refactor(ocr): get rid of require imports

apps/server/src/services/ocr/processors/image_processor.ts

@@ -1,8 +1,9 @@
 import Tesseract from 'tesseract.js';
-import { FileProcessor } from './file_processor.js';
-import { OCRResult, OCRProcessingOptions } from '../ocr_service.js';
+
 import log from '../../log.js';
 import options from '../../options.js';
+import { OCRProcessingOptions,OCRResult } from '../ocr_service.js';
+import { FileProcessor } from './file_processor.js';
 
 /**
  * Image processor for extracting text from image files using Tesseract
@@ -135,7 +136,6 @@ export class ImageProcessor extends FileProcessor {
      */
     private getDefaultOCRLanguage(): string {
         try {
-            const options = require('../../options.js').default;
             const ocrLanguage = options.getOption('ocrLanguage');
             if (!ocrLanguage) {
                 throw new Error('OCR language not configured in user settings');
@@ -161,8 +161,8 @@ export class ImageProcessor extends FileProcessor {
             };
         }
 
-        let filteredWords: string[] = [];
-        let validConfidences: number[] = [];
+        const filteredWords: string[] = [];
+        const validConfidences: number[] = [];
 
         // Tesseract provides word-level data
         if (data.words && Array.isArray(data.words)) {
@@ -182,13 +182,12 @@ export class ImageProcessor extends FileProcessor {
                     filteredText: data.text.trim(),
                     overallConfidence
                 };
-            } else {
-                log.info(`Entire text filtered out due to low confidence ${overallConfidence} (below threshold ${minConfidence})`);
-                return {
-                    filteredText: '',
-                    overallConfidence
-                };
             }
+            log.info(`Entire text filtered out due to low confidence ${overallConfidence} (below threshold ${minConfidence})`);
+            return {
+                filteredText: '',
+                overallConfidence
+            };
         }
 
         // Calculate average confidence of accepted words

apps/server/src/services/ocr/processors/office_processor.ts

@@ -1,8 +1,10 @@
 import * as officeParser from 'officeparser';
-import { FileProcessor } from './file_processor.js';
-import { OCRResult, OCRProcessingOptions } from '../ocr_service.js';
-import { ImageProcessor } from './image_processor.js';
+
 import log from '../../log.js';
+import options from '../../options.js';
+import { OCRProcessingOptions,OCRResult } from '../ocr_service.js';
+import { FileProcessor } from './file_processor.js';
+import { ImageProcessor } from './image_processor.js';
 
 /**
  * Office document processor for extracting text and images from DOCX/XLSX/PPTX files
@@ -51,9 +53,9 @@ export class OfficeProcessor extends FileProcessor {
 
             const result: OCRResult = {
                 text: combinedText,
-                confidence: confidence,
+                confidence,
                 extractedAt: new Date().toISOString(),
-                language: language,
+                language,
                 pageCount: 1 // Office documents are treated as single logical document
             };
 
@@ -97,7 +99,6 @@ export class OfficeProcessor extends FileProcessor {
      */
     private getDefaultOCRLanguage(): string {
         try {
-            const options = require('../../options.js').default;
             const ocrLanguage = options.getOption('ocrLanguage');
             if (!ocrLanguage) {
                 throw new Error('OCR language not configured in user settings');

apps/server/src/services/ocr/processors/pdf_processor.ts

@@ -1,9 +1,10 @@
 import * as pdfParse from 'pdf-parse';
-import { FileProcessor } from './file_processor.js';
-import { OCRResult, OCRProcessingOptions } from '../ocr_service.js';
-import { ImageProcessor } from './image_processor.js';
+
 import log from '../../log.js';
-import sharp from 'sharp';
+import options from '../../options.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
@@ -58,7 +59,7 @@ export class PDFProcessor extends FileProcessor {
     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
@@ -77,15 +78,15 @@ export class PDFProcessor extends FileProcessor {
             // 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,
@@ -112,7 +113,6 @@ export class PDFProcessor extends FileProcessor {
      */
     private getDefaultOCRLanguage(): string {
         try {
-            const options = require('../../options.js').default;
             const ocrLanguage = options.getOption('ocrLanguage');
             if (!ocrLanguage) {
                 throw new Error('OCR language not configured in user settings');
@@ -132,16 +132,16 @@ export class PDFProcessor extends FileProcessor {
         if (!language || typeof language !== 'string') {
             return false;
         }
-        
+
         // Split by '+' for multi-language format
         const languages = language.split('+');
-        
+
         // Check each language code (should be 2-7 characters, alphanumeric with underscores)
         const validLanguagePattern = /^[a-zA-Z]{2,3}(_[a-zA-Z]{2,3})?$/;
-        
+
         return languages.every(lang => {
             const trimmed = lang.trim();
             return trimmed.length > 0 && validLanguagePattern.test(trimmed);
         });
     }
-}
+}

apps/server/src/services/ocr/processors/tiff_processor.ts

@@ -1,8 +1,10 @@
 import sharp from 'sharp';
-import { FileProcessor } from './file_processor.js';
-import { OCRResult, OCRProcessingOptions } from '../ocr_service.js';
-import { ImageProcessor } from './image_processor.js';
+
 import log from '../../log.js';
+import options from '../../options.js';
+import { OCRProcessingOptions,OCRResult } from '../ocr_service.js';
+import { FileProcessor } from './file_processor.js';
+import { ImageProcessor } from './image_processor.js';
 
 /**
  * TIFF processor for extracting text from multi-page TIFF files
@@ -45,7 +47,7 @@ export class TIFFProcessor extends FileProcessor {
             for (let page = 0; page < pageCount; page++) {
                 try {
                     log.info(`Processing TIFF page ${page + 1}/${pageCount}...`);
-                    
+
                     // Extract page as PNG buffer
                     const pageBuffer = await sharp(buffer, { page })
                         .png()
@@ -53,10 +55,10 @@ export class TIFFProcessor extends FileProcessor {
 
                     // OCR the page
                     const pageResult = await this.imageProcessor.extractText(pageBuffer, options);
-                    
+
                     if (pageResult.text.trim().length > 0) {
                         if (combinedText.length > 0) {
-                            combinedText += '\n\n--- Page ' + (page + 1) + ' ---\n';
+                            combinedText += `\n\n--- Page ${page + 1} ---\n`;
                         }
                         combinedText += pageResult.text;
                         totalConfidence += pageResult.confidence;
@@ -74,7 +76,7 @@ export class TIFFProcessor extends FileProcessor {
                 confidence: averageConfidence,
                 extractedAt: new Date().toISOString(),
                 language: options.language || this.getDefaultOCRLanguage(),
-                pageCount: pageCount
+                pageCount
             };
 
             log.info(`TIFF text extraction completed. Pages: ${pageCount}, Confidence: ${averageConfidence}%, Text length: ${result.text.length}`);
@@ -99,7 +101,6 @@ export class TIFFProcessor extends FileProcessor {
      */
     private getDefaultOCRLanguage(): string {
         try {
-            const options = require('../../options.js').default;
             const ocrLanguage = options.getOption('ocrLanguage');
             if (!ocrLanguage) {
                 throw new Error('OCR language not configured in user settings');
@@ -119,16 +120,16 @@ export class TIFFProcessor extends FileProcessor {
         if (!language || typeof language !== 'string') {
             return false;
         }
-        
+
         // Split by '+' for multi-language format
         const languages = language.split('+');
-        
+
         // Check each language code (should be 2-7 characters, alphanumeric with underscores)
         const validLanguagePattern = /^[a-zA-Z]{2,3}(_[a-zA-Z]{2,3})?$/;
-        
+
         return languages.every(lang => {
             const trimmed = lang.trim();
             return trimmed.length > 0 && validLanguagePattern.test(trimmed);
         });
     }
-}
+}