formatferry 1.0.21 → 1.0.23

package/dist/index.js

@@ -19720,7 +19720,7 @@ var InputSanitizer = class {
       ],
       // Include data-page-marker for PDF page break markers (Req 3.5)
       // Include type and checked for task list checkbox support
-      ALLOWED_ATTR: ["href", "src", "alt", "title", "style", "class", "data-page-marker", "type", "checked", "data-math"]
+      ALLOWED_ATTR: ["href", "src", "alt", "title", "style", "class", "data-page-marker", "data-docx-table-kind", "type", "checked", "data-math"]
     };
     switch (this.config.level) {
       case "strict":
@@ -20729,6 +20729,22 @@ ${plainText}
     });
     return lines.join("\n");
   }
+  flattenTables(html) {
+    if (typeof DOMParser === "undefined") return html;
+    const parser = new DOMParser();
+    const doc = parser.parseFromString(`<div>${html}</div>`, "text/html");
+    Array.from(doc.querySelectorAll("table")).forEach((table) => {
+      const replacement = doc.createElement("div");
+      const lines = this.tableToPlainText(table).split("\n").filter(Boolean);
+      lines.forEach((line) => {
+        const paragraph = doc.createElement("p");
+        paragraph.textContent = line;
+        replacement.appendChild(paragraph);
+      });
+      table.replaceWith(replacement);
+    });
+    return doc.body.firstElementChild?.innerHTML ?? html;
+  }
   addGoogleDocsRules(service) {
     service.addRule("googleDocsBold", {
       filter: (node) => {
@@ -20860,10 +20876,7 @@ ${plainText}
         let headerProcessed = false;
         rows.forEach((row, index) => {
           const cells = Array.from(row.querySelectorAll("th, td"));
-          const cellContents = cells.map((cell) => {
-            const cellContent = cell.textContent || "";
-            return cellContent.trim().replace(/\|/g, "\\|");
-          });
+          const cellContents = cells.map((cell) => this.escapeTableCell(this.renderTableCell(cell)));
           if (cellContents.length > 0) {
             markdown += "| " + cellContents.join(" | ") + " |\n";
             if (!headerProcessed && (row.querySelector("th") || index === 0)) {
@@ -20877,6 +20890,73 @@ ${plainText}
       }
     });
   }
+  renderTableCell(cell) {
+    const blocks = [];
+    let inlineBuffer = "";
+    cell.childNodes.forEach((child) => {
+      if (child.nodeType === Node.ELEMENT_NODE && this.isTableCellBlock(child)) {
+        if (inlineBuffer.trim()) {
+          blocks.push(inlineBuffer.trim());
+          inlineBuffer = "";
+        }
+        blocks.push(this.renderTableBlock(child));
+        return;
+      }
+      inlineBuffer += this.renderInlineNode(child);
+    });
+    if (inlineBuffer.trim()) blocks.push(inlineBuffer.trim());
+    if (blocks.length === 0) return (cell.textContent || "").trim();
+    return blocks.filter(Boolean).join("<br>");
+  }
+  isTableCellBlock(element) {
+    return ["P", "DIV", "UL", "OL", "H1", "H2", "H3", "H4", "H5", "H6", "BLOCKQUOTE", "PRE"].includes(element.nodeName);
+  }
+  renderTableBlock(element) {
+    if (element.nodeName === "UL" || element.nodeName === "OL") {
+      return Array.from(element.children).filter((child) => child.nodeName === "LI").map((child) => `- ${this.renderInlineNode(child).trim()}`).join("<br>");
+    }
+    return this.renderInlineNode(element).trim();
+  }
+  renderInlineNode(node) {
+    if (node.nodeType === Node.TEXT_NODE) return node.textContent || "";
+    if (node.nodeType !== Node.ELEMENT_NODE) return "";
+    const element = node;
+    const content = Array.from(element.childNodes).map((child) => this.renderInlineNode(child)).join("");
+    const trimmed = content.trim();
+    switch (element.nodeName) {
+      case "BR":
+        return "<br>";
+      case "STRONG":
+      case "B":
+        return trimmed ? `**${trimmed}**` : content;
+      case "EM":
+      case "I":
+        return trimmed ? `_${trimmed}_` : content;
+      case "CODE":
+        return trimmed ? this.renderCodeSpan(trimmed) : content;
+      case "A": {
+        const href = element.getAttribute("href") || "";
+        return href && trimmed ? `[${this.escapeMarkdownLinkText(trimmed)}](<${this.escapeMarkdownLinkHref(href)}>)` : content;
+      }
+      default:
+        return content;
+    }
+  }
+  renderCodeSpan(content) {
+    if (!content.includes("`")) return `\`${content}\``;
+    const runs = content.match(/`+/g) || ["`"];
+    const delimiter = "`".repeat(Math.max(...runs.map((run) => run.length)) + 1);
+    return `${delimiter} ${content} ${delimiter}`;
+  }
+  escapeMarkdownLinkText(text) {
+    return text.replace(/\\/g, "\\\\").replace(/\[/g, "\\[").replace(/\]/g, "\\]");
+  }
+  escapeMarkdownLinkHref(href) {
+    return href.replace(/</g, "%3C").replace(/>/g, "%3E");
+  }
+  escapeTableCell(content) {
+    return content.replace(/\|/g, "\\|").replace(/\s*<br>\s*/g, "<br>").trim();
+  }
   addListRules(service) {
     service.addRule("enhancedList", {
       filter: ["ul", "ol"],
@@ -21025,7 +21105,8 @@ ${plainText}
         service = this.githubService;
         break;
     }
-    let markdown = service.turndown(sanitizedHtml);
+    const htmlForConversion = options.preserveTables ? sanitizedHtml : this.flattenTables(sanitizedHtml);
+    let markdown = service.turndown(htmlForConversion);
     markdown = this.applyFormatPostProcessing(markdown, options.format);
     const endTime = performance.now();
     const metadata = {
@@ -23089,7 +23170,7 @@ var PdfAdapter = class {
       const hasMinimalTextWithImages = totalTextLength < MIN_TEXT_CHARS_THRESHOLD && hasAnyImages;
       if (hasNoText || hasMinimalTextWithImages) {
         throw new PdfParseError(
-          "This PDF contains only images. Text extraction requires a text layer. OCR is not supported.",
+          "PDF contains only images \u2014 no text layer available.",
           "IMAGE_ONLY"
         );
       }
@@ -25419,8 +25500,8 @@ import { readFileSync as readFileSync4 } from "fs";
 import { resolve, dirname as dirname3 } from "path";
 import { fileURLToPath } from "url";
 var getVersion = () => {
-  if ("1.0.21") {
-    return "1.0.21";
+  if ("1.0.23") {
+    return "1.0.23";
   }
   try {
     const __dirname = dirname3(fileURLToPath(import.meta.url));

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "formatferry",
-  "version": "1.0.21",
+  "version": "1.0.23",
   "description": "FormatFerry CLI — Convert HTML, DOCX, PDF, XLSX, CSV, JSON, XML, PPTX to Markdown",
   "keywords": [
     "markdown",