officeparser 7.1.0 → 7.2.0

package/dist/utils/xmlUtils.js

@@ -11,7 +11,7 @@
  * @module xmlUtils
  */
 Object.defineProperty(exports, "__esModule", { value: true });
-exports.decodeXmlEntities = 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);
@@ -375,6 +396,37 @@ 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.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "officeparser",
-  "version": "7.1.0",
+  "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",