Merge branch 'main' into advanced-doclist

src/app/api/nlp/route.ts

@@ -1,31 +1,23 @@
+import { NextRequest, NextResponse } from 'next/server';
 import nlp from 'compromise';
 
-// Text preprocessing function to clean and normalize text
-export const preprocessSentenceForAudio = (text: string): string => {
+const MAX_BLOCK_LENGTH = 300;
+
+const preprocessSentenceForAudio = (text: string): string => {
   return text
-    // Replace URLs with descriptive text including domain
     .replace(/\S*(?:https?:\/\/|www\.)([^\/\s]+)(?:\/\S*)?/gi, '- (link to $1) -')
-    // Remove special characters except basic punctuation
-    //.replace(/[^\w\s.,!?;:'"()-]/g, ' ')
-    // Fix hyphenated words at line breaks (word- word -> wordword)
     .replace(/(\w+)-\s+(\w+)/g, '$1$2')
-    // Replace multiple spaces with single space
     .replace(/\s+/g, ' ')
-    // Trim whitespace
     .trim();
 };
 
-const MAX_BLOCK_LENGTH = 300; // Maximum characters per block
-
-export const splitIntoSentences = (text: string): string[] => {
-  // Split text into paragraphs first
+const splitIntoSentences = (text: string): string[] => {
   const paragraphs = text.split(/\n+/);
   const blocks: string[] = [];
 
   for (const paragraph of paragraphs) {
     if (!paragraph.trim()) continue;
 
-    // Preprocess each paragraph
     const cleanedText = preprocessSentenceForAudio(paragraph);
     const doc = nlp(cleanedText);
     const rawSentences = doc.sentences().out('array') as string[];
@@ -35,23 +27,42 @@ export const splitIntoSentences = (text: string): string[] => {
     for (const sentence of rawSentences) {
       const trimmedSentence = sentence.trim();
       
-      // If adding this sentence would exceed the limit, start a new block
       if (currentBlock && (currentBlock.length + trimmedSentence.length + 1) > MAX_BLOCK_LENGTH) {
         blocks.push(currentBlock.trim());
         currentBlock = trimmedSentence;
       } else {
-        // Add to current block with a space if not empty
         currentBlock = currentBlock 
           ? `${currentBlock} ${trimmedSentence}`
           : trimmedSentence;
       }
     }
 
-    // Add the last block if not empty
     if (currentBlock) {
       blocks.push(currentBlock.trim());
     }
   }
   
   return blocks;
-};
+};
+
+export async function POST(req: NextRequest) {
+  try {
+    const { text } = await req.json();
+    if (!text) {
+      return NextResponse.json({ error: 'No text provided' }, { status: 400 });
+    }
+
+    if (text.length <= MAX_BLOCK_LENGTH) {
+      // Single sentence preprocessing
+      const cleanedText = preprocessSentenceForAudio(text);
+      return NextResponse.json({ sentences: [cleanedText] });
+    }
+
+    // Full text splitting into sentences
+    const sentences = splitIntoSentences(text);
+    return NextResponse.json({ sentences });
+  } catch (error) {
+    console.error('Error processing text:', error);
+    return NextResponse.json({ error: 'Failed to process text' }, { status: 500 });
+  }
+}

src/contexts/PDFContext.tsx

@@ -33,6 +33,8 @@ import {
   handleTextClick,
 } from '@/utils/pdf';
 
+import type { PDFDocumentProxy } from 'pdfjs-dist';
+
 /**
  * Interface defining all available methods and properties in the PDF context
  */
@@ -43,11 +45,12 @@ interface PDFContextType {
   currDocPages: number | undefined;
   currDocPage: number;
   currDocText: string | undefined;
+  pdfDocument: PDFDocumentProxy | undefined;
   setCurrentDocument: (id: string) => Promise<void>;
   clearCurrDoc: () => void;
 
   // PDF functionality
-  onDocumentLoadSuccess: ({ numPages }: { numPages: number }) => void;
+  onDocumentLoadSuccess: (pdf: PDFDocumentProxy) => void;
   highlightPattern: (text: string, pattern: string, containerRef: React.RefObject<HTMLDivElement>) => void;
   clearHighlights: () => void;
   handleTextClick: (
@@ -75,15 +78,17 @@ export function PDFProvider({ children }: { children: ReactNode }) {
   const [currDocURL, setCurrDocURL] = useState<string>();
   const [currDocName, setCurrDocName] = useState<string>();
   const [currDocText, setCurrDocText] = useState<string>();
+  const [pdfDocument, setPdfDocument] = useState<PDFDocumentProxy>();
 
   /**
    * Handles successful document load
    * 
    * @param {Object} param0 - Object containing numPages
    */
-  const onDocumentLoadSuccess = useCallback(({ numPages }: { numPages: number }) => {
-    console.log('Document loaded:', numPages);
-    setCurrDocPages(numPages);
+  const onDocumentLoadSuccess = useCallback((pdf: PDFDocumentProxy) => {
+    console.log('Document loaded:', pdf.numPages);
+    setCurrDocPages(pdf.numPages);
+    setPdfDocument(pdf);
   }, [setCurrDocPages]);
 
   /**
@@ -91,15 +96,15 @@ export function PDFProvider({ children }: { children: ReactNode }) {
    */
   const loadCurrDocText = useCallback(async () => {
     try {
-      if (!currDocURL) return;
-      const text = await extractTextFromPDF(currDocURL, currDocPage);
+      if (!pdfDocument) return;
+      const text = await extractTextFromPDF(pdfDocument, currDocPage);
       setCurrDocText(text);
       setTTSText(text);
 
     } catch (error) {
       console.error('Error loading PDF text:', error);
     }
-  }, [currDocURL, currDocPage, setTTSText]);
+  }, [pdfDocument, currDocPage, setTTSText]);
 
   /**
    * Updates the current document text when the page changes
@@ -136,6 +141,7 @@ export function PDFProvider({ children }: { children: ReactNode }) {
     setCurrDocURL(undefined);
     setCurrDocText(undefined);
     setCurrDocPages(undefined);
+    setPdfDocument(undefined);
     stop();
   }, [setCurrDocPages, stop]);
 
@@ -153,6 +159,7 @@ export function PDFProvider({ children }: { children: ReactNode }) {
       highlightPattern,
       clearHighlights,
       handleTextClick,
+      pdfDocument,
     }),
     [
       onDocumentLoadSuccess,
@@ -163,6 +170,7 @@ export function PDFProvider({ children }: { children: ReactNode }) {
       currDocPage,
       currDocText,
       clearCurrDoc,
+      pdfDocument,
     ]
   );
 

src/contexts/TTSContext.tsx

@@ -29,7 +29,6 @@ import toast from 'react-hot-toast';
 import { useParams } from 'next/navigation';
 
 import { useConfig } from '@/contexts/ConfigContext';
-import { splitIntoSentences, preprocessSentenceForAudio } from '@/utils/nlp';
 import { audioBufferToURL } from '@/utils/audio';
 import { useAudioCache } from '@/hooks/audio/useAudioCache';
 import { useVoiceManagement } from '@/hooks/audio/useVoiceManagement';
@@ -152,51 +151,71 @@ export function TTSProvider({ children }: { children: React.ReactNode }) {
    */
   const setText = useCallback((text: string) => {
     console.log('Setting page text:', text);
-    const newSentences = splitIntoSentences(text);
-
-    // If skipBlank is enabled and there's no text
-    if (isPlaying && skipBlank && newSentences.length === 0) {
-      if (isEPUB && locationChangeHandlerRef.current) {
-        // For EPUB, use the location handler to move to next section
-        locationChangeHandlerRef.current('next');
-        
-        toast.success('Skipping blank section', {
-          id: `epub-section-skip`,
-          iconTheme: {
-            primary: 'var(--accent)',
-            secondary: 'var(--background)',
-          },
-          style: {
-            background: 'var(--background)',
-            color: 'var(--accent)',
-          },
-          duration: 1000,
-          position: 'top-center',
-        });
-        return;
-      } else if (currDocPageNumber < currDocPages!) {
-        // For PDF, increment the page
-        incrementPage();
-        
-        toast.success(`Skipping blank page ${currDocPageNumber}`, {
-          id: `page-${currDocPageNumber}`,
-          iconTheme: {
-            primary: 'var(--accent)',
-            secondary: 'var(--background)',
-          },
-          style: {
-            background: 'var(--background)',
-            color: 'var(--accent)',
-          },
-          duration: 1000,
-          position: 'top-center',
-        });
-        return;
-      }
-    }
     
-    setSentences(newSentences);
-    setNextPageLoading(false);
+    fetch('/api/nlp', {
+      method: 'POST',
+      headers: { 'Content-Type': 'application/json' },
+      body: JSON.stringify({ text }),
+    })
+      .then(response => {
+        if (!response.ok) {
+          throw new Error('Failed to process text');
+        }
+        return response.json();
+      })
+      .then(({ sentences: newSentences }) => {
+        // If skipBlank is enabled and there's no text
+        if (isPlaying && skipBlank && newSentences.length === 0) {
+          if (isEPUB && locationChangeHandlerRef.current) {
+            locationChangeHandlerRef.current('next');
+            
+            toast.success('Skipping blank section', {
+              id: `epub-section-skip`,
+              iconTheme: {
+                primary: 'var(--accent)',
+                secondary: 'var(--background)',
+              },
+              style: {
+                background: 'var(--background)',
+                color: 'var(--accent)',
+              },
+              duration: 1000,
+              position: 'top-center',
+            });
+            return;
+          } else if (currDocPageNumber < currDocPages!) {
+            incrementPage();
+            
+            toast.success(`Skipping blank page ${currDocPageNumber}`, {
+              id: `page-${currDocPageNumber}`,
+              iconTheme: {
+                primary: 'var(--accent)',
+                secondary: 'var(--background)',
+              },
+              style: {
+                background: 'var(--background)',
+                color: 'var(--accent)',
+              },
+              duration: 1000,
+              position: 'top-center',
+            });
+            return;
+          }
+        }
+        
+        setSentences(newSentences);
+        setNextPageLoading(false);
+      })
+      .catch(error => {
+        console.error('Error processing text:', error);
+        toast.error('Failed to process text', {
+          style: {
+            background: 'var(--background)',
+            color: 'var(--accent)',
+          },
+          duration: 3000,
+        });
+      });
   }, [isPlaying, skipBlank, currDocPageNumber, currDocPages, incrementPage, isEPUB]);
 
   /**
@@ -437,9 +456,8 @@ export function TTSProvider({ children }: { children: React.ReactNode }) {
     // Only set processing state if not preloading
     if (!preload) setIsProcessing(true);
   
-    const cleanedSentence = preprocessSentenceForAudio(sentence);
-    const audioBuffer = await getAudio(cleanedSentence);
-    
+    // No need to preprocess again since setText already did it
+    const audioBuffer = await getAudio(sentence);
     return audioBufferToURL(audioBuffer!);
   }, [isProcessing, audioContext, getAudio]);
 

src/utils/pdf.ts

@@ -2,9 +2,11 @@ import { pdfjs } from 'react-pdf';
 import nlp from 'compromise';
 import stringSimilarity from 'string-similarity';
 import type { TextItem } from 'pdfjs-dist/types/src/display/api';
+import type { PDFDocumentProxy } from 'pdfjs-dist';
 
-// Set worker from public directory
+// Set worker from public directory and compatibility mode
 pdfjs.GlobalWorkerOptions.workerSrc = '/pdf.worker.mjs';
+pdfjs.GlobalWorkerOptions.workerPort = null;
 
 interface TextMatch {
   elements: HTMLElement[];
@@ -24,17 +26,8 @@ export function convertPDFDataToURL(pdfData: Blob): Promise<string> {
 }
 
 // Text Processing functions
-export async function extractTextFromPDF(pdfURL: string, pageNumber: number): Promise<string> {
+export async function extractTextFromPDF(pdf: PDFDocumentProxy, pageNumber: number): Promise<string> {
   try {
-    const base64Data = pdfURL.split(',')[1];
-    const binaryData = atob(base64Data);
-    const bytes = new Uint8Array(binaryData.length);
-    for (let i = 0; i < binaryData.length; i++) {
-      bytes[i] = binaryData.charCodeAt(i);
-    }
-
-    const loadingTask = pdfjs.getDocument({ data: bytes });
-    const pdf = await loadingTask.promise;
     const page = await pdf.getPage(pageNumber);
     const textContent = await page.getTextContent();