gaschz/trilium
1
0
Fork 0
mirror of https://github.com/zadam/trilium.git synced 2026-02-05 21:34:23 +01:00
Code Issues Actions 10 Packages Projects Releases Wiki Activity

refactor(ocr): unnecessary initialization logic

Browse Source
This commit is contained in:
Elian Doran 2025-07-26 11:56:48 +03:00
parent 090b175152
commit c55aa6ee88
No known key found for this signature in database
1 changed files with 6 additions and 63 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

69
apps/server/src/services/ocr/ocr_service.ts
View File

@ -35,34 +35,16 @@ interface OCRBlobRow {
* Uses Tesseract.js for text recognition
*/
class OCRService {
private isInitialized = false;
private worker: Tesseract.Worker | null = null;
private isProcessing = false;
private processors: Map<string, FileProcessor> = new Map();
/**
* Initialize the OCR service
*/
async initialize(): Promise<void> {
if (this.isInitialized) {
return;
}
try {
log.info('Initializing OCR service with file processors...');
// Initialize file processors
this.processors.set('image', new ImageProcessor());
this.processors.set('pdf', new PDFProcessor());
this.processors.set('tiff', new TIFFProcessor());
this.processors.set('office', new OfficeProcessor());
this.isInitialized = true;
log.info('OCR service initialized successfully');
} catch (error) {
log.error(`Failed to initialize OCR service: ${error}`);
throw error;
}
constructor() {
// Initialize file processors
this.processors.set('image', new ImageProcessor());
this.processors.set('pdf', new PDFProcessor());
this.processors.set('tiff', new TIFFProcessor());
this.processors.set('office', new OfficeProcessor());
}
/**
@ -101,10 +83,6 @@ class OCRService {
* Extract text from file buffer using appropriate processor
*/
async extractTextFromFile(fileBuffer: Buffer, mimeType: string, options: OCRProcessingOptions = {}): Promise<OCRResult> {
if (!this.isInitialized) {
await this.initialize();
}
try {
log.info(`Starting OCR text extraction for MIME type: ${mimeType}`);
this.isProcessing = true;
@ -143,10 +121,6 @@ class OCRService {
return null;
}
if (!this.isInitialized) {
await this.initialize();
}
// Check if note type and MIME type are supported for OCR
if (note.type === 'image') {
if (!this.isSupportedMimeType(note.mime)) {
@ -205,10 +179,6 @@ class OCRService {
return null;
}
if (!this.isInitialized) {
await this.initialize();
}
// Check if attachment role and MIME type are supported for OCR
if (attachment.role === 'image') {
if (!this.isSupportedMimeType(attachment.mime)) {
@ -422,7 +392,6 @@ class OCRService {
await this.worker.terminate();
this.worker = null;
}
this.isInitialized = false;
log.info('OCR service cleaned up');
}
@ -563,32 +532,6 @@ class OCRService {
getAllSupportedMimeTypes(): string[] {
const supportedTypes = new Set<string>();
// Initialize processors if not already done
if (!this.isInitialized) {
// Return a static list if not initialized to avoid async issues
// This covers all known supported types
return [
// Images
'image/jpeg',
'image/jpg',
'image/png',
'image/gif',
'image/bmp',
'image/tiff',
'image/tif',
'image/webp',
// Documents
'application/pdf',
'application/vnd.openxmlformats-officedocument.wordprocessingml.document',
'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet',
'application/vnd.openxmlformats-officedocument.presentationml.presentation',
'application/msword',
'application/vnd.ms-excel',
'application/vnd.ms-powerpoint',
'application/rtf'
];
}
// Gather MIME types from all registered processors
for (const processor of this.processors.values()) {
const processorTypes = processor.getSupportedMimeTypes();