html-to-org 0.1.0 → 0.2.0

package/dist/chunk-ASXHHGOB.cjs

@@ -0,0 +1,294 @@
+"use strict";Object.defineProperty(exports, "__esModule", {value: true}); function _optionalChain(ops) { let lastAccessLHS = undefined; let value = ops[0]; let i = 1; while (i < ops.length) { const op = ops[i]; const fn = ops[i + 1]; i += 2; if ((op === 'optionalAccess' || op === 'optionalCall') && value == null) { return undefined; } if (op === 'access' || op === 'optionalAccess') { lastAccessLHS = value; value = fn(value); } else if (op === 'call' || op === 'optionalCall') { value = fn((...args) => value.call(lastAccessLHS, ...args)); lastAccessLHS = undefined; } } return value; }// src/convert.ts
+function domToOrg(node, baseUrl = "") {
+  const ctx = { baseUrl, listDepth: 0, orderedIndex: [], indentWidth: 0 };
+  const raw = convertNode(node, ctx);
+  return raw.replace(/\n{3,}/g, "\n\n").trim();
+}
+function convertNode(node, ctx) {
+  if (node.nodeType === 3) {
+    return collapseWhitespace(node.textContent || "");
+  }
+  if (node.nodeType !== 1) return "";
+  const tag = (node.tagName || "").toLowerCase();
+  if (tag === "script" || tag === "style" || tag === "noscript") return "";
+  switch (tag) {
+    case "h1":
+    case "h2":
+    case "h3":
+    case "h4":
+    case "h5":
+    case "h6":
+      return convertHeading(node, tag, ctx);
+    case "p":
+      return convertParagraph(node, ctx);
+    case "strong":
+    case "b":
+      return wrapInline("*", node, ctx);
+    case "em":
+    case "i":
+      return wrapInline("/", node, ctx);
+    case "u":
+    case "ins":
+      return wrapInline("_", node, ctx);
+    case "s":
+    case "del":
+    case "strike":
+      return wrapInline("+", node, ctx);
+    case "code":
+      return wrapInline("=", node, ctx);
+    case "mark":
+      return wrapInline("=", node, ctx);
+    case "sup":
+      return `^{${convertChildren(node, ctx)}}`;
+    case "sub":
+      return `_{${convertChildren(node, ctx)}}`;
+    case "a":
+      return convertLink(node, ctx);
+    case "img":
+      return convertImage(node, ctx);
+    case "ul":
+      return convertList(node, false, ctx);
+    case "ol":
+      return convertList(node, true, ctx);
+    case "li":
+      return convertChildren(node, ctx);
+    case "pre":
+      return convertPre(node, ctx);
+    case "blockquote":
+      return convertBlockquote(node, ctx);
+    case "table":
+      return convertTable(node, ctx);
+    case "hr":
+      return "\n\n-----\n\n";
+    case "br":
+      return "\n";
+    default:
+      return convertChildren(node, ctx);
+  }
+}
+function convertChildren(node, ctx) {
+  let out = "";
+  for (const child of node.childNodes) {
+    if (child.nodeType === 3) {
+      const text = child.textContent || "";
+      if (!text.trim()) {
+        const prev = child.previousSibling;
+        const next = child.nextSibling;
+        if (isBlockElement(prev) || isBlockElement(next)) continue;
+      }
+    }
+    out += convertNode(child, ctx);
+  }
+  return out;
+}
+var BLOCK_TAGS = /* @__PURE__ */ new Set([
+  "p",
+  "div",
+  "h1",
+  "h2",
+  "h3",
+  "h4",
+  "h5",
+  "h6",
+  "ul",
+  "ol",
+  "li",
+  "pre",
+  "blockquote",
+  "table",
+  "hr",
+  "section",
+  "article",
+  "main",
+  "header",
+  "footer",
+  "nav",
+  "aside",
+  "figure",
+  "figcaption",
+  "details",
+  "summary"
+]);
+function isBlockElement(node) {
+  if (!node || node.nodeType !== 1) return false;
+  return BLOCK_TAGS.has((node.tagName || "").toLowerCase());
+}
+function convertHeading(node, tag, ctx) {
+  const level = parseInt(tag[1], 10);
+  const text = convertChildren(node, ctx).trim();
+  return `
+
+${"*".repeat(level)} ${text}
+
+`;
+}
+function convertParagraph(node, ctx) {
+  const text = convertChildren(node, ctx).trim();
+  if (!text) return "";
+  return `
+
+${text}
+
+`;
+}
+function convertBlockquote(node, ctx) {
+  const inner = convertChildren(node, ctx).trim();
+  return `
+
+#+BEGIN_QUOTE
+${inner}
+#+END_QUOTE
+
+`;
+}
+function convertPre(node, _ctx) {
+  const codeChild = _optionalChain([node, 'access', _ => _.querySelector, 'optionalCall', _2 => _2("code")]);
+  if (codeChild) {
+    const lang = detectLanguage(codeChild);
+    const code = (codeChild.textContent || "").replace(/\n$/, "");
+    const langSuffix = lang ? ` ${lang}` : "";
+    return `
+
+#+BEGIN_SRC${langSuffix}
+${code}
+#+END_SRC
+
+`;
+  }
+  const text = (node.textContent || "").replace(/\n$/, "");
+  return `
+
+#+BEGIN_EXAMPLE
+${text}
+#+END_EXAMPLE
+
+`;
+}
+function detectLanguage(codeNode) {
+  const cls = _optionalChain([codeNode, 'access', _3 => _3.getAttribute, 'optionalCall', _4 => _4("class")]) || "";
+  const match = cls.match(/(?:^|\s)language-(\S+)/);
+  return match ? match[1] : "";
+}
+function convertList(node, ordered, ctx) {
+  const items = [];
+  let counter = 1;
+  const indent = ctx.listDepth > 0 ? " ".repeat(ctx.indentWidth || 2) : "";
+  const prefixWidth = ordered ? 3 : 2;
+  for (const child of node.childNodes) {
+    if (child.nodeType !== 1 || (child.tagName || "").toLowerCase() !== "li") continue;
+    const currentPrefix = ordered ? `${counter}. ` : "- ";
+    let textParts = [];
+    let nestedLists = [];
+    for (const liChild of child.childNodes) {
+      const liChildTag = (liChild.tagName || "").toLowerCase();
+      if (liChildTag === "ul" || liChildTag === "ol") {
+        const nestedCtx = {
+          ...ctx,
+          listDepth: ctx.listDepth + 1,
+          indentWidth: (ctx.indentWidth || 0) + prefixWidth
+        };
+        nestedLists.push(convertList(liChild, liChildTag === "ol", nestedCtx));
+      } else {
+        textParts.push(convertNode(liChild, ctx));
+      }
+    }
+    const text = textParts.join("").trim();
+    let item = `${indent}${currentPrefix}${text}`;
+    if (nestedLists.length > 0) {
+      item += "\n" + nestedLists.join("\n");
+    }
+    items.push(item);
+    counter++;
+  }
+  const result = items.join("\n");
+  return ctx.listDepth === 0 ? `
+
+${result}
+
+` : result;
+}
+function wrapInline(marker, node, ctx) {
+  const inner = convertChildren(node, ctx);
+  if (!inner.trim()) return inner;
+  return `${marker}${inner}${marker}`;
+}
+function convertLink(node, ctx) {
+  const href = resolveUrl(_optionalChain([node, 'access', _5 => _5.getAttribute, 'optionalCall', _6 => _6("href")]) || "", ctx.baseUrl);
+  const text = convertChildren(node, ctx).trim();
+  if (!href) return text;
+  if (!text || text === href) return `[[${href}]]`;
+  return `[[${href}][${text}]]`;
+}
+function convertImage(node, ctx) {
+  const src = resolveUrl(_optionalChain([node, 'access', _7 => _7.getAttribute, 'optionalCall', _8 => _8("src")]) || "", ctx.baseUrl);
+  if (!src) return "";
+  return `[[${src}]]`;
+}
+function convertTable(node, ctx) {
+  const rows = [];
+  let headerRowCount = 0;
+  const thead = _optionalChain([node, 'access', _9 => _9.querySelector, 'optionalCall', _10 => _10("thead")]);
+  if (thead) {
+    for (const tr of _optionalChain([thead, 'access', _11 => _11.querySelectorAll, 'optionalCall', _12 => _12("tr")]) || []) {
+      rows.push(extractRowCells(tr, ctx));
+      headerRowCount++;
+    }
+  }
+  const tbody = _optionalChain([node, 'access', _13 => _13.querySelector, 'optionalCall', _14 => _14("tbody")]);
+  const bodyContainer = tbody || node;
+  for (const tr of _optionalChain([bodyContainer, 'access', _15 => _15.querySelectorAll, 'optionalCall', _16 => _16("tr")]) || []) {
+    if (thead && tr.parentNode === thead) continue;
+    rows.push(extractRowCells(tr, ctx));
+  }
+  if (rows.length === 0) return "";
+  const colCount = Math.max(...rows.map((r) => r.length));
+  const colWidths = new Array(colCount).fill(0);
+  for (const row of rows) {
+    for (let i = 0; i < colCount; i++) {
+      colWidths[i] = Math.max(colWidths[i], (row[i] || "").length);
+    }
+  }
+  const formatRow = (row) => {
+    const cells = [];
+    for (let i = 0; i < colCount; i++) {
+      cells.push((row[i] || "").padEnd(colWidths[i]));
+    }
+    return "| " + cells.join(" | ") + " |";
+  };
+  const separatorRow = "|" + colWidths.map((w) => `-${"-".repeat(w)}-`).join("+") + "|";
+  const lines = [];
+  for (let i = 0; i < rows.length; i++) {
+    lines.push(formatRow(rows[i]));
+    if (i === headerRowCount - 1 && headerRowCount > 0) {
+      lines.push(separatorRow);
+    }
+  }
+  return "\n\n" + lines.join("\n") + "\n\n";
+}
+function extractRowCells(tr, ctx) {
+  const cells = [];
+  for (const cell of tr.childNodes) {
+    const cellTag = (cell.tagName || "").toLowerCase();
+    if (cellTag === "td" || cellTag === "th") {
+      cells.push(convertChildren(cell, ctx).trim());
+    }
+  }
+  return cells;
+}
+function collapseWhitespace(text) {
+  return text.replace(/\s+/g, " ");
+}
+function resolveUrl(url, baseUrl) {
+  if (!url) return "";
+  if (/^https?:\/\//.test(url) || url.startsWith("mailto:")) return url;
+  if (!baseUrl) return url;
+  try {
+    return new URL(url, baseUrl).href;
+  } catch (e) {
+    return url;
+  }
+}
+
+
+
+exports.domToOrg = domToOrg;

package/dist/chunk-ZZ4VEN6R.js

@@ -0,0 +1,294 @@
+// src/convert.ts
+function domToOrg(node, baseUrl = "") {
+  const ctx = { baseUrl, listDepth: 0, orderedIndex: [], indentWidth: 0 };
+  const raw = convertNode(node, ctx);
+  return raw.replace(/\n{3,}/g, "\n\n").trim();
+}
+function convertNode(node, ctx) {
+  if (node.nodeType === 3) {
+    return collapseWhitespace(node.textContent || "");
+  }
+  if (node.nodeType !== 1) return "";
+  const tag = (node.tagName || "").toLowerCase();
+  if (tag === "script" || tag === "style" || tag === "noscript") return "";
+  switch (tag) {
+    case "h1":
+    case "h2":
+    case "h3":
+    case "h4":
+    case "h5":
+    case "h6":
+      return convertHeading(node, tag, ctx);
+    case "p":
+      return convertParagraph(node, ctx);
+    case "strong":
+    case "b":
+      return wrapInline("*", node, ctx);
+    case "em":
+    case "i":
+      return wrapInline("/", node, ctx);
+    case "u":
+    case "ins":
+      return wrapInline("_", node, ctx);
+    case "s":
+    case "del":
+    case "strike":
+      return wrapInline("+", node, ctx);
+    case "code":
+      return wrapInline("=", node, ctx);
+    case "mark":
+      return wrapInline("=", node, ctx);
+    case "sup":
+      return `^{${convertChildren(node, ctx)}}`;
+    case "sub":
+      return `_{${convertChildren(node, ctx)}}`;
+    case "a":
+      return convertLink(node, ctx);
+    case "img":
+      return convertImage(node, ctx);
+    case "ul":
+      return convertList(node, false, ctx);
+    case "ol":
+      return convertList(node, true, ctx);
+    case "li":
+      return convertChildren(node, ctx);
+    case "pre":
+      return convertPre(node, ctx);
+    case "blockquote":
+      return convertBlockquote(node, ctx);
+    case "table":
+      return convertTable(node, ctx);
+    case "hr":
+      return "\n\n-----\n\n";
+    case "br":
+      return "\n";
+    default:
+      return convertChildren(node, ctx);
+  }
+}
+function convertChildren(node, ctx) {
+  let out = "";
+  for (const child of node.childNodes) {
+    if (child.nodeType === 3) {
+      const text = child.textContent || "";
+      if (!text.trim()) {
+        const prev = child.previousSibling;
+        const next = child.nextSibling;
+        if (isBlockElement(prev) || isBlockElement(next)) continue;
+      }
+    }
+    out += convertNode(child, ctx);
+  }
+  return out;
+}
+var BLOCK_TAGS = /* @__PURE__ */ new Set([
+  "p",
+  "div",
+  "h1",
+  "h2",
+  "h3",
+  "h4",
+  "h5",
+  "h6",
+  "ul",
+  "ol",
+  "li",
+  "pre",
+  "blockquote",
+  "table",
+  "hr",
+  "section",
+  "article",
+  "main",
+  "header",
+  "footer",
+  "nav",
+  "aside",
+  "figure",
+  "figcaption",
+  "details",
+  "summary"
+]);
+function isBlockElement(node) {
+  if (!node || node.nodeType !== 1) return false;
+  return BLOCK_TAGS.has((node.tagName || "").toLowerCase());
+}
+function convertHeading(node, tag, ctx) {
+  const level = parseInt(tag[1], 10);
+  const text = convertChildren(node, ctx).trim();
+  return `
+
+${"*".repeat(level)} ${text}
+
+`;
+}
+function convertParagraph(node, ctx) {
+  const text = convertChildren(node, ctx).trim();
+  if (!text) return "";
+  return `
+
+${text}
+
+`;
+}
+function convertBlockquote(node, ctx) {
+  const inner = convertChildren(node, ctx).trim();
+  return `
+
+#+BEGIN_QUOTE
+${inner}
+#+END_QUOTE
+
+`;
+}
+function convertPre(node, _ctx) {
+  const codeChild = node.querySelector?.("code");
+  if (codeChild) {
+    const lang = detectLanguage(codeChild);
+    const code = (codeChild.textContent || "").replace(/\n$/, "");
+    const langSuffix = lang ? ` ${lang}` : "";
+    return `
+
+#+BEGIN_SRC${langSuffix}
+${code}
+#+END_SRC
+
+`;
+  }
+  const text = (node.textContent || "").replace(/\n$/, "");
+  return `
+
+#+BEGIN_EXAMPLE
+${text}
+#+END_EXAMPLE
+
+`;
+}
+function detectLanguage(codeNode) {
+  const cls = codeNode.getAttribute?.("class") || "";
+  const match = cls.match(/(?:^|\s)language-(\S+)/);
+  return match ? match[1] : "";
+}
+function convertList(node, ordered, ctx) {
+  const items = [];
+  let counter = 1;
+  const indent = ctx.listDepth > 0 ? " ".repeat(ctx.indentWidth || 2) : "";
+  const prefixWidth = ordered ? 3 : 2;
+  for (const child of node.childNodes) {
+    if (child.nodeType !== 1 || (child.tagName || "").toLowerCase() !== "li") continue;
+    const currentPrefix = ordered ? `${counter}. ` : "- ";
+    let textParts = [];
+    let nestedLists = [];
+    for (const liChild of child.childNodes) {
+      const liChildTag = (liChild.tagName || "").toLowerCase();
+      if (liChildTag === "ul" || liChildTag === "ol") {
+        const nestedCtx = {
+          ...ctx,
+          listDepth: ctx.listDepth + 1,
+          indentWidth: (ctx.indentWidth || 0) + prefixWidth
+        };
+        nestedLists.push(convertList(liChild, liChildTag === "ol", nestedCtx));
+      } else {
+        textParts.push(convertNode(liChild, ctx));
+      }
+    }
+    const text = textParts.join("").trim();
+    let item = `${indent}${currentPrefix}${text}`;
+    if (nestedLists.length > 0) {
+      item += "\n" + nestedLists.join("\n");
+    }
+    items.push(item);
+    counter++;
+  }
+  const result = items.join("\n");
+  return ctx.listDepth === 0 ? `
+
+${result}
+
+` : result;
+}
+function wrapInline(marker, node, ctx) {
+  const inner = convertChildren(node, ctx);
+  if (!inner.trim()) return inner;
+  return `${marker}${inner}${marker}`;
+}
+function convertLink(node, ctx) {
+  const href = resolveUrl(node.getAttribute?.("href") || "", ctx.baseUrl);
+  const text = convertChildren(node, ctx).trim();
+  if (!href) return text;
+  if (!text || text === href) return `[[${href}]]`;
+  return `[[${href}][${text}]]`;
+}
+function convertImage(node, ctx) {
+  const src = resolveUrl(node.getAttribute?.("src") || "", ctx.baseUrl);
+  if (!src) return "";
+  return `[[${src}]]`;
+}
+function convertTable(node, ctx) {
+  const rows = [];
+  let headerRowCount = 0;
+  const thead = node.querySelector?.("thead");
+  if (thead) {
+    for (const tr of thead.querySelectorAll?.("tr") || []) {
+      rows.push(extractRowCells(tr, ctx));
+      headerRowCount++;
+    }
+  }
+  const tbody = node.querySelector?.("tbody");
+  const bodyContainer = tbody || node;
+  for (const tr of bodyContainer.querySelectorAll?.("tr") || []) {
+    if (thead && tr.parentNode === thead) continue;
+    rows.push(extractRowCells(tr, ctx));
+  }
+  if (rows.length === 0) return "";
+  const colCount = Math.max(...rows.map((r) => r.length));
+  const colWidths = new Array(colCount).fill(0);
+  for (const row of rows) {
+    for (let i = 0; i < colCount; i++) {
+      colWidths[i] = Math.max(colWidths[i], (row[i] || "").length);
+    }
+  }
+  const formatRow = (row) => {
+    const cells = [];
+    for (let i = 0; i < colCount; i++) {
+      cells.push((row[i] || "").padEnd(colWidths[i]));
+    }
+    return "| " + cells.join(" | ") + " |";
+  };
+  const separatorRow = "|" + colWidths.map((w) => `-${"-".repeat(w)}-`).join("+") + "|";
+  const lines = [];
+  for (let i = 0; i < rows.length; i++) {
+    lines.push(formatRow(rows[i]));
+    if (i === headerRowCount - 1 && headerRowCount > 0) {
+      lines.push(separatorRow);
+    }
+  }
+  return "\n\n" + lines.join("\n") + "\n\n";
+}
+function extractRowCells(tr, ctx) {
+  const cells = [];
+  for (const cell of tr.childNodes) {
+    const cellTag = (cell.tagName || "").toLowerCase();
+    if (cellTag === "td" || cellTag === "th") {
+      cells.push(convertChildren(cell, ctx).trim());
+    }
+  }
+  return cells;
+}
+function collapseWhitespace(text) {
+  return text.replace(/\s+/g, " ");
+}
+function resolveUrl(url, baseUrl) {
+  if (!url) return "";
+  if (/^https?:\/\//.test(url) || url.startsWith("mailto:")) return url;
+  if (!baseUrl) return url;
+  try {
+    return new URL(url, baseUrl).href;
+  } catch {
+    return url;
+  }
+}
+
+export {
+  domToOrg
+};

package/dist/dom.cjs

@@ -0,0 +1,6 @@
+"use strict";Object.defineProperty(exports, "__esModule", {value: true});
+
+var _chunkASXHHGOBcjs = require('./chunk-ASXHHGOB.cjs');
+
+
+exports.domToOrg = _chunkASXHHGOBcjs.domToOrg;

package/dist/dom.d.cts

@@ -0,0 +1,10 @@
+/**
+ * Convert a DOM node to Org-mode format.
+ *
+ * @param node - A DOM node (Element, Document, or DocumentFragment)
+ * @param baseUrl - Base URL for resolving relative links
+ * @returns Org-mode formatted string
+ */
+declare function domToOrg(node: any, baseUrl?: string): string;
+
+export { domToOrg };

package/dist/dom.d.ts

@@ -0,0 +1,10 @@
+/**
+ * Convert a DOM node to Org-mode format.
+ *
+ * @param node - A DOM node (Element, Document, or DocumentFragment)
+ * @param baseUrl - Base URL for resolving relative links
+ * @returns Org-mode formatted string
+ */
+declare function domToOrg(node: any, baseUrl?: string): string;
+
+export { domToOrg };

package/dist/dom.js

@@ -0,0 +1,6 @@
+import {
+  domToOrg
+} from "./chunk-ZZ4VEN6R.js";
+export {
+  domToOrg
+};

package/dist/index.cjs

@@ -1,351 +1,15 @@
-"use strict";
-var __defProp = Object.defineProperty;
-var __getOwnPropDesc = Object.getOwnPropertyDescriptor;
-var __getOwnPropNames = Object.getOwnPropertyNames;
-var __hasOwnProp = Object.prototype.hasOwnProperty;
-var __export = (target, all) => {
-  for (var name in all)
-    __defProp(target, name, { get: all[name], enumerable: true });
-};
-var __copyProps = (to, from, except, desc) => {
-  if (from && typeof from === "object" || typeof from === "function") {
-    for (let key of __getOwnPropNames(from))
-      if (!__hasOwnProp.call(to, key) && key !== except)
-        __defProp(to, key, { get: () => from[key], enumerable: !(desc = __getOwnPropDesc(from, key)) || desc.enumerable });
-  }
-  return to;
-};
-var __toCommonJS = (mod) => __copyProps(__defProp({}, "__esModule", { value: true }), mod);
+"use strict";Object.defineProperty(exports, "__esModule", {value: true});
+
+var _chunkASXHHGOBcjs = require('./chunk-ASXHHGOB.cjs');
 
 // src/index.ts
-var index_exports = {};
-__export(index_exports, {
-  htmlToOrg: () => htmlToOrg
-});
-module.exports = __toCommonJS(index_exports);
-var import_linkedom = require("linkedom");
+var _linkedom = require('linkedom');
 function htmlToOrg(html, baseUrl = "") {
   if (!html || !html.trim()) return "";
-  const { document } = (0, import_linkedom.parseHTML)(`<!DOCTYPE html><html><body>${html}</body></html>`);
-  const body = document.body;
-  const ctx = { baseUrl, listDepth: 0, orderedIndex: [], indentWidth: 0 };
-  const raw = convertNode(body, ctx);
-  return raw.replace(/\n{3,}/g, "\n\n").trim();
-}
-function convertNode(node, ctx) {
-  if (node.nodeType === 3) {
-    return collapseWhitespace(node.textContent || "");
-  }
-  if (node.nodeType !== 1) return "";
-  const tag = (node.tagName || "").toLowerCase();
-  if (tag === "script" || tag === "style" || tag === "noscript") return "";
-  switch (tag) {
-    case "h1":
-    case "h2":
-    case "h3":
-    case "h4":
-    case "h5":
-    case "h6":
-      return convertHeading(node, tag, ctx);
-    case "p":
-      return convertParagraph(node, ctx);
-    case "strong":
-    case "b":
-      return wrapInline("*", node, ctx);
-    case "em":
-    case "i":
-      return wrapInline("/", node, ctx);
-    case "u":
-    case "ins":
-      return wrapInline("_", node, ctx);
-    case "s":
-    case "del":
-    case "strike":
-      return wrapInline("+", node, ctx);
-    case "code":
-      return wrapInline("=", node, ctx);
-    case "mark":
-      return wrapInline("=", node, ctx);
-    case "sup":
-      return `^{${convertChildren(node, ctx)}}`;
-    case "sub":
-      return `_{${convertChildren(node, ctx)}}`;
-    case "a":
-      return convertLink(node, ctx);
-    case "img":
-      return convertImage(node, ctx);
-    case "ul":
-      return convertList(node, false, ctx);
-    case "ol":
-      return convertList(node, true, ctx);
-    case "li":
-      return convertChildren(node, ctx);
-    case "pre":
-      return convertPre(node, ctx);
-    case "blockquote":
-      return convertBlockquote(node, ctx);
-    case "table":
-      return convertTable(node, ctx);
-    case "hr":
-      return "\n\n-----\n\n";
-    case "br":
-      return "\n";
-    // Transparent wrappers — just recurse
-    case "div":
-    case "section":
-    case "article":
-    case "main":
-    case "header":
-    case "footer":
-    case "nav":
-    case "aside":
-    case "figure":
-    case "figcaption":
-    case "details":
-    case "summary":
-    case "span":
-    case "small":
-    case "time":
-    case "abbr":
-    case "thead":
-    case "tbody":
-    case "tfoot":
-    case "html":
-    case "body":
-      return convertChildren(node, ctx);
-    default:
-      return convertChildren(node, ctx);
-  }
-}
-function convertChildren(node, ctx) {
-  let out = "";
-  for (const child of node.childNodes) {
-    if (child.nodeType === 3) {
-      const text = child.textContent || "";
-      if (!text.trim()) {
-        const prev = child.previousSibling;
-        const next = child.nextSibling;
-        if (isBlockElement(prev) || isBlockElement(next)) {
-          continue;
-        }
-      }
-    }
-    out += convertNode(child, ctx);
-  }
-  return out;
-}
-var BLOCK_TAGS = /* @__PURE__ */ new Set([
-  "p",
-  "div",
-  "h1",
-  "h2",
-  "h3",
-  "h4",
-  "h5",
-  "h6",
-  "ul",
-  "ol",
-  "li",
-  "pre",
-  "blockquote",
-  "table",
-  "hr",
-  "section",
-  "article",
-  "main",
-  "header",
-  "footer",
-  "nav",
-  "aside",
-  "figure",
-  "figcaption",
-  "details",
-  "summary"
-]);
-function isBlockElement(node) {
-  if (!node || node.nodeType !== 1) return false;
-  return BLOCK_TAGS.has((node.tagName || "").toLowerCase());
-}
-function convertHeading(node, tag, ctx) {
-  const level = parseInt(tag[1], 10);
-  const stars = "*".repeat(level);
-  const text = convertChildren(node, ctx).trim();
-  return `
-
-${stars} ${text}
-
-`;
-}
-function convertParagraph(node, ctx) {
-  const text = convertChildren(node, ctx).trim();
-  if (!text) return "";
-  return `
-
-${text}
-
-`;
-}
-function convertBlockquote(node, ctx) {
-  const inner = convertChildren(node, ctx).trim();
-  return `
-
-#+BEGIN_QUOTE
-${inner}
-#+END_QUOTE
-
-`;
+  const { document } = _linkedom.parseHTML.call(void 0, `<!DOCTYPE html><html><body>${html}</body></html>`);
+  return _chunkASXHHGOBcjs.domToOrg.call(void 0, document.body, baseUrl);
 }
-function convertPre(node, _ctx) {
-  const codeChild = node.querySelector?.("code");
-  if (codeChild) {
-    const lang = detectLanguage(codeChild);
-    const code = (codeChild.textContent || "").replace(/\n$/, "");
-    const langSuffix = lang ? ` ${lang}` : "";
-    return `
 
-#+BEGIN_SRC${langSuffix}
-${code}
-#+END_SRC
 
-`;
-  }
-  const text = (node.textContent || "").replace(/\n$/, "");
-  return `
 
-#+BEGIN_EXAMPLE
-${text}
-#+END_EXAMPLE
-
-`;
-}
-function detectLanguage(codeNode) {
-  const cls = codeNode.getAttribute?.("class") || "";
-  const match = cls.match(/(?:^|\s)language-(\S+)/);
-  return match ? match[1] : "";
-}
-function convertList(node, ordered, ctx) {
-  const items = [];
-  let counter = 1;
-  const indent = ctx.listDepth > 0 ? " ".repeat(ctx.indentWidth || 2) : "";
-  const prefixWidth = ordered ? 3 : 2;
-  for (const child of node.childNodes) {
-    if (child.nodeType !== 1 || (child.tagName || "").toLowerCase() !== "li") continue;
-    const currentPrefix = ordered ? `${counter}. ` : "- ";
-    let textParts = [];
-    let nestedLists = [];
-    for (const liChild of child.childNodes) {
-      const liChildTag = (liChild.tagName || "").toLowerCase();
-      if (liChildTag === "ul" || liChildTag === "ol") {
-        const nestedCtx = {
-          ...ctx,
-          listDepth: ctx.listDepth + 1,
-          indentWidth: (ctx.indentWidth || 0) + prefixWidth
-        };
-        nestedLists.push(convertList(liChild, liChildTag === "ol", nestedCtx));
-      } else {
-        textParts.push(convertNode(liChild, ctx));
-      }
-    }
-    const text = textParts.join("").trim();
-    let item = `${indent}${currentPrefix}${text}`;
-    if (nestedLists.length > 0) {
-      item += "\n" + nestedLists.join("\n");
-    }
-    items.push(item);
-    counter++;
-  }
-  const result = items.join("\n");
-  return ctx.listDepth === 0 ? `
-
-${result}
-
-` : result;
-}
-function wrapInline(marker, node, ctx) {
-  const inner = convertChildren(node, ctx);
-  if (!inner.trim()) return inner;
-  return `${marker}${inner}${marker}`;
-}
-function convertLink(node, ctx) {
-  const href = resolveUrl(node.getAttribute?.("href") || "", ctx.baseUrl);
-  const text = convertChildren(node, ctx).trim();
-  if (!href) return text;
-  if (!text || text === href) return `[[${href}]]`;
-  return `[[${href}][${text}]]`;
-}
-function convertImage(node, ctx) {
-  const src = resolveUrl(node.getAttribute?.("src") || "", ctx.baseUrl);
-  if (!src) return "";
-  return `[[${src}]]`;
-}
-function convertTable(node, ctx) {
-  const rows = [];
-  let headerRowCount = 0;
-  const thead = node.querySelector?.("thead");
-  if (thead) {
-    for (const tr of thead.querySelectorAll?.("tr") || []) {
-      const cells = extractRowCells(tr, ctx);
-      rows.push(cells);
-      headerRowCount++;
-    }
-  }
-  const tbody = node.querySelector?.("tbody");
-  const bodyContainer = tbody || node;
-  for (const tr of bodyContainer.querySelectorAll?.("tr") || []) {
-    if (thead && tr.parentNode === thead) continue;
-    const cells = extractRowCells(tr, ctx);
-    rows.push(cells);
-  }
-  if (rows.length === 0) return "";
-  const colCount = Math.max(...rows.map((r) => r.length));
-  const colWidths = new Array(colCount).fill(0);
-  for (const row of rows) {
-    for (let i = 0; i < colCount; i++) {
-      colWidths[i] = Math.max(colWidths[i], (row[i] || "").length);
-    }
-  }
-  const formatRow = (row) => {
-    const cells = [];
-    for (let i = 0; i < colCount; i++) {
-      cells.push((row[i] || "").padEnd(colWidths[i]));
-    }
-    return "| " + cells.join(" | ") + " |";
-  };
-  const sepParts = colWidths.map((w) => "-".repeat(w));
-  const separatorRow = "|" + sepParts.map((s) => `-${s}-`).join("+") + "|";
-  const lines = [];
-  for (let i = 0; i < rows.length; i++) {
-    lines.push(formatRow(rows[i]));
-    if (i === headerRowCount - 1 && headerRowCount > 0) {
-      lines.push(separatorRow);
-    }
-  }
-  return "\n\n" + lines.join("\n") + "\n\n";
-}
-function extractRowCells(tr, ctx) {
-  const cells = [];
-  for (const cell of tr.childNodes) {
-    const cellTag = (cell.tagName || "").toLowerCase();
-    if (cellTag === "td" || cellTag === "th") {
-      cells.push(convertChildren(cell, ctx).trim());
-    }
-  }
-  return cells;
-}
-function collapseWhitespace(text) {
-  return text.replace(/\s+/g, " ");
-}
-function resolveUrl(url, baseUrl) {
-  if (!url) return "";
-  if (/^https?:\/\//.test(url) || url.startsWith("mailto:")) return url;
-  if (!baseUrl) return url;
-  try {
-    return new URL(url, baseUrl).href;
-  } catch {
-    return url;
-  }
-}
-// Annotate the CommonJS export names for ESM import in node:
-0 && (module.exports = {
-  htmlToOrg
-});
+exports.domToOrg = _chunkASXHHGOBcjs.domToOrg; exports.htmlToOrg = htmlToOrg;

package/dist/index.d.cts

@@ -1,6 +1,11 @@
+export { domToOrg } from './dom.cjs';
+
 /**
  * Convert HTML to Org-mode format.
  *
+ * Uses linkedom for parsing. In browser environments, import `domToOrg`
+ * from `html-to-org/dom` to avoid bundling linkedom.
+ *
  * @param html - HTML string to convert
  * @param baseUrl - Base URL for resolving relative links
  * @returns Org-mode formatted string

package/dist/index.d.ts

@@ -1,6 +1,11 @@
+export { domToOrg } from './dom.js';
+
 /**
  * Convert HTML to Org-mode format.
  *
+ * Uses linkedom for parsing. In browser environments, import `domToOrg`
+ * from `html-to-org/dom` to avoid bundling linkedom.
+ *
  * @param html - HTML string to convert
  * @param baseUrl - Base URL for resolving relative links
  * @returns Org-mode formatted string

package/dist/index.js

@@ -1,326 +1,15 @@
+import {
+  domToOrg
+} from "./chunk-ZZ4VEN6R.js";
+
 // src/index.ts
 import { parseHTML } from "linkedom";
 function htmlToOrg(html, baseUrl = "") {
   if (!html || !html.trim()) return "";
   const { document } = parseHTML(`<!DOCTYPE html><html><body>${html}</body></html>`);
-  const body = document.body;
-  const ctx = { baseUrl, listDepth: 0, orderedIndex: [], indentWidth: 0 };
-  const raw = convertNode(body, ctx);
-  return raw.replace(/\n{3,}/g, "\n\n").trim();
-}
-function convertNode(node, ctx) {
-  if (node.nodeType === 3) {
-    return collapseWhitespace(node.textContent || "");
-  }
-  if (node.nodeType !== 1) return "";
-  const tag = (node.tagName || "").toLowerCase();
-  if (tag === "script" || tag === "style" || tag === "noscript") return "";
-  switch (tag) {
-    case "h1":
-    case "h2":
-    case "h3":
-    case "h4":
-    case "h5":
-    case "h6":
-      return convertHeading(node, tag, ctx);
-    case "p":
-      return convertParagraph(node, ctx);
-    case "strong":
-    case "b":
-      return wrapInline("*", node, ctx);
-    case "em":
-    case "i":
-      return wrapInline("/", node, ctx);
-    case "u":
-    case "ins":
-      return wrapInline("_", node, ctx);
-    case "s":
-    case "del":
-    case "strike":
-      return wrapInline("+", node, ctx);
-    case "code":
-      return wrapInline("=", node, ctx);
-    case "mark":
-      return wrapInline("=", node, ctx);
-    case "sup":
-      return `^{${convertChildren(node, ctx)}}`;
-    case "sub":
-      return `_{${convertChildren(node, ctx)}}`;
-    case "a":
-      return convertLink(node, ctx);
-    case "img":
-      return convertImage(node, ctx);
-    case "ul":
-      return convertList(node, false, ctx);
-    case "ol":
-      return convertList(node, true, ctx);
-    case "li":
-      return convertChildren(node, ctx);
-    case "pre":
-      return convertPre(node, ctx);
-    case "blockquote":
-      return convertBlockquote(node, ctx);
-    case "table":
-      return convertTable(node, ctx);
-    case "hr":
-      return "\n\n-----\n\n";
-    case "br":
-      return "\n";
-    // Transparent wrappers — just recurse
-    case "div":
-    case "section":
-    case "article":
-    case "main":
-    case "header":
-    case "footer":
-    case "nav":
-    case "aside":
-    case "figure":
-    case "figcaption":
-    case "details":
-    case "summary":
-    case "span":
-    case "small":
-    case "time":
-    case "abbr":
-    case "thead":
-    case "tbody":
-    case "tfoot":
-    case "html":
-    case "body":
-      return convertChildren(node, ctx);
-    default:
-      return convertChildren(node, ctx);
-  }
-}
-function convertChildren(node, ctx) {
-  let out = "";
-  for (const child of node.childNodes) {
-    if (child.nodeType === 3) {
-      const text = child.textContent || "";
-      if (!text.trim()) {
-        const prev = child.previousSibling;
-        const next = child.nextSibling;
-        if (isBlockElement(prev) || isBlockElement(next)) {
-          continue;
-        }
-      }
-    }
-    out += convertNode(child, ctx);
-  }
-  return out;
-}
-var BLOCK_TAGS = /* @__PURE__ */ new Set([
-  "p",
-  "div",
-  "h1",
-  "h2",
-  "h3",
-  "h4",
-  "h5",
-  "h6",
-  "ul",
-  "ol",
-  "li",
-  "pre",
-  "blockquote",
-  "table",
-  "hr",
-  "section",
-  "article",
-  "main",
-  "header",
-  "footer",
-  "nav",
-  "aside",
-  "figure",
-  "figcaption",
-  "details",
-  "summary"
-]);
-function isBlockElement(node) {
-  if (!node || node.nodeType !== 1) return false;
-  return BLOCK_TAGS.has((node.tagName || "").toLowerCase());
-}
-function convertHeading(node, tag, ctx) {
-  const level = parseInt(tag[1], 10);
-  const stars = "*".repeat(level);
-  const text = convertChildren(node, ctx).trim();
-  return `
-
-${stars} ${text}
-
-`;
-}
-function convertParagraph(node, ctx) {
-  const text = convertChildren(node, ctx).trim();
-  if (!text) return "";
-  return `
-
-${text}
-
-`;
-}
-function convertBlockquote(node, ctx) {
-  const inner = convertChildren(node, ctx).trim();
-  return `
-
-#+BEGIN_QUOTE
-${inner}
-#+END_QUOTE
-
-`;
-}
-function convertPre(node, _ctx) {
-  const codeChild = node.querySelector?.("code");
-  if (codeChild) {
-    const lang = detectLanguage(codeChild);
-    const code = (codeChild.textContent || "").replace(/\n$/, "");
-    const langSuffix = lang ? ` ${lang}` : "";
-    return `
-
-#+BEGIN_SRC${langSuffix}
-${code}
-#+END_SRC
-
-`;
-  }
-  const text = (node.textContent || "").replace(/\n$/, "");
-  return `
-
-#+BEGIN_EXAMPLE
-${text}
-#+END_EXAMPLE
-
-`;
-}
-function detectLanguage(codeNode) {
-  const cls = codeNode.getAttribute?.("class") || "";
-  const match = cls.match(/(?:^|\s)language-(\S+)/);
-  return match ? match[1] : "";
-}
-function convertList(node, ordered, ctx) {
-  const items = [];
-  let counter = 1;
-  const indent = ctx.listDepth > 0 ? " ".repeat(ctx.indentWidth || 2) : "";
-  const prefixWidth = ordered ? 3 : 2;
-  for (const child of node.childNodes) {
-    if (child.nodeType !== 1 || (child.tagName || "").toLowerCase() !== "li") continue;
-    const currentPrefix = ordered ? `${counter}. ` : "- ";
-    let textParts = [];
-    let nestedLists = [];
-    for (const liChild of child.childNodes) {
-      const liChildTag = (liChild.tagName || "").toLowerCase();
-      if (liChildTag === "ul" || liChildTag === "ol") {
-        const nestedCtx = {
-          ...ctx,
-          listDepth: ctx.listDepth + 1,
-          indentWidth: (ctx.indentWidth || 0) + prefixWidth
-        };
-        nestedLists.push(convertList(liChild, liChildTag === "ol", nestedCtx));
-      } else {
-        textParts.push(convertNode(liChild, ctx));
-      }
-    }
-    const text = textParts.join("").trim();
-    let item = `${indent}${currentPrefix}${text}`;
-    if (nestedLists.length > 0) {
-      item += "\n" + nestedLists.join("\n");
-    }
-    items.push(item);
-    counter++;
-  }
-  const result = items.join("\n");
-  return ctx.listDepth === 0 ? `
-
-${result}
-
-` : result;
-}
-function wrapInline(marker, node, ctx) {
-  const inner = convertChildren(node, ctx);
-  if (!inner.trim()) return inner;
-  return `${marker}${inner}${marker}`;
-}
-function convertLink(node, ctx) {
-  const href = resolveUrl(node.getAttribute?.("href") || "", ctx.baseUrl);
-  const text = convertChildren(node, ctx).trim();
-  if (!href) return text;
-  if (!text || text === href) return `[[${href}]]`;
-  return `[[${href}][${text}]]`;
-}
-function convertImage(node, ctx) {
-  const src = resolveUrl(node.getAttribute?.("src") || "", ctx.baseUrl);
-  if (!src) return "";
-  return `[[${src}]]`;
-}
-function convertTable(node, ctx) {
-  const rows = [];
-  let headerRowCount = 0;
-  const thead = node.querySelector?.("thead");
-  if (thead) {
-    for (const tr of thead.querySelectorAll?.("tr") || []) {
-      const cells = extractRowCells(tr, ctx);
-      rows.push(cells);
-      headerRowCount++;
-    }
-  }
-  const tbody = node.querySelector?.("tbody");
-  const bodyContainer = tbody || node;
-  for (const tr of bodyContainer.querySelectorAll?.("tr") || []) {
-    if (thead && tr.parentNode === thead) continue;
-    const cells = extractRowCells(tr, ctx);
-    rows.push(cells);
-  }
-  if (rows.length === 0) return "";
-  const colCount = Math.max(...rows.map((r) => r.length));
-  const colWidths = new Array(colCount).fill(0);
-  for (const row of rows) {
-    for (let i = 0; i < colCount; i++) {
-      colWidths[i] = Math.max(colWidths[i], (row[i] || "").length);
-    }
-  }
-  const formatRow = (row) => {
-    const cells = [];
-    for (let i = 0; i < colCount; i++) {
-      cells.push((row[i] || "").padEnd(colWidths[i]));
-    }
-    return "| " + cells.join(" | ") + " |";
-  };
-  const sepParts = colWidths.map((w) => "-".repeat(w));
-  const separatorRow = "|" + sepParts.map((s) => `-${s}-`).join("+") + "|";
-  const lines = [];
-  for (let i = 0; i < rows.length; i++) {
-    lines.push(formatRow(rows[i]));
-    if (i === headerRowCount - 1 && headerRowCount > 0) {
-      lines.push(separatorRow);
-    }
-  }
-  return "\n\n" + lines.join("\n") + "\n\n";
-}
-function extractRowCells(tr, ctx) {
-  const cells = [];
-  for (const cell of tr.childNodes) {
-    const cellTag = (cell.tagName || "").toLowerCase();
-    if (cellTag === "td" || cellTag === "th") {
-      cells.push(convertChildren(cell, ctx).trim());
-    }
-  }
-  return cells;
-}
-function collapseWhitespace(text) {
-  return text.replace(/\s+/g, " ");
-}
-function resolveUrl(url, baseUrl) {
-  if (!url) return "";
-  if (/^https?:\/\//.test(url) || url.startsWith("mailto:")) return url;
-  if (!baseUrl) return url;
-  try {
-    return new URL(url, baseUrl).href;
-  } catch {
-    return url;
-  }
+  return domToOrg(document.body, baseUrl);
 }
 export {
+  domToOrg,
   htmlToOrg
 };

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "html-to-org",
-  "version": "0.1.0",
+  "version": "0.2.0",
   "description": "Convert HTML to Org-mode format",
   "type": "module",
   "main": "./dist/index.cjs",
@@ -16,11 +16,28 @@
         "types": "./dist/index.d.cts",
         "default": "./dist/index.cjs"
       }
+    },
+    "./dom": {
+      "import": {
+        "types": "./dist/dom.d.ts",
+        "default": "./dist/dom.js"
+      },
+      "require": {
+        "types": "./dist/dom.d.cts",
+        "default": "./dist/dom.cjs"
+      }
     }
   },
   "files": [
     "dist"
   ],
+  "scripts": {
+    "build": "tsup",
+    "test": "vitest run",
+    "test:watch": "vitest",
+    "test:coverage": "vitest run --coverage",
+    "prepublishOnly": "pnpm build"
+  },
   "keywords": [
     "html",
     "org-mode",
@@ -36,6 +53,7 @@
     "url": "https://github.com/zzzhizhia/html-to-org.git"
   },
   "homepage": "https://github.com/zzzhizhia/html-to-org",
+  "packageManager": "pnpm@10.33.0",
   "engines": {
     "node": ">=20"
   },
@@ -46,11 +64,5 @@
   },
   "dependencies": {
     "linkedom": "^0.18.12"
-  },
-  "scripts": {
-    "build": "tsup",
-    "test": "vitest run",
-    "test:watch": "vitest",
-    "test:coverage": "vitest run --coverage"
   }
-}
+}