+        }
+        for (const word of splitIntoWords(span.text)) {
+          const width = drawnWidth(span.font, word, span.fontSize);
+          line += width;
+          if (word.trim().length > 0) longestWord = Math.max(longestWord, width);
+        }
+        longestLine = Math.max(longestLine, line);
+      }
+      minimum[colIdx] = Math.max(minimum[colIdx], longestWord + pad);
+      preferred[colIdx] = Math.max(preferred[colIdx], longestLine + pad);
+    });
+  }
+  const share = tableWidth / numCols;
+  const floor = minimum.map((width) => Math.min(width, share));
+  const wish = preferred.map((width, colIdx) => Math.max(width, floor[colIdx]));
+  const sum = (values) => values.reduce((total, value) => total + value, 0);
+  const wished = sum(wish);
+  if (wished <= tableWidth) return wish.map((width) => width / wished * tableWidth);
+  const spare = tableWidth - sum(floor);
+  const wants = wish.map((width, colIdx) => width - floor[colIdx]);
+  const wanted = sum(wants);
+  return floor.map(
+    (width, colIdx) => wanted > 0 ? width + spare * wants[colIdx] / wanted : width + spare / numCols
+  );
+}
 function renderThematicBreak(ctx, extraIndent) {
   ctx.y -= PARAGRAPH_SPACING;
   ensureSpace(ctx, 10);

package/dist/docx/index.js

@@ -4,10 +4,10 @@ import {
   docxToDoc,
   docxToMarkdownDoc,
   markdownDocToDocx
-} from "../chunk-BIERSSDF.js";
-import "../chunk-CJUOAGGS.js";
+} from "../chunk-KROH7FNT.js";
+import "../chunk-DHAK4PII.js";
 import "../chunk-VCZUXQUL.js";
-import "../chunk-AVOZAKGP.js";
+import "../chunk-BZ5SHMOY.js";
 import "../chunk-ILCJ3WFD.js";
 import "../chunk-JU2RHXUB.js";
 import "../chunk-S5PCVMKU.js";

package/dist/epub/index.js

@@ -1,10 +1,10 @@
 import {
   docToEpub,
   markdownDocToEpub
-} from "../chunk-GESOAVYN.js";
+} from "../chunk-SVYTUYZM.js";
 import "../chunk-WMIDXNGR.js";
 import "../chunk-VCZUXQUL.js";
-import "../chunk-AVOZAKGP.js";
+import "../chunk-BZ5SHMOY.js";
 import "../chunk-JU2RHXUB.js";
 import "../chunk-BXWNU4T5.js";
 import "../chunk-LBXBRFBZ.js";

package/dist/index.js

@@ -9,7 +9,7 @@ import {
 import {
   docToEpub,
   markdownDocToEpub
-} from "./chunk-GESOAVYN.js";
+} from "./chunk-SVYTUYZM.js";
 import {
   BUILTIN_FORMAT_IDS
 } from "./chunk-2P5HJAQL.js";
@@ -28,7 +28,7 @@ import {
   renderOutsideInDocument,
   withOutsideInMarkdownEditing,
   withOutsideInMetadata
-} from "./chunk-AOY3UPE3.js";
+} from "./chunk-IYNDHW2C.js";
 import {
   DEFAULT_CONVERSION_LIMITS,
   convert,
@@ -37,7 +37,7 @@ import {
   defaultRegistry,
   prepareConversion,
   resolveConversionLimits
-} from "./chunk-X6WFQDZD.js";
+} from "./chunk-ALDCFHZT.js";
 import {
   ConversionError
 } from "./chunk-KXOZMWBS.js";
@@ -52,23 +52,23 @@ import {
   docxToDoc,
   docxToMarkdownDoc,
   markdownDocToDocx
-} from "./chunk-BIERSSDF.js";
+} from "./chunk-KROH7FNT.js";
 import {
   docToPptx,
   markdownDocToPptx,
   pptxToDoc,
   pptxToMarkdownDoc
-} from "./chunk-QRDJOYT3.js";
+} from "./chunk-QU4EVC62.js";
 import "./chunk-6N2J7C2B.js";
 import "./chunk-WMIDXNGR.js";
-import "./chunk-CJUOAGGS.js";
+import "./chunk-DHAK4PII.js";
 import "./chunk-VCZUXQUL.js";
 import {
   docToXlsx,
   markdownDocToXlsx,
   xlsxToDoc
-} from "./chunk-ZULJFSA4.js";
-import "./chunk-AVOZAKGP.js";
+} from "./chunk-4A3EHQJP.js";
+import "./chunk-BZ5SHMOY.js";
 import {
   gridToTables,
   xlsxToMarkdownDoc,
@@ -94,7 +94,7 @@ import {
   markdownDocToPdf,
   pdfToDoc,
   pdfToMarkdownDoc
-} from "./chunk-GLIUMDYD.js";
+} from "./chunk-TMOXUUTG.js";
 import "./chunk-34KECNZO.js";
 import "./chunk-BXWNU4T5.js";
 import "./chunk-IIQYS2YH.js";

package/dist/outside-in/index.js

@@ -6,8 +6,8 @@ import {
   renderOutsideInDocument,
   withOutsideInMarkdownEditing,
   withOutsideInMetadata
-} from "../chunk-AOY3UPE3.js";
-import "../chunk-X6WFQDZD.js";
+} from "../chunk-IYNDHW2C.js";
+import "../chunk-ALDCFHZT.js";
 import "../chunk-KXOZMWBS.js";
 import {
   OUTSIDE_IN_FORMAT_IDS,

package/dist/pdf/index.js

@@ -5,7 +5,7 @@ import {
   pdfToContainer,
   pdfToDoc,
   pdfToMarkdownDoc
-} from "../chunk-GLIUMDYD.js";
+} from "../chunk-TMOXUUTG.js";
 import "../chunk-34KECNZO.js";
 import "../chunk-BXWNU4T5.js";
 import "../chunk-IIQYS2YH.js";

package/dist/pptx/index.js

@@ -4,15 +4,15 @@ import {
   pptxToContainer,
   pptxToDoc,
   pptxToMarkdownDoc
-} from "../chunk-QRDJOYT3.js";
+} from "../chunk-QU4EVC62.js";
 import {
   analyzePptxLayouts,
   inspectPptxLayouts
 } from "../chunk-6N2J7C2B.js";
 import "../chunk-WMIDXNGR.js";
-import "../chunk-CJUOAGGS.js";
+import "../chunk-DHAK4PII.js";
 import "../chunk-VCZUXQUL.js";
-import "../chunk-AVOZAKGP.js";
+import "../chunk-BZ5SHMOY.js";
 import "../chunk-ILCJ3WFD.js";
 import "../chunk-JU2RHXUB.js";
 import "../chunk-S5PCVMKU.js";

package/dist/registry/index.js

@@ -9,7 +9,7 @@ import {
   defaultRegistry,
   prepareConversion,
   resolveConversionLimits
-} from "../chunk-X6WFQDZD.js";
+} from "../chunk-ALDCFHZT.js";
 import {
   ConversionError
 } from "../chunk-KXOZMWBS.js";

package/dist/xlsx/index.js

@@ -4,8 +4,8 @@ import {
   markdownDocToXlsx,
   patchXlsxCellValues,
   xlsxToDoc
-} from "../chunk-ZULJFSA4.js";
-import "../chunk-AVOZAKGP.js";
+} from "../chunk-4A3EHQJP.js";
+import "../chunk-BZ5SHMOY.js";
 import {
   formatCellRef,
   gridToTables,

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@bendyline/squisq-formats",
-  "version": "2.8.5",
+  "version": "2.8.7",
   "description": "Document format converters — DOCX, PDF, OOXML import/export",
   "license": "MIT",
   "author": "Bendyline",
@@ -125,7 +125,7 @@
     "typecheck": "tsc --noEmit"
   },
   "dependencies": {
-    "@bendyline/squisq": "2.13.4",
+    "@bendyline/squisq": "2.13.6",
     "@pdf-lib/fontkit": "1.1.1",
     "@pdf-lib/upng": "1.0.1",
     "@xmldom/xmldom": "0.9.12",