Dockery/Trilium
1
0
Fork 0
mirror of https://github.com/zadam/trilium.git synced 2026-06-26 19:21:42 +02:00
Code Issues Packages Projects Releases Activity

chore(ocr): remove unimplemented logic

Browse Source
This commit is contained in:
Elian Doran
2026-04-02 12:36:10 +03:00
parent cf73a4ef43
commit ad29375975
1 changed files with 13 additions and 75 deletions
Download Patch File Download Diff File Expand all files Collapse all files

88
apps/server/src/services/ocr/processors/pdf_processor.ts
View File

@@ -1,96 +1,34 @@
import pdfParse from 'pdf-parse';
import log from '../../log.js';
import { OCRProcessingOptions,OCRResult } from '../ocr_service.js';
import { OCRProcessingOptions, OCRResult } from '../ocr_service.js';
import { FileProcessor } from './file_processor.js';
import { ImageProcessor } from './image_processor.js';
/**
* PDF processor for extracting text from PDF files
* First tries to extract existing text, then falls back to OCR on images
* PDF processor for extracting embedded text from PDF files using pdf-parse.
*/
export class PDFProcessor extends FileProcessor {
private imageProcessor: ImageProcessor;
private readonly supportedTypes = ['application/pdf'];
constructor() {
super();
this.imageProcessor = new ImageProcessor();
}
canProcess(mimeType: string): boolean {
return mimeType.toLowerCase() === 'application/pdf';
}
getSupportedMimeTypes(): string[] {
return [...this.supportedTypes];
return ['application/pdf'];
}
async extractText(buffer: Buffer, options: OCRProcessingOptions = {}): Promise<OCRResult> {
try {
log.info('Starting PDF text extraction...');
log.info('Starting PDF text extraction...');
// First try to extract existing text from PDF
if (options.enablePDFTextExtraction !== false) {
const textResult = await this.extractTextFromPDF(buffer, options);
if (textResult.text.trim().length > 0) {
log.info(`PDF text extraction successful. Length: ${textResult.text.length}`);
return textResult;
}
}
const data = await pdfParse(buffer);
// Fall back to OCR if no text found or PDF text extraction is disabled
log.info('No text found in PDF or text extraction disabled, falling back to OCR...');
return await this.extractTextViaOCR(buffer, options);
} catch (error) {
log.error(`PDF text extraction failed: ${error}`);
throw error;
}
}
private async extractTextFromPDF(buffer: Buffer, options: OCRProcessingOptions): Promise<OCRResult> {
try {
const data = await pdfParse(buffer);
return {
text: data.text.trim(),
confidence: 0.99, // High confidence for direct text extraction
extractedAt: new Date().toISOString(),
language: options.language || "eng",
pageCount: data.numpages
};
} catch (error) {
log.error(`PDF text extraction failed: ${error}`);
throw error;
}
}
private async extractTextViaOCR(buffer: Buffer, options: OCRProcessingOptions): Promise<OCRResult> {
try {
// Convert PDF to images and OCR each page
// For now, we'll use a simple approach - convert first page to image
// In a full implementation, we'd convert all pages
// This is a simplified implementation
// In practice, you might want to use pdf2pic or similar library
// to convert PDF pages to images for OCR
// For now, we'll return a placeholder result
// indicating that OCR on PDF is not fully implemented
log.info('PDF to image conversion not fully implemented, returning placeholder');
return {
text: '[PDF OCR not fully implemented - would convert PDF pages to images and OCR each page]',
confidence: 0.0,
extractedAt: new Date().toISOString(),
language: options.language || "eng",
pageCount: 1
};
} catch (error) {
log.error(`PDF OCR extraction failed: ${error}`);
throw error;
}
return {
text: data.text.trim(),
confidence: 0.99,
extractedAt: new Date().toISOString(),
language: options.language || "eng",
pageCount: data.numpages
};
}
getProcessingType(): string {
@@ -98,6 +36,6 @@ export class PDFProcessor extends FileProcessor {
}
async cleanup(): Promise<void> {
await this.imageProcessor.cleanup();
// Nothing to clean up.
}
}