@leolionart/n8n-nodes-pdf-extractor 1.1.0 → 1.2.0

package/dist/nodes/PdfExtractor/PdfExtractor.node.js

@@ -2,10 +2,8 @@
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.PdfExtractor = void 0;
 const n8n_workflow_1 = require("n8n-workflow");
-// Use legacy build for Node.js compatibility
-const pdf_mjs_1 = require("pdfjs-dist/legacy/build/pdf.mjs");
-// Disable worker for Node.js environment
-pdf_mjs_1.GlobalWorkerOptions.workerSrc = '';
+// unpdf provides a simple API for PDF text extraction with password support
+const unpdf_1 = require("unpdf");
 /**
  * Parse page range string into array of page numbers
  * Supports: "1-5", "1,3,5", "1-3,7,9-11", or empty for all pages
@@ -154,40 +152,38 @@ class PdfExtractor {
                 // Validate binary data exists
                 const binaryData = this.helpers.assertBinaryData(itemIndex, binaryPropertyName);
                 const buffer = await this.helpers.getBinaryDataBuffer(itemIndex, binaryPropertyName);
-                // Convert buffer to Uint8Array for pdfjs
+                // Convert buffer to Uint8Array
                 const pdfData = new Uint8Array(buffer);
-                // Load PDF document
-                let pdfDocument;
+                // Get document info first to know total pages
+                let pdf;
                 try {
-                    const loadingTask = (0, pdf_mjs_1.getDocument)({
-                        data: pdfData,
+                    pdf = await (0, unpdf_1.getDocumentProxy)(pdfData, {
                         password: password || undefined,
-                        useSystemFonts: true,
                     });
-                    pdfDocument = await loadingTask.promise;
                 }
                 catch (error) {
                     const errorMessage = error.message || String(error);
-                    if (errorMessage.includes('Invalid password') || errorMessage.includes('Incorrect Password')) {
-                        throw new n8n_workflow_1.NodeOperationError(this.getNode(), 'Invalid password for PDF file', { itemIndex });
-                    }
-                    if (errorMessage.includes('password')) {
-                        throw new n8n_workflow_1.NodeOperationError(this.getNode(), 'This PDF is password-protected. Please provide the correct password.', { itemIndex });
+                    if (errorMessage.toLowerCase().includes('password')) {
+                        if (password) {
+                            throw new n8n_workflow_1.NodeOperationError(this.getNode(), 'Invalid password for PDF file', { itemIndex });
+                        }
+                        else {
+                            throw new n8n_workflow_1.NodeOperationError(this.getNode(), 'This PDF is password-protected. Please provide the password.', { itemIndex });
+                        }
                     }
                     throw new n8n_workflow_1.NodeOperationError(this.getNode(), `Failed to load PDF: ${errorMessage}`, { itemIndex });
                 }
-                const numPages = pdfDocument.numPages;
-                // Parse page range
+                const numPages = pdf.numPages;
                 const pagesToExtract = parsePageRange(options.pageRange || '', numPages);
-                // Extract text from each page
+                // Extract text from selected pages
                 const pageTexts = [];
                 for (const pageNum of pagesToExtract) {
                     try {
-                        const page = await pdfDocument.getPage(pageNum);
+                        const page = await pdf.getPage(pageNum);
                         const textContent = await page.getTextContent();
-                        // Extract text items and join them
+                        // Join text items
                         const pageText = textContent.items
-                            .filter((item) => 'str' in item)
+                            .filter((item) => typeof item === 'object' && item !== null && 'str' in item)
                             .map((item) => item.str)
                             .join(' ')
                             .replace(/\s+/g, ' ')
@@ -195,16 +191,14 @@ class PdfExtractor {
                         pageTexts.push({ page: pageNum, text: pageText });
                     }
                     catch (pageError) {
-                        // Continue with other pages if one fails
-                        console.warn(`Failed to extract text from page ${pageNum}: ${pageError}`);
+                        console.warn(`Failed to extract page ${pageNum}: ${pageError}`);
                         pageTexts.push({ page: pageNum, text: '' });
                     }
                 }
                 const outputProperty = options.outputProperty || 'text';
-                const joinPages = options.joinPages !== false; // Default to true
+                const joinPages = options.joinPages !== false;
                 let outputData;
                 if (joinPages) {
-                    // Join all pages with separator
                     const separator = options.pageSeparator || '\n\n--- Page {page} ---\n\n';
                     const fullText = pageTexts
                         .map((p, index) => {
@@ -225,7 +219,6 @@ class PdfExtractor {
                     };
                 }
                 else {
-                    // Return array of pages
                     const pagesOutput = options.includePageNumbers
                         ? pageTexts
                         : pageTexts.map(p => p.text);

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@leolionart/n8n-nodes-pdf-extractor",
-  "version": "1.1.0",
+  "version": "1.2.0",
   "description": "n8n community node to extract text from password-protected PDFs - no external dependencies required",
   "keywords": [
     "n8n-community-node-package",
@@ -59,6 +59,6 @@
     "n8n-workflow": "*"
   },
   "dependencies": {
-    "pdfjs-dist": "^4.9.155"
+    "unpdf": "^0.12.1"
   }
 }