xml-to-html-converter 0.2.0 → 0.3.0

package/dist/index.cjs

@@ -0,0 +1,258 @@
+"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);
+
+// src/index.ts
+var src_exports = {};
+__export(src_exports, {
+  isMalformed: () => isMalformed,
+  render: () => render,
+  scaffold: () => scaffold
+});
+module.exports = __toCommonJS(src_exports);
+
+// src/modules/render/render.ts
+function render(nodes) {
+  return nodes.map(renderNode).join("");
+}
+function renderNode(node) {
+  if (node.role === "textLeaf") return node.raw;
+  if (node.role === "comment") return node.raw;
+  if (node.role === "processingInstruction") return "";
+  if (node.role === "doctype") return "";
+  if (node.role === "closeTag") return "";
+  const tag = node.xmlTag ?? "";
+  const attrs = buildDataAttrs(node);
+  const attrsHtml = attrs ? ` ${attrs}` : "";
+  if (node.role === "selfTag") {
+    return `<div data-tag="${tag}"${attrsHtml}></div>`;
+  }
+  const children = render(node.children ?? []);
+  return `<div data-tag="${tag}"${attrsHtml}>${children}</div>`;
+}
+function buildDataAttrs(node) {
+  if (!node.xmlAttributes || node.xmlAttributes.length === 0) return "";
+  return node.xmlAttributes.map(({ name, value }) => `data-attrs-${name}="${value}"`).join(" ");
+}
+
+// src/modules/scaffold/scaffold.ts
+function parseXmlAttributes(xmlInner) {
+  const attributes = [];
+  let i = 0;
+  const s = xmlInner.trim();
+  while (i < s.length) {
+    while (i < s.length && /\s/.test(s[i])) i++;
+    if (i >= s.length) break;
+    const nameStart = i;
+    while (i < s.length && s[i] !== "=" && !/\s/.test(s[i])) i++;
+    const name = s.slice(nameStart, i).trim();
+    if (!name) break;
+    while (i < s.length && /\s/.test(s[i])) i++;
+    if (s[i] !== "=") break;
+    i++;
+    while (i < s.length && /\s/.test(s[i])) i++;
+    const quote = s[i];
+    if (quote !== '"' && quote !== "'") break;
+    i++;
+    const valueStart = i;
+    while (i < s.length && s[i] !== quote) i++;
+    const value = s.slice(valueStart, i);
+    i++;
+    attributes.push({ name, value });
+  }
+  return attributes.length > 0 ? attributes : void 0;
+}
+var MAX_DEPTH = 500;
+function scaffold(xml) {
+  const counter = { value: 0 };
+  const { xmlNodes } = collectXmlNodes(xml, 0, null, counter, 0);
+  return xmlNodes;
+}
+function collectXmlNodes(xml, position, parentTag, counter, depth) {
+  if (depth > MAX_DEPTH) return { xmlNodes: [], position, closed: false };
+  const xmlNodes = [];
+  while (position < xml.length) {
+    const xmlNodeData = extractXmlNodes(xml, position);
+    if (xmlNodeData.role === "textLeaf" && xmlNodeData.raw.trim() === "") {
+      position = xmlNodeData.end;
+      continue;
+    }
+    if (xmlNodeData.role === "closeTag") {
+      if (xmlNodeData.tag === parentTag)
+        return { xmlNodes, position: xmlNodeData.end, closed: true };
+      xmlNodes.push({
+        role: "closeTag",
+        raw: xmlNodeData.raw,
+        xmlTag: xmlNodeData.tag || void 0,
+        xmlInner: xmlNodeData.xmlInner,
+        globalIndex: counter.value++,
+        localIndex: xmlNodes.length,
+        malformed: true
+      });
+      position = xmlNodeData.end;
+      continue;
+    }
+    if (xmlNodeData.role === "openTag" && !xmlNodeData.malformed) {
+      const globalIndex = counter.value++;
+      const localIndex = xmlNodes.length;
+      const nested = collectXmlNodes(
+        xml,
+        xmlNodeData.end,
+        xmlNodeData.tag,
+        counter,
+        depth + 1
+      );
+      const xmlNode2 = {
+        role: "openTag",
+        raw: xmlNodeData.raw,
+        xmlTag: xmlNodeData.tag || void 0,
+        xmlInner: xmlNodeData.xmlInner,
+        xmlAttributes: xmlNodeData.xmlAttributes,
+        globalIndex,
+        localIndex,
+        children: nested.xmlNodes
+      };
+      if (!nested.closed) xmlNode2.malformed = true;
+      xmlNodes.push(xmlNode2);
+      position = nested.position;
+      continue;
+    }
+    const xmlNode = {
+      role: xmlNodeData.role,
+      raw: xmlNodeData.raw,
+      xmlTag: xmlNodeData.tag || void 0,
+      xmlInner: xmlNodeData.xmlInner,
+      xmlAttributes: xmlNodeData.xmlAttributes,
+      globalIndex: counter.value++,
+      localIndex: xmlNodes.length
+    };
+    if (xmlNodeData.malformed) xmlNode.malformed = true;
+    if (xmlNodeData.role === "openTag") xmlNode.children = [];
+    xmlNodes.push(xmlNode);
+    position = xmlNodeData.end;
+  }
+  return { xmlNodes, position, closed: parentTag === null };
+}
+function findTagClose(xml, position) {
+  let i = position;
+  while (i < xml.length) {
+    const ch = xml[i];
+    if (ch === '"' || ch === "'") {
+      const closeQuote = xml.indexOf(ch, i + 1);
+      i = closeQuote === -1 ? xml.length : closeQuote + 1;
+      continue;
+    }
+    if (ch === ">") return i;
+    i++;
+  }
+  return -1;
+}
+function extractXmlNodes(xml, position) {
+  if (xml[position] !== "<") {
+    const end2 = xml.indexOf("<", position);
+    return {
+      raw: xml.slice(position, end2 === -1 ? xml.length : end2),
+      role: "textLeaf",
+      tag: "",
+      end: end2 === -1 ? xml.length : end2
+    };
+  }
+  if (xml[position + 1] === "?") {
+    const end2 = xml.indexOf("?>", position + 2);
+    return end2 === -1 ? {
+      raw: xml.slice(position),
+      role: "processingInstruction",
+      tag: "",
+      end: xml.length
+    } : {
+      raw: xml.slice(position, end2 + 2),
+      role: "processingInstruction",
+      tag: "",
+      end: end2 + 2
+    };
+  }
+  if (xml[position + 1] === "!" && xml[position + 2] === "[") {
+    const end2 = xml.indexOf("]]>", position + 3);
+    return end2 === -1 ? { raw: xml.slice(position), role: "textLeaf", tag: "", end: xml.length } : {
+      raw: xml.slice(position, end2 + 3),
+      role: "textLeaf",
+      tag: "",
+      end: end2 + 3
+    };
+  }
+  if (xml[position + 1] === "!" && xml[position + 2] === "-" && xml[position + 3] === "-") {
+    const end2 = xml.indexOf("-->", position + 4);
+    return end2 === -1 ? { raw: xml.slice(position), role: "comment", tag: "", end: xml.length } : {
+      raw: xml.slice(position, end2 + 3),
+      role: "comment",
+      tag: "",
+      end: end2 + 3
+    };
+  }
+  if (xml.startsWith("<!DOCTYPE", position)) {
+    const bracketOpen = xml.indexOf("[", position);
+    const firstClose = xml.indexOf(">", position);
+    const hasBracket = bracketOpen !== -1 && bracketOpen < firstClose;
+    if (hasBracket) {
+      const bracketClose = xml.indexOf("]>", bracketOpen);
+      const end3 = bracketClose === -1 ? xml.length : bracketClose + 2;
+      return { raw: xml.slice(position, end3), role: "doctype", tag: "", end: end3 };
+    }
+    const end2 = firstClose === -1 ? xml.length : firstClose + 1;
+    return { raw: xml.slice(position, end2), role: "doctype", tag: "", end: end2 };
+  }
+  const closeAt = findTagClose(xml, position + 1);
+  if (closeAt === -1)
+    return {
+      raw: xml.slice(position),
+      role: "openTag",
+      tag: "",
+      end: xml.length,
+      malformed: true
+    };
+  const raw = xml.slice(position, closeAt + 1);
+  const end = closeAt + 1;
+  const inner = xml.slice(position + 1, closeAt).trim();
+  if (inner.startsWith("/")) {
+    const tag2 = inner.slice(1).trim().split(/\s/)[0] ?? "";
+    const xmlInner2 = inner.slice(1).trim().slice(tag2.length).trim() || void 0;
+    return { raw, role: "closeTag", tag: tag2, xmlInner: xmlInner2, end };
+  }
+  if (inner.endsWith("/")) {
+    const trimmed = inner.slice(0, -1).trim();
+    const tag2 = trimmed.split(/\s/)[0] ?? "";
+    const xmlInner2 = trimmed.slice(tag2.length).trim() || void 0;
+    const xmlAttributes2 = xmlInner2 ? parseXmlAttributes(xmlInner2) : void 0;
+    return { raw, role: "selfTag", tag: tag2, xmlInner: xmlInner2, xmlAttributes: xmlAttributes2, end };
+  }
+  const tag = inner.split(/\s/)[0] ?? "";
+  const xmlInner = inner.slice(tag.length).trim() || void 0;
+  const xmlAttributes = xmlInner ? parseXmlAttributes(xmlInner) : void 0;
+  return { raw, role: "openTag", tag, xmlInner, xmlAttributes, end };
+}
+
+// src/modules/scaffold/types.ts
+function isMalformed(node) {
+  return node.malformed === true;
+}
+// Annotate the CommonJS export names for ESM import in node:
+0 && (module.exports = {
+  isMalformed,
+  render,
+  scaffold
+});

package/dist/index.d.cts

@@ -0,0 +1,26 @@
+interface XmlAttribute {
+    name: string;
+    value: string;
+}
+type XmlNodeRole = "closeTag" | "comment" | "doctype" | "openTag" | "processingInstruction" | "selfTag" | "textLeaf";
+interface XmlNode {
+    role: XmlNodeRole;
+    raw: string;
+    xmlTag?: string;
+    xmlInner?: string;
+    xmlAttributes?: XmlAttribute[];
+    globalIndex: number;
+    localIndex: number;
+    children?: XmlNode[];
+    malformed?: true;
+}
+type MalformedXmlNode = XmlNode & {
+    malformed: true;
+};
+declare function isMalformed(node: XmlNode): node is MalformedXmlNode;
+
+declare function render(nodes: XmlNode[]): string;
+
+declare function scaffold(xml: string): XmlNode[];
+
+export { type MalformedXmlNode, type XmlAttribute, type XmlNode, type XmlNodeRole, isMalformed, render, scaffold };

package/package.json

@@ -1,14 +1,16 @@
 {
   "name": "xml-to-html-converter",
-  "version": "0.2.0",
+  "version": "0.3.0",
   "description": "Zero dependency XML to HTML converter for Node environments",
   "type": "module",
-  "main": "./dist/index.js",
+  "main": "./dist/index.cjs",
+  "module": "./dist/index.js",
   "types": "./dist/index.d.ts",
   "exports": {
     ".": {
       "types": "./dist/index.d.ts",
-      "import": "./dist/index.js"
+      "import": "./dist/index.js",
+      "require": "./dist/index.cjs"
     }
   },
   "files": [