officeparser 4.1.0 → 4.1.2

package/README.md

@@ -13,6 +13,7 @@ A Node.js library to parse text out of any office file.
 
 
 #### Update
+* 2024/05/06 - Replaced pdf parsing support from pdf-parse library to natively building it using pdf.js library from Mozilla by analyzing its output. Added pdfjs-dist build as a local library.
 * 2023/11/25 - Fixed error catching when an error occurs within the parsing of a file, especially after decompressing it. Also fixed the problem with parallel parsing of files as we were using only timestamp in file names.
 * 2023/10/24 - Revamped content parsing code. Fixed order of content in files, especially in word files where table information would always land up at the end of the text. Added config object as argument for parseOffice which can be used to set new line delimiter and multiple other configurations. Added support for parsing pdf files using the popular npm library pdf-parse. Removed support for individual file parsing functions.
 * 2023/04/26 - Added support for file buffers as argument for filepath for parseOffice and parseOfficeAsync

package/officeParser.js

@@ -126,6 +126,7 @@ function parsePowerPoint(filepath, callback, config) {
     // Files regex that hold our content of interest
     const allFilesRegex = /ppt\/(notesSlides|slides)\/(notesSlide|slide)\d+.xml/g;
     const slidesRegex   = /ppt\/slides\/slide\d+.xml/g;
+    const slideNumberRegex = /lide(\d+)\.xml/;
 
     /** The decompress location which contains the filename in it */
     const decompressLocation = `${config.tempFilesLocation}/${filepath.split("/").pop()}`;
@@ -134,6 +135,17 @@ function parsePowerPoint(filepath, callback, config) {
         { filter: x => x.path.match(config.ignoreNotes ? slidesRegex : allFilesRegex) }
     )
     .then(files => {
+        // Sort files by slide number and their notes (if any).
+        files.sort((a, b) => {
+            const matchedANumber = parseInt(a.path.match(slideNumberRegex)?.at(1), 10);
+            const matchedBNumber = parseInt(b.path.match(slideNumberRegex)?.at(1), 10);
+
+            const aNumber = isNaN(matchedANumber) ? Infinity : matchedANumber;
+            const bNumber = isNaN(matchedBNumber) ? Infinity : matchedBNumber;
+
+            return aNumber - bNumber || Number(a.path.includes('notes')) - Number(b.path.includes('notes'));
+        });
+
         // Verify if atleast the slides xml files exist in the extracted files list.
         if (files.length == 0 || !files.map(file => file.path).some(filename => filename.match(slidesRegex)))
             throw ERRORMSG.fileCorrupted(filepath);
@@ -218,18 +230,45 @@ function parseExcel(filepath, callback, config) {
     })
     // ********************************** excel xml files explanation ***************************************
     // Structure of xmlContent of an excel file is a bit complex.
-    // We have a sharedStrings.xml file which has strings inside t tags
+    // We usually have a sharedStrings.xml file which has strings inside t tags
+    // However, this file is not necessary to be present. It is sometimes absent if the file has no shared strings indices represented in v nodes.
     // Each sheet has an individual sheet xml file which has numbers in v tags (probably value) inside c tags (probably cell)
     // Each value of v tag is to be used as it is if the "t" attribute (probably type) of c tag is not "s" (probably shared string)
     // If the "t" attribute of c tag is "s", then we use the value to select value from sharedStrings array with the value as its index.
+    // However, if the "t" attribute of c tag is "inlineStr", strings can be inline inside "is"(probably inside String) > "t".
+    // We extract either the inline strings or use the value to get numbers of text from shared strings.
     // Drawing files contain all text for each drawing and have text nodes in a:t and paragraph nodes in a:p.
     // ******************************************************************************************************
     .then(xmlContentFilesObject => {
         /** Store all the text content to respond */
         let responseText = [];
 
-        /** Find text nodes with t tags in sharedStrings xml file */
-        const sharedStringsXmlTNodesList = parseString(xmlContentFilesObject.sharedStringsFile).getElementsByTagName("t");
+        /** Function to check if the given c node is a valid inline string node. */
+        function isValidInlineStringCNode(cNode) {
+            // Initial check to see if the passed node is a cNode
+            if (cNode.tagName.toLowerCase() != 'c')
+                return false;
+            if (cNode.getAttribute("t") != 'inlineStr')
+                return false;
+            const childNodesNamedIs = cNode.getElementsByTagName('is');
+            if (childNodesNamedIs.length != 1)
+                return false;
+            const childNodesNamedT = childNodesNamedIs[0].getElementsByTagName('t');
+            if (childNodesNamedT.length != 1)
+                return false;
+            return childNodesNamedT[0].childNodes[0] && childNodesNamedT[0].childNodes[0].nodeValue != '';
+        }
+
+        /** Function to check if the given c node has a valid v node */
+        function hasValidVNodeInCNode(cNode) {
+            return cNode.getElementsByTagName("v")[0]
+                && cNode.getElementsByTagName("v")[0].childNodes[0]
+                && cNode.getElementsByTagName("v")[0].childNodes[0].nodeValue != ''
+        }
+
+        /** Find text nodes with t tags in sharedStrings xml file. If the sharedStringsFile is not present, we return an empty array. */
+        const sharedStringsXmlTNodesList = xmlContentFilesObject.sharedStringsFile != undefined ? parseString(xmlContentFilesObject.sharedStringsFile).getElementsByTagName("t")
+                                                                                                : [];
         /** Create shared string array. This will be used as a map to get strings from within sheet files. */
         const sharedStrings = Array.from(sharedStringsXmlTNodesList)
                                 .map(tNode => tNode.childNodes[0]?.nodeValue ?? '');
@@ -238,25 +277,33 @@ function parseExcel(filepath, callback, config) {
         xmlContentFilesObject.sheetFiles.forEach(sheetXmlContent => {
             /** Find text nodes with c tags in sharedStrings xml file */
             const sheetsXmlCNodesList = parseString(sheetXmlContent).getElementsByTagName("c");
-            // Traverse through the nodes list and fill responseText with either the number value in its v node or find a mapped string from sharedStrings.
+            // Traverse through the nodes list and fill responseText with either the number value in its v node or find a mapped string from sharedStrings or an inline string.
             responseText.push(
                 Array.from(sheetsXmlCNodesList)
-                    // Filter c nodes than do not have any valid v nodes
-                    .filter(cNode => cNode.getElementsByTagName("v")[0]
-                                        && cNode.getElementsByTagName("v")[0].childNodes[0]
-                                        && cNode.getElementsByTagName("v")[0].childNodes[0].nodeValue)
+                    // Filter out invalid c nodes
+                    .filter(cNode => isValidInlineStringCNode(cNode) || hasValidVNodeInCNode(cNode))
                     .map(cNode => {
-                        /** Flag whether this node's value represents a string index */
-                        const isString = cNode.getAttribute("t") == "s";
-                        /** Find value nodes represented by v tags */
-                        const value = cNode.getElementsByTagName("v")[0].childNodes[0].nodeValue;
-                        // Validate text
-                        if (isString && value >= sharedStrings.length)
-                            throw ERRORMSG.fileCorrupted(filepath);
-
-                        return isString
-                                ? sharedStrings[value]
-                                : value;
+                        // Processing if this is a valid inline string c node.
+                        if (isValidInlineStringCNode(cNode))
+                            return cNode.getElementsByTagName('is')[0].getElementsByTagName('t')[0].childNodes[0].nodeValue;
+
+                        // Processing if this c node has a valid v node.
+                        if (hasValidVNodeInCNode(cNode)) {
+                            /** Flag whether this node's value represents an index in the shared string array */
+                            const isIndexInSharedStrings = cNode.getAttribute("t") == "s";
+                            /** Find value nodes represented by v tags */
+                            const value = cNode.getElementsByTagName("v")[0].childNodes[0].nodeValue;
+                            // Validate text
+                            if (isIndexInSharedStrings && value >= sharedStrings.length)
+                                throw ERRORMSG.fileCorrupted(filepath);
+
+                            return isIndexInSharedStrings
+                                    ? sharedStrings[value]
+                                    : value;
+                        }
+                        // TODO: Add debug asserts for if we reach here which would mean we are filtering more items than we are processing.
+                        // Not the case now but it could happen and it is better to be safe.
+                        return '';
                     })
                     // Join each cell text within a sheet with a space.
                     .join(config.newlineDelimiter ?? "\n")
@@ -454,6 +501,7 @@ function parsePdf(filepath, callback, config) {
         const responseText = textContentArray
                                 .map(textContent => textContent.items)      // Get all the items
                                 .flat()                                     // Flatten all the items object
+                                .filter(item => item.str != '')             // Ignore the empty string items.
                                 .reduce((a, v) => (
                                     {
                                         text: a.text + (v.transform[5] != a.transform5 ? (config.newlineDelimiter ?? "\n") : '') + v.str,
@@ -555,7 +603,7 @@ function parseOffice(file, callback, config = {}) {
                     break;
 
                 default:
-                    throw ERRORMSG.extensionUnsupported(extension);
+                    internalCallback(undefined, ERRORMSG.extensionUnsupported(extension));  // Call the internalCallback function which removes the temp files if required.
             }
 
             /** Internal callback function that calls the user's callback function passed in argument and removes the temp files if required */
@@ -633,7 +681,7 @@ module.exports.parseOfficeAsync = parseOfficeAsync;
 
 
 // Run this library on CLI
-if ((process.argv[0].split('/').pop() == "node" || process.argv[0].split('/').pop() == "npx") && (process.argv[1].split('/').pop() == "officeParser.js" || process.argv[1].split('/').pop() == "officeparser")) {
+if ((process.argv[0].split('/').pop() == "node" || process.argv[0].split('/').pop() == "npx") && (process.argv[1].split('/').pop() == "officeParser.js" || process.argv[1].split('/').pop().toLowerCase() == "officeparser")) {
     if (process.argv.length == 2) {
         // continue
     }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "officeparser",
-  "version": "4.1.0",
+  "version": "4.1.2",
   "description": "A Node.js library to parse text out of any office file. Currently supports docx, pptx, xlsx, odt, odp, ods, pdf files.",
   "main": "officeParser.js",
   "files": [