stegdoc 1.0.0

package/src/lib/xml-utils.js

@@ -0,0 +1,115 @@
+/**
+ * Shared XML parsing utilities for DOCX and XLSX handlers
+ * Handles namespace prefix variations (w:, ns0:, ns1:, etc.)
+ */
+
+const AdmZip = require('adm-zip');
+const { XMLParser } = require('fast-xml-parser');
+
+// Reusable XML parser configured to strip namespace prefixes
+const xmlParser = new XMLParser({
+  ignoreAttributes: false,
+  attributeNamePrefix: '@_',
+  removeNSPrefix: true, // Strips ns0:, ns1:, w:, etc.
+});
+
+/**
+ * Parse XML string with namespace-agnostic parser
+ * @param {string} xmlString - XML content to parse
+ * @returns {object} Parsed XML as JavaScript object
+ */
+function parseXml(xmlString) {
+  return xmlParser.parse(xmlString);
+}
+
+/**
+ * Read and parse an XML file from a ZIP archive (DOCX/XLSX)
+ * @param {string|Buffer} zipPath - Path to ZIP file or Buffer
+ * @param {string} entryPath - Path to XML file within ZIP
+ * @returns {object|null} Parsed XML or null if entry not found
+ */
+function parseXmlFromZip(zipPath, entryPath) {
+  const zip = new AdmZip(zipPath);
+  const entry = zip.getEntry(entryPath);
+
+  if (!entry) {
+    return null;
+  }
+
+  const xmlString = entry.getData().toString('utf8');
+  return parseXml(xmlString);
+}
+
+/**
+ * Get raw XML string from a ZIP archive
+ * @param {string|Buffer} zipPath - Path to ZIP file or Buffer
+ * @param {string} entryPath - Path to XML file within ZIP
+ * @returns {string|null} XML string or null if entry not found
+ */
+function getXmlStringFromZip(zipPath, entryPath) {
+  const zip = new AdmZip(zipPath);
+  const entry = zip.getEntry(entryPath);
+
+  if (!entry) {
+    return null;
+  }
+
+  return entry.getData().toString('utf8');
+}
+
+/**
+ * List all entries in a ZIP file
+ * @param {string|Buffer} zipPath - Path to ZIP file or Buffer
+ * @returns {string[]} Array of entry paths
+ */
+function listZipEntries(zipPath) {
+  const zip = new AdmZip(zipPath);
+  return zip.getEntries().map(e => e.entryName);
+}
+
+/**
+ * Ensure value is an array (handles single item vs array in XML parsing)
+ * @param {*} value - Value that might be an array or single item
+ * @returns {Array} Always returns an array
+ */
+function ensureArray(value) {
+  if (value === undefined || value === null) return [];
+  return Array.isArray(value) ? value : [value];
+}
+
+/**
+ * Safely get nested property from object
+ * @param {object} obj - Object to traverse
+ * @param {string} path - Dot-separated path (e.g., 'worksheet.sheetData.row')
+ * @returns {*} Value at path or undefined
+ */
+function getNestedValue(obj, path) {
+  return path.split('.').reduce((current, key) => {
+    return current && current[key] !== undefined ? current[key] : undefined;
+  }, obj);
+}
+
+/**
+ * Extract text content from a parsed XML text node
+ * Handles both simple strings and objects with #text
+ * @param {*} textNode - Text node from parsed XML
+ * @returns {string} Extracted text
+ */
+function extractTextContent(textNode) {
+  if (typeof textNode === 'string') return textNode;
+  if (typeof textNode === 'number') return String(textNode);
+  if (textNode && typeof textNode === 'object') {
+    if (textNode['#text'] !== undefined) return String(textNode['#text']);
+  }
+  return '';
+}
+
+module.exports = {
+  parseXml,
+  parseXmlFromZip,
+  getXmlStringFromZip,
+  listZipEntries,
+  ensureArray,
+  getNestedValue,
+  extractTextContent,
+};