pdf upload fix

utils/pdfConverter.ts

@@ -1,13 +1,58 @@
-// This assumes pdfjsLib is available on the window, loaded from index.html
-declare const pdfjsLib: any;
+/**
+ * Renders a single PDF page to a canvas and returns it as a JPEG File object.
+ * @param pdfPage The PDF page object from pdf.js.
+ * @param pageNumber The page number (1-based).
+ * @param originalFileName The name of the original PDF file.
+ * @param scale The scale at which to render the page.
+ * @returns A promise that resolves to an image File object.
+ */
+const renderPageToImageFile = async (
+  pdfPage: any, // pdf.js page object
+  pageNumber: number,
+  originalFileName: string,
+  scale: number
+): Promise<File> => {
+  const viewport = pdfPage.getViewport({ scale });
 
-// Add a global declaration for `window.pdfjsWorker` to inform TypeScript
-// that this property is available, as it's loaded via a script in index.html.
-declare global {
-  interface Window {
-    pdfjsWorker: string;
+  const canvas = document.createElement('canvas');
+  const context = canvas.getContext('2d');
+  canvas.height = viewport.height;
+  canvas.width = viewport.width;
+
+  if (!context) {
+    throw new Error('Could not get canvas context');
   }
-}
+
+  await pdfPage.render({ canvasContext: context, viewport: viewport }).promise;
+
+  // Promisify canvas.toBlob for async/await usage
+  const blob = await new Promise<Blob | null>((resolve) => {
+    canvas.toBlob(resolve, 'image/jpeg', 0.9);
+  });
+
+  if (!blob) {
+    throw new Error(`Failed to convert page ${pageNumber} of PDF to blob.`);
+  }
+
+  const newFileName = originalFileName.replace(/\.pdf$/i, `_page_${pageNumber}.jpeg`);
+  return new File([blob], newFileName, { type: 'image/jpeg' });
+};
+
+/**
+ * Fetches a PDF document from a File object.
+ * @param pdfFile The PDF file.
+ * @returns A promise that resolves to the pdf.js document object.
+ */
+const getPdfDocument = async (pdfFile: File) => {
+  // @ts-ignore - pdfjsLib is globally available from the script tag in index.html
+  if (typeof pdfjsLib === 'undefined') {
+    throw new Error('pdf.js library is not loaded. Please check the script tag in index.html.');
+  }
+  const arrayBuffer = await pdfFile.arrayBuffer();
+  // @ts-ignore - pdfjsLib is globally available
+  const pdf = await pdfjsLib.getDocument(arrayBuffer).promise;
+  return pdf;
+};
 
 /**
  * Converts all pages of a PDF file into an array of image File objects.
@@ -19,48 +64,30 @@ export const convertPdfToImageFiles = async (
   pdfFile: File,
   onProgress?: (currentPage: number, totalPages: number) => void
 ): Promise<{ imageFiles: File[], pageCount: number }> => {
-    if (typeof pdfjsLib === 'undefined' || !pdfjsLib.GlobalWorkerOptions.workerSrc) {
-        // Set workerSrc if not already set by the script in index.html
-        pdfjsLib.GlobalWorkerOptions.workerSrc = window.pdfjsWorker;
-    }
+  const pdf = await getPdfDocument(pdfFile);
+  const pageCount = pdf.numPages;
+  const imageFiles: File[] = [];
+  const scale = 1.5;
+
+  // Create an array of promises, one for each page rendering task.
+  const pagePromises = Array.from({ length: pageCount }, async (_, i) => {
+    const pageNumber = i + 1;
+    const page = await pdf.getPage(pageNumber);
+    const imageFile = await renderPageToImageFile(page, pageNumber, pdfFile.name, scale);
     
-    const arrayBuffer = await pdfFile.arrayBuffer();
-    const pdf = await pdfjsLib.getDocument(arrayBuffer).promise;
-    const pageCount = pdf.numPages;
-    const imageFiles: File[] = [];
-
-    for (let i = 1; i <= pageCount; i++) {
-        onProgress?.(i, pageCount);
-        const page = await pdf.getPage(i);
-        const viewport = page.getViewport({ scale: 1.5 });
-        
-        const canvas = document.createElement('canvas');
-        const context = canvas.getContext('2d');
-        canvas.height = viewport.height;
-        canvas.width = viewport.width;
-
-        if (!context) {
-            throw new Error('Could not get canvas context');
-        }
-
-        await page.render({ canvasContext: context, viewport: viewport }).promise;
-
-        const blob = await new Promise<Blob | null>((resolve) => {
-            canvas.toBlob(resolve, 'image/jpeg', 0.9);
-        });
-
-        if (blob) {
-            const newFileName = pdfFile.name.replace(/\.pdf$/i, `_page_${i}.jpeg`);
-            const imageFile = new File([blob], newFileName, { type: 'image/jpeg' });
-            imageFiles.push(imageFile);
-        } else {
-            console.warn(`Failed to convert page ${i} of PDF to blob.`);
-        }
-    }
+    // Report progress as each page finishes.
+    onProgress?.(pageNumber, pageCount);
     
-    if (imageFiles.length === 0 && pageCount > 0) {
-      throw new Error('PDF conversion resulted in zero images, though the PDF has pages. It might be corrupted or contain non-standard content.');
-    }
+    return imageFile;
+  });
 
-    return { imageFiles, pageCount };
+  // Process all pages in parallel and collect the results.
+  const results = await Promise.all(pagePromises);
+  imageFiles.push(...results);
+
+  if (imageFiles.length === 0 && pageCount > 0) {
+    throw new Error('PDF conversion resulted in zero images, though the PDF has pages. It might be corrupted or contain non-standard content.');
+  }
+
+  return { imageFiles, pageCount };
 };