officeparser 7.0.3 → 7.2.0

package/dist/utils/xmlUtils.d.ts

@@ -144,3 +144,20 @@ export declare const parseOfficeMetadata: (xmlContent: string) => OfficeMetadata
  * ```
  */
 export declare const parseOOXMLCustomProperties: (xmlContent: string) => Record<string, string | number | boolean | Date>;
+/**
+ * Parses OOXML application properties from `docProps/app.xml`.
+ *
+ * Application properties contain document statistics and application settings.
+ *
+ * @param xmlContent - Raw XML string from `docProps/app.xml`
+ * @returns A record of property name -> typed value
+ */
+export declare const parseOOXMLAppProperties: (xmlContent: string) => Record<string, string | number | boolean>;
+/**
+ * Decodes XML entities (standard named entities, decimal, and hexadecimal entities) in a string.
+ * Useful when parsing content with regular expressions instead of a full DOM parser.
+ *
+ * @param text - The XML-encoded string
+ * @returns The decoded string
+ */
+export declare const decodeXmlEntities: (text: string) => string;

package/dist/utils/xmlUtils.js

@@ -11,7 +11,7 @@
  * @module xmlUtils
  */
 Object.defineProperty(exports, "__esModule", { value: true });
-exports.parseOOXMLCustomProperties = exports.parseOfficeMetadata = exports.getDirectChildren = exports.getAttribute = exports.getFirstElementByTagName = exports.getRawContent = exports.getSourceSubstring = exports.serializeXml = exports.getElementsByTagName = exports.parseXmlString = exports.isElement = void 0;
+exports.decodeXmlEntities = exports.parseOOXMLAppProperties = exports.parseOOXMLCustomProperties = exports.parseOfficeMetadata = exports.getDirectChildren = exports.getAttribute = exports.getFirstElementByTagName = exports.getRawContent = exports.getSourceSubstring = exports.serializeXml = exports.getElementsByTagName = exports.parseXmlString = exports.isElement = void 0;
 const xmldom_1 = require("@xmldom/xmldom");
 const dateUtils_js_1 = require("./dateUtils.js");
 /**
@@ -221,6 +221,13 @@ const parseOfficeMetadata = (xmlContent) => {
     // Check for OOXML Core Properties
     const coreProperties = (0, exports.getElementsByTagName)(xml, "cp:coreProperties")[0];
     if (coreProperties) {
+        metadata.nativeProperties = {};
+        for (let i = 0; i < coreProperties.childNodes.length; i++) {
+            const child = coreProperties.childNodes[i];
+            if ((0, exports.isElement)(child)) {
+                metadata.nativeProperties[child.tagName] = child.textContent;
+            }
+        }
         // Step 3: Extract title (Dublin Core element)
         const title = (0, exports.getElementsByTagName)(coreProperties, "dc:title")[0];
         if (title && title.textContent)
@@ -248,11 +255,21 @@ const parseOfficeMetadata = (xmlContent) => {
         const subject = (0, exports.getElementsByTagName)(coreProperties, "dc:subject")[0];
         if (subject && subject.textContent)
             metadata.subject = subject.textContent;
+        const keywords = (0, exports.getElementsByTagName)(coreProperties, "cp:keywords")[0];
+        if (keywords && keywords.textContent)
+            metadata.keywords = keywords.textContent;
         return metadata;
     }
     // Check for ODF Meta
     const officeMeta = (0, exports.getElementsByTagName)(xml, "office:meta")[0];
     if (officeMeta) {
+        metadata.nativeProperties = {};
+        for (let i = 0; i < officeMeta.childNodes.length; i++) {
+            const child = officeMeta.childNodes[i];
+            if ((0, exports.isElement)(child)) {
+                metadata.nativeProperties[child.tagName] = child.textContent;
+            }
+        }
         const title = (0, exports.getElementsByTagName)(officeMeta, "dc:title")[0];
         if (title && title.textContent)
             metadata.title = title.textContent;
@@ -265,6 +282,10 @@ const parseOfficeMetadata = (xmlContent) => {
         const subject = (0, exports.getElementsByTagName)(officeMeta, "dc:subject")[0];
         if (subject && subject.textContent)
             metadata.subject = subject.textContent;
+        const keywordElements = (0, exports.getElementsByTagName)(officeMeta, "meta:keyword");
+        if (keywordElements.length > 0) {
+            metadata.keywords = keywordElements.map(k => k.textContent).filter(Boolean).join(', ');
+        }
         const created = (0, exports.getElementsByTagName)(officeMeta, "meta:creation-date")[0];
         if (created && created.textContent)
             metadata.created = (0, dateUtils_js_1.parseOfficeDate)(created.textContent);
@@ -375,3 +396,66 @@ const parseOOXMLCustomProperties = (xmlContent) => {
     return result;
 };
 exports.parseOOXMLCustomProperties = parseOOXMLCustomProperties;
+/**
+ * Parses OOXML application properties from `docProps/app.xml`.
+ *
+ * Application properties contain document statistics and application settings.
+ *
+ * @param xmlContent - Raw XML string from `docProps/app.xml`
+ * @returns A record of property name -> typed value
+ */
+const parseOOXMLAppProperties = (xmlContent) => {
+    const xml = (0, exports.parseXmlString)(xmlContent);
+    const result = {};
+    const appProperties = (0, exports.getElementsByTagName)(xml, "Properties")[0];
+    if (appProperties) {
+        for (let i = 0; i < appProperties.childNodes.length; i++) {
+            const child = appProperties.childNodes[i];
+            if ((0, exports.isElement)(child)) {
+                const text = child.textContent || '';
+                if (text.toLowerCase() === 'true')
+                    result[child.tagName] = true;
+                else if (text.toLowerCase() === 'false')
+                    result[child.tagName] = false;
+                else if (!isNaN(Number(text)) && text.trim() !== '')
+                    result[child.tagName] = Number(text);
+                else
+                    result[child.tagName] = text;
+            }
+        }
+    }
+    return result;
+};
+exports.parseOOXMLAppProperties = parseOOXMLAppProperties;
+/**
+ * Decodes XML entities (standard named entities, decimal, and hexadecimal entities) in a string.
+ * Useful when parsing content with regular expressions instead of a full DOM parser.
+ *
+ * @param text - The XML-encoded string
+ * @returns The decoded string
+ */
+const decodeXmlEntities = (text) => {
+    return text.replace(/&([^;]+);/g, (match, entity) => {
+        if (entity.startsWith('#')) {
+            if (entity[1] === 'x' || entity[1] === 'X') {
+                const hex = entity.slice(2);
+                const code = parseInt(hex, 16);
+                return !isNaN(code) ? String.fromCodePoint(code) : match;
+            }
+            else {
+                const dec = entity.slice(1);
+                const code = parseInt(dec, 10);
+                return !isNaN(code) ? String.fromCodePoint(code) : match;
+            }
+        }
+        switch (entity) {
+            case 'amp': return '&';
+            case 'lt': return '<';
+            case 'gt': return '>';
+            case 'quot': return '"';
+            case 'apos': return "'";
+            default: return match;
+        }
+    });
+};
+exports.decodeXmlEntities = decodeXmlEntities;

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "officeparser",
-  "version": "7.0.3",
+  "version": "7.2.0",
   "description": "A robust, strictly-typed Node.js and Browser library for parsing office files (.docx, .pptx, .xlsx, .odt, .odp, .ods, .pdf, .rtf, .csv, .md, .html) and generating high-fidelity outputs in Markdown, HTML, CSV, RTF, and RAG-focused chunks.",
   "funding": "https://github.com/sponsors/harshankur",
   "main": "dist/index.js",
@@ -29,7 +29,7 @@
     "build:browser:types": "dts-bundle-generator --no-check -o dist/officeparser.browser.d.ts src/index.ts",
     "build:browser": "node build_browser.js && npm run sync:docs",
     "sync:versions": "node scripts/sync-pdfjs-versions.js",
-    "sync:docs": "mkdir -p docs/dist && cp dist/officeparser.browser.iife.js docs/dist/ && cp dist/officeparser.browser.mjs docs/dist/",
+    "sync:docs": "mkdir -p docs/dist && cp dist/officeparser.browser.iife.js docs/dist/ && cp dist/officeparser.browser.mjs docs/dist/ && mkdir -p docs/test/files && cp test/files/* docs/test/files/",
     "lint": "eslint src",
     "test": "npm run lint && npm run test:clean && npm run build && npm run test:license && npm run test:artifacts && npm run test:parser && npm run test:generator",
     "test:baseline": "npm run test:parser:baseline && npm run test:generator:baseline",
@@ -38,6 +38,7 @@
     "test:generator": "npx tsx test/generator/testOfficeGenerator.ts",
     "test:generator:baseline": "npx tsx test/generator/testOfficeGenerator.ts baseline",
     "test:artifacts": "npx tsx test/testShippingArtifacts.ts",
+    "test:visualizer": "node test/testVisualizer.js",
     "test:license": "npm run sbom && node scripts/validate-licenses.js",
     "test:clean": "rm -rf test/results test/generator/results test/generator/output test/parser/results test/parser/output",
     "clean": "rm -rf dist && npm run test:clean",