officeparser 6.0.7 → 6.1.1

package/dist/parsers/ExcelParser.d.ts

@@ -21,7 +21,7 @@
  * @module ExcelParser
  * @see https://www.ecma-international.org/publications-and-standards/standards/ecma-376/
  */
-import { OfficeParserAST, OfficeParserConfig } from '../types';
+import { OfficeParserAST, OfficeParserConfig } from '../types.js';
 /**
  * Parses an Excel spreadsheet (.xlsx) and extracts sheets, rows, and cells.
  *

package/dist/parsers/ExcelParser.js

@@ -24,12 +24,12 @@
  */
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.parseExcel = void 0;
-const chartUtils_1 = require("../utils/chartUtils");
-const errorUtils_1 = require("../utils/errorUtils");
-const imageUtils_1 = require("../utils/imageUtils");
-const ocrUtils_1 = require("../utils/ocrUtils");
-const xmlUtils_1 = require("../utils/xmlUtils");
-const zipUtils_1 = require("../utils/zipUtils");
+const chartUtils_js_1 = require("../utils/chartUtils.js");
+const errorUtils_js_1 = require("../utils/errorUtils.js");
+const imageUtils_js_1 = require("../utils/imageUtils.js");
+const ocrUtils_js_1 = require("../utils/ocrUtils.js");
+const xmlUtils_js_1 = require("../utils/xmlUtils.js");
+const zipUtils_js_1 = require("../utils/zipUtils.js");
 /**
  * Parses an Excel spreadsheet (.xlsx) and extracts sheets, rows, and cells.
  *
@@ -44,9 +44,10 @@ const parseExcel = async (buffer, config) => {
     const stringsFilePath = 'xl/sharedStrings.xml';
     const mediaFileRegex = /xl\/media\/.*/;
     const corePropsFileRegex = /docProps\/core\.xml/;
+    const customPropsFileRegex = /docProps\/custom\.xml/;
     const relsRegex = /xl\/worksheets\/_rels\/sheet\d+\.xml\.rels/g;
     const drawingRelsRegex = /xl\/drawings\/_rels\/drawing\d+\.xml\.rels/g;
-    const files = await (0, zipUtils_1.extractFiles)(buffer, (x) => !!x.match(sheetsRegex) ||
+    const files = await (0, zipUtils_js_1.extractFiles)(buffer, (x) => !!x.match(sheetsRegex) ||
         !!x.match(drawingsRegex) ||
         !!x.match(chartsRegex) ||
         x === stringsFilePath ||
@@ -54,47 +55,48 @@ const parseExcel = async (buffer, config) => {
         x === 'xl/workbook.xml' ||
         x === 'xl/_rels/workbook.xml.rels' ||
         !!x.match(corePropsFileRegex) ||
+        !!x.match(customPropsFileRegex) ||
         (!!config.extractAttachments && (!!x.match(mediaFileRegex) || !!x.match(relsRegex) || !!x.match(drawingRelsRegex))));
     const sharedStringsFile = files.find(f => f.path === stringsFilePath);
     // Updated to store structured content (rich text runs) or simple string
     const sharedStrings = [];
     if (sharedStringsFile) {
-        const xml = (0, xmlUtils_1.parseXmlString)(sharedStringsFile.content.toString());
-        const siNodes = (0, xmlUtils_1.getElementsByTagName)(xml, "si");
+        const xml = (0, xmlUtils_js_1.parseXmlString)(sharedStringsFile.content.toString());
+        const siNodes = (0, xmlUtils_js_1.getElementsByTagName)(xml, "si");
         for (const si of siNodes) {
-            const runNodes = (0, xmlUtils_1.getElementsByTagName)(si, "r");
+            const runNodes = (0, xmlUtils_js_1.getElementsByTagName)(si, "r");
             if (runNodes.length > 0) {
                 // Rich text with runs
                 const runs = [];
                 for (const run of runNodes) {
-                    const tNode = (0, xmlUtils_1.getElementsByTagName)(run, "t")[0];
+                    const tNode = (0, xmlUtils_js_1.getElementsByTagName)(run, "t")[0];
                     if (tNode) {
                         const text = tNode.textContent || '';
                         // Extract run formatting 
-                        const rPr = (0, xmlUtils_1.getElementsByTagName)(run, "rPr")[0];
+                        const rPr = (0, xmlUtils_js_1.getElementsByTagName)(run, "rPr")[0];
                         const formatting = {};
                         if (rPr) {
-                            if ((0, xmlUtils_1.getElementsByTagName)(rPr, "b").length > 0)
+                            if ((0, xmlUtils_js_1.getElementsByTagName)(rPr, "b").length > 0)
                                 formatting.bold = true;
-                            if ((0, xmlUtils_1.getElementsByTagName)(rPr, "i").length > 0)
+                            if ((0, xmlUtils_js_1.getElementsByTagName)(rPr, "i").length > 0)
                                 formatting.italic = true;
-                            if ((0, xmlUtils_1.getElementsByTagName)(rPr, "u").length > 0)
+                            if ((0, xmlUtils_js_1.getElementsByTagName)(rPr, "u").length > 0)
                                 formatting.underline = true;
-                            if ((0, xmlUtils_1.getElementsByTagName)(rPr, "strike").length > 0)
+                            if ((0, xmlUtils_js_1.getElementsByTagName)(rPr, "strike").length > 0)
                                 formatting.strikethrough = true;
-                            const sz = (0, xmlUtils_1.getElementsByTagName)(rPr, "sz")[0];
+                            const sz = (0, xmlUtils_js_1.getElementsByTagName)(rPr, "sz")[0];
                             if (sz)
                                 formatting.size = sz.getAttribute("val") + 'pt';
-                            const color = (0, xmlUtils_1.getElementsByTagName)(rPr, "color")[0];
+                            const color = (0, xmlUtils_js_1.getElementsByTagName)(rPr, "color")[0];
                             if (color) {
                                 const rgb = color.getAttribute("rgb");
                                 if (rgb)
                                     formatting.color = '#' + rgb.substring(2);
                             }
-                            const rFont = (0, xmlUtils_1.getElementsByTagName)(rPr, "rFont")[0];
+                            const rFont = (0, xmlUtils_js_1.getElementsByTagName)(rPr, "rFont")[0];
                             if (rFont)
                                 formatting.font = rFont.getAttribute("val") || undefined;
-                            const vertAlign = (0, xmlUtils_1.getElementsByTagName)(rPr, "vertAlign")[0];
+                            const vertAlign = (0, xmlUtils_js_1.getElementsByTagName)(rPr, "vertAlign")[0];
                             if (vertAlign) {
                                 const val = vertAlign.getAttribute("val");
                                 if (val === "subscript")
@@ -114,7 +116,7 @@ const parseExcel = async (buffer, config) => {
             }
             else {
                 // Simple text case
-                const tNodes = (0, xmlUtils_1.getElementsByTagName)(si, "t");
+                const tNodes = (0, xmlUtils_js_1.getElementsByTagName)(si, "t");
                 let text = '';
                 for (const t of tNodes) {
                     text += t.textContent || '';
@@ -127,41 +129,41 @@ const parseExcel = async (buffer, config) => {
     const stylesFile = files.find(f => f.path === 'xl/styles.xml');
     const cellFormatMap = {};
     if (stylesFile) {
-        const xml = (0, xmlUtils_1.parseXmlString)(stylesFile.content.toString());
+        const xml = (0, xmlUtils_js_1.parseXmlString)(stylesFile.content.toString());
         // Parse fonts
-        const fontsNode = (0, xmlUtils_1.getElementsByTagName)(xml, "fonts")[0];
+        const fontsNode = (0, xmlUtils_js_1.getElementsByTagName)(xml, "fonts")[0];
         const fonts = [];
         if (fontsNode) {
-            const fontNodes = (0, xmlUtils_1.getElementsByTagName)(fontsNode, "font");
+            const fontNodes = (0, xmlUtils_js_1.getElementsByTagName)(fontsNode, "font");
             for (const font of fontNodes) {
                 const formatting = {};
-                if ((0, xmlUtils_1.getElementsByTagName)(font, "b").length > 0)
+                if ((0, xmlUtils_js_1.getElementsByTagName)(font, "b").length > 0)
                     formatting.bold = true;
-                if ((0, xmlUtils_1.getElementsByTagName)(font, "i").length > 0)
+                if ((0, xmlUtils_js_1.getElementsByTagName)(font, "i").length > 0)
                     formatting.italic = true;
-                if ((0, xmlUtils_1.getElementsByTagName)(font, "u").length > 0)
+                if ((0, xmlUtils_js_1.getElementsByTagName)(font, "u").length > 0)
                     formatting.underline = true;
-                if ((0, xmlUtils_1.getElementsByTagName)(font, "strike").length > 0)
+                if ((0, xmlUtils_js_1.getElementsByTagName)(font, "strike").length > 0)
                     formatting.strikethrough = true;
-                const szNode = (0, xmlUtils_1.getElementsByTagName)(font, "sz")[0];
+                const szNode = (0, xmlUtils_js_1.getElementsByTagName)(font, "sz")[0];
                 if (szNode) {
                     const val = szNode.getAttribute("val");
                     if (val)
                         formatting.size = val + 'pt';
                 }
-                const colorNode = (0, xmlUtils_1.getElementsByTagName)(font, "color")[0];
+                const colorNode = (0, xmlUtils_js_1.getElementsByTagName)(font, "color")[0];
                 if (colorNode) {
                     const rgb = colorNode.getAttribute("rgb");
                     if (rgb)
                         formatting.color = '#' + rgb.substring(2); // Remove alpha channel
                 }
-                const nameNode = (0, xmlUtils_1.getElementsByTagName)(font, "name")[0];
+                const nameNode = (0, xmlUtils_js_1.getElementsByTagName)(font, "name")[0];
                 if (nameNode) {
                     const val = nameNode.getAttribute("val");
                     if (val)
                         formatting.font = val;
                 }
-                const vertAlignNode = (0, xmlUtils_1.getElementsByTagName)(font, "vertAlign")[0];
+                const vertAlignNode = (0, xmlUtils_js_1.getElementsByTagName)(font, "vertAlign")[0];
                 if (vertAlignNode) {
                     const val = vertAlignNode.getAttribute("val");
                     if (val === "subscript")
@@ -173,15 +175,15 @@ const parseExcel = async (buffer, config) => {
             }
         }
         // Parse fills (for background color)
-        const fillsNode = (0, xmlUtils_1.getElementsByTagName)(xml, "fills")[0];
+        const fillsNode = (0, xmlUtils_js_1.getElementsByTagName)(xml, "fills")[0];
         const fills = [];
         if (fillsNode) {
-            const fillNodes = (0, xmlUtils_1.getElementsByTagName)(fillsNode, "fill");
+            const fillNodes = (0, xmlUtils_js_1.getElementsByTagName)(fillsNode, "fill");
             for (const fill of fillNodes) {
                 const formatting = {};
-                const patternFill = (0, xmlUtils_1.getElementsByTagName)(fill, "patternFill")[0];
+                const patternFill = (0, xmlUtils_js_1.getElementsByTagName)(fill, "patternFill")[0];
                 if (patternFill) {
-                    const fgColor = (0, xmlUtils_1.getElementsByTagName)(patternFill, "fgColor")[0];
+                    const fgColor = (0, xmlUtils_js_1.getElementsByTagName)(patternFill, "fgColor")[0];
                     if (fgColor) {
                         const rgb = fgColor.getAttribute("rgb");
                         const theme = fgColor.getAttribute("theme");
@@ -207,9 +209,9 @@ const parseExcel = async (buffer, config) => {
             }
         }
         // Parse cellXfs (cell format definitions)
-        const cellXfsNode = (0, xmlUtils_1.getElementsByTagName)(xml, "cellXfs")[0];
+        const cellXfsNode = (0, xmlUtils_js_1.getElementsByTagName)(xml, "cellXfs")[0];
         if (cellXfsNode) {
-            const xfNodes = (0, xmlUtils_1.getElementsByTagName)(cellXfsNode, "xf");
+            const xfNodes = (0, xmlUtils_js_1.getElementsByTagName)(cellXfsNode, "xf");
             for (let i = 0; i < xfNodes.length; i++) {
                 const xf = xfNodes[i];
                 const formatting = {};
@@ -227,7 +229,7 @@ const parseExcel = async (buffer, config) => {
                         formatting.backgroundColor = fills[fillIdx].backgroundColor;
                     }
                 }
-                const alignmentNode = (0, xmlUtils_1.getElementsByTagName)(xf, "alignment")[0];
+                const alignmentNode = (0, xmlUtils_js_1.getElementsByTagName)(xf, "alignment")[0];
                 if (alignmentNode) {
                     const horizontal = alignmentNode.getAttribute("horizontal");
                     if (horizontal === 'center' || horizontal === 'right' || horizontal === 'justify' || horizontal === 'left') {
@@ -249,8 +251,8 @@ const parseExcel = async (buffer, config) => {
         for (const relFile of drawingRelsFiles) {
             const drawingFilename = relFile.path.split('/').pop()?.replace('.rels', '') || '';
             const drawingPath = `xl/drawings/${drawingFilename}`;
-            const relsXml = (0, xmlUtils_1.parseXmlString)(relFile.content.toString());
-            const relationships = (0, xmlUtils_1.getElementsByTagName)(relsXml, "Relationship");
+            const relsXml = (0, xmlUtils_js_1.parseXmlString)(relFile.content.toString());
+            const relationships = (0, xmlUtils_js_1.getElementsByTagName)(relsXml, "Relationship");
             if (!drawingImageMap[drawingPath]) {
                 drawingImageMap[drawingPath] = {};
             }
@@ -267,15 +269,15 @@ const parseExcel = async (buffer, config) => {
         // 2. Parse Drawings to get Alt Text and link to Rels
         const drawingFiles = files.filter(f => f.path.match(drawingsRegex));
         for (const drawingFile of drawingFiles) {
-            const xml = (0, xmlUtils_1.parseXmlString)(drawingFile.content.toString());
-            const pics = (0, xmlUtils_1.getElementsByTagName)(xml, "xdr:pic"); // SpreadsheetML drawing
+            const xml = (0, xmlUtils_js_1.parseXmlString)(drawingFile.content.toString());
+            const pics = (0, xmlUtils_js_1.getElementsByTagName)(xml, "xdr:pic"); // SpreadsheetML drawing
             const rels = drawingImageMap[drawingFile.path] || {};
             for (const pic of pics) {
-                const blipFill = (0, xmlUtils_1.getElementsByTagName)(pic, "xdr:blipFill")[0];
-                const blip = blipFill ? (0, xmlUtils_1.getElementsByTagName)(blipFill, "a:blip")[0] : null;
+                const blipFill = (0, xmlUtils_js_1.getElementsByTagName)(pic, "xdr:blipFill")[0];
+                const blip = blipFill ? (0, xmlUtils_js_1.getElementsByTagName)(blipFill, "a:blip")[0] : null;
                 const embedId = blip ? blip.getAttribute("r:embed") : null;
-                const nvPicPr = (0, xmlUtils_1.getElementsByTagName)(pic, "xdr:nvPicPr")[0];
-                const cNvPr = nvPicPr ? (0, xmlUtils_1.getElementsByTagName)(nvPicPr, "xdr:cNvPr")[0] : null;
+                const nvPicPr = (0, xmlUtils_js_1.getElementsByTagName)(pic, "xdr:nvPicPr")[0];
+                const cNvPr = nvPicPr ? (0, xmlUtils_js_1.getElementsByTagName)(nvPicPr, "xdr:cNvPr")[0] : null;
                 const altText = cNvPr ? (cNvPr.getAttribute("descr") || cNvPr.getAttribute("name")) : undefined;
                 if (embedId && rels[embedId]) {
                     rels[embedId].altText = altText || '';
@@ -284,7 +286,7 @@ const parseExcel = async (buffer, config) => {
         }
         // 3. Process Media Files
         for (const media of mediaFiles) {
-            const attachment = (0, imageUtils_1.createAttachment)(media.path.split('/').pop() || 'image', media.content);
+            const attachment = (0, imageUtils_js_1.createAttachment)(media.path.split('/').pop() || 'image', media.content);
             // Try to find alt text for this media
             let altText = '';
             for (const drawingPath in drawingImageMap) {
@@ -303,13 +305,13 @@ const parseExcel = async (buffer, config) => {
             if (config.ocr) {
                 if (attachment.mimeType.startsWith('image/')) {
                     try {
-                        const ocrText = await (0, ocrUtils_1.performOcr)(media.content, config.ocrLanguage);
-                        if (ocrText.trim()) {
-                            attachment.ocrText = ocrText.trim();
+                        const ocrText = (await (0, ocrUtils_js_1.performOcr)(media.content, { language: config.ocrLanguage, ...config.ocrConfig })).trim();
+                        if (ocrText) {
+                            attachment.ocrText = ocrText;
                         }
                     }
                     catch (e) {
-                        (0, errorUtils_1.logWarning)(`OCR failed for ${attachment.name}:`, config, e);
+                        (0, errorUtils_js_1.logWarning)(`OCR failed for ${attachment.name}:`, config, e);
                     }
                 }
             }
@@ -324,11 +326,11 @@ const parseExcel = async (buffer, config) => {
             };
             // Extract structured chart data
             try {
-                const chartData = (0, chartUtils_1.extractChartData)(chart.content);
+                const chartData = (0, chartUtils_js_1.extractChartData)(chart.content);
                 attachment.chartData = chartData;
             }
             catch (e) {
-                (0, errorUtils_1.logWarning)(`Failed to extract chart data from ${chart.path}:`, config, e);
+                (0, errorUtils_js_1.logWarning)(`Failed to extract chart data from ${chart.path}:`, config, e);
             }
             attachments.push(attachment);
         }
@@ -340,8 +342,8 @@ const parseExcel = async (buffer, config) => {
         for (const relFile of drawingRelsFiles) {
             const drawingFilename = relFile.path.split('/').pop()?.replace('.rels', '') || '';
             const drawingPath = `xl/drawings/${drawingFilename}`;
-            const relsXml = (0, xmlUtils_1.parseXmlString)(relFile.content.toString());
-            const relationships = (0, xmlUtils_1.getElementsByTagName)(relsXml, "Relationship");
+            const relsXml = (0, xmlUtils_js_1.parseXmlString)(relFile.content.toString());
+            const relationships = (0, xmlUtils_js_1.getElementsByTagName)(relsXml, "Relationship");
             if (!drawingChartMap[drawingPath]) {
                 drawingChartMap[drawingPath] = {};
             }
@@ -363,8 +365,8 @@ const parseExcel = async (buffer, config) => {
     const workbookRelsFile = files.find(f => f.path === 'xl/_rels/workbook.xml.rels');
     if (workbookFile && workbookRelsFile) {
         // Parse rels to get rId -> file mapping
-        const relsXml = (0, xmlUtils_1.parseXmlString)(workbookRelsFile.content.toString());
-        const relationships = (0, xmlUtils_1.getElementsByTagName)(relsXml, "Relationship");
+        const relsXml = (0, xmlUtils_js_1.parseXmlString)(workbookRelsFile.content.toString());
+        const relationships = (0, xmlUtils_js_1.getElementsByTagName)(relsXml, "Relationship");
         const rIdToFile = {};
         for (const rel of relationships) {
             const rId = rel.getAttribute("Id");
@@ -376,8 +378,8 @@ const parseExcel = async (buffer, config) => {
             }
         }
         // Parse workbook.xml to get sheet name -> rId mapping
-        const workbookXml = (0, xmlUtils_1.parseXmlString)(workbookFile.content.toString());
-        const sheets = (0, xmlUtils_1.getElementsByTagName)(workbookXml, "sheet");
+        const workbookXml = (0, xmlUtils_js_1.parseXmlString)(workbookFile.content.toString());
+        const sheets = (0, xmlUtils_js_1.getElementsByTagName)(workbookXml, "sheet");
         for (const sheet of sheets) {
             const name = sheet.getAttribute("name");
             const rId = sheet.getAttribute("r:id");
@@ -420,10 +422,10 @@ const parseExcel = async (buffer, config) => {
                     if (cMatches) {
                         for (const cXml of cMatches) {
                             // Extract cell value
-                            const typeMatch = cXml.match(/t="([a-z]+)"/);
+                            const typeMatch = cXml.match(/t="([a-zA-Z]+)"/);
                             const type = typeMatch ? typeMatch[1] : 'n'; // n = number (default)
-                            const vMatch = cXml.match(/<v>(.*?)<\/v>/);
-                            const tMatch = cXml.match(/<t>(.*?)<\/t>/);
+                            const vMatch = cXml.match(/<v>([\s\S]*?)<\/v>/);
+                            const tMatch = cXml.match(/<t>([\s\S]*?)<\/t>/);
                             let text = '';
                             let cellNodes = [];
                             if (type === 's' && vMatch) {
@@ -440,10 +442,10 @@ const parseExcel = async (buffer, config) => {
                                 }
                             }
                             else if (type === 'inlineStr' && tMatch) {
-                                text = tMatch[1];
+                                text = tMatch[1].trim();
                             }
                             else if (vMatch) {
-                                text = vMatch[1];
+                                text = vMatch[1].trim();
                             }
                             // Parse cell coordinate
                             const coordMatch = cXml.match(/r="([A-Z]+)(\d+)"/);
@@ -515,8 +517,8 @@ const parseExcel = async (buffer, config) => {
                 const relsFile = files.find(f => f.path === relsFilename);
                 const drawingMap = {}; // rId -> drawingPath
                 if (relsFile) {
-                    const relsXml = (0, xmlUtils_1.parseXmlString)(relsFile.content.toString());
-                    const relationships = (0, xmlUtils_1.getElementsByTagName)(relsXml, "Relationship");
+                    const relsXml = (0, xmlUtils_js_1.parseXmlString)(relsFile.content.toString());
+                    const relationships = (0, xmlUtils_js_1.getElementsByTagName)(relsXml, "Relationship");
                     for (const rel of relationships) {
                         const id = rel.getAttribute("Id");
                         const target = rel.getAttribute("Target");
@@ -588,7 +590,13 @@ const parseExcel = async (buffer, config) => {
         }
     }
     const corePropsFile = files.find(f => f.path.match(corePropsFileRegex));
-    const metadata = corePropsFile ? (0, xmlUtils_1.parseOfficeMetadata)(corePropsFile.content.toString()) : {};
+    const metadata = corePropsFile ? (0, xmlUtils_js_1.parseOfficeMetadata)(corePropsFile.content.toString()) : {};
+    const customPropsFile = files.find(f => f.path.match(customPropsFileRegex));
+    if (customPropsFile) {
+        const customProperties = (0, xmlUtils_js_1.parseOOXMLCustomProperties)(customPropsFile.content.toString());
+        if (Object.keys(customProperties).length > 0)
+            metadata.customProperties = customProperties;
+    }
     // Link OCR text and chart data to content nodes (like PPTX parser)
     const assignAttachmentData = (nodes) => {
         for (const node of nodes) {

package/dist/parsers/OpenOfficeParser.d.ts

@@ -20,7 +20,7 @@
  *
  * @module OpenOfficeParser
  */
-import { OfficeParserAST, OfficeParserConfig } from '../types';
+import { OfficeParserAST, OfficeParserConfig } from '../types.js';
 /**
  * Parses an OpenOffice document (.odt, .odp, .ods) and extracts content.
  *