feat: add receipt extraction proxy endpoint (refs #130)

Add POST /api/ocr/extract/receipt endpoint that proxies to the Python
OCR service's /extract/receipt for receipt-specific field extraction.

- ReceiptExtractionResponse type with receiptType, extractedFields, rawText
- OcrClient.extractReceipt() with optional receipt_type form field
- OcrService.extractReceipt() with 10MB max, image-only validation
- OcrController.extractReceipt() with file upload and error mapping
- Route with auth middleware
- 9 unit tests covering normal, edge, and error scenarios

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
Eric Gullickson
2026-02-11 09:26:57 -06:00
parent e98b45eb3a
commit e0e578a627
7 changed files with 444 additions and 1 deletions

View File

@@ -15,6 +15,14 @@ const SUPPORTED_TYPES = new Set([
'application/pdf',
]);
/** Image-only MIME types for receipt extraction (no PDF) */
const SUPPORTED_IMAGE_TYPES = new Set([
'image/jpeg',
'image/png',
'image/heic',
'image/heif',
]);
export class OcrController {
/**
* POST /api/ocr/extract
@@ -223,6 +231,111 @@ export class OcrController {
}
}
/**
* POST /api/ocr/extract/receipt
* Extract data from a receipt image using receipt-specific OCR.
*/
async extractReceipt(
request: FastifyRequest,
reply: FastifyReply
) {
const userId = (request as any).user?.sub as string;
logger.info('Receipt extract requested', {
operation: 'ocr.controller.extractReceipt',
userId,
});
const file = await (request as any).file({ limits: { files: 1 } });
if (!file) {
logger.warn('No file provided for receipt extraction', {
operation: 'ocr.controller.extractReceipt.no_file',
userId,
});
return reply.code(400).send({
error: 'Bad Request',
message: 'No file provided',
});
}
const contentType = file.mimetype as string;
if (!SUPPORTED_IMAGE_TYPES.has(contentType)) {
logger.warn('Unsupported file type for receipt extraction', {
operation: 'ocr.controller.extractReceipt.unsupported_type',
userId,
contentType,
fileName: file.filename,
});
return reply.code(415).send({
error: 'Unsupported Media Type',
message: `Unsupported file type: ${contentType}. Supported: JPEG, PNG, HEIC`,
});
}
const chunks: Buffer[] = [];
for await (const chunk of file.file) {
chunks.push(chunk);
}
const fileBuffer = Buffer.concat(chunks);
if (fileBuffer.length === 0) {
logger.warn('Empty file provided for receipt extraction', {
operation: 'ocr.controller.extractReceipt.empty_file',
userId,
fileName: file.filename,
});
return reply.code(400).send({
error: 'Bad Request',
message: 'Empty file provided',
});
}
// Get optional receipt_type from form fields
const receiptType = file.fields?.receipt_type?.value as string | undefined;
try {
const result = await ocrService.extractReceipt(userId, {
fileBuffer,
contentType,
receiptType,
});
logger.info('Receipt extract completed', {
operation: 'ocr.controller.extractReceipt.success',
userId,
success: result.success,
receiptType: result.receiptType,
processingTimeMs: result.processingTimeMs,
});
return reply.code(200).send(result);
} catch (error: any) {
if (error.statusCode === 413) {
return reply.code(413).send({
error: 'Payload Too Large',
message: error.message,
});
}
if (error.statusCode === 415) {
return reply.code(415).send({
error: 'Unsupported Media Type',
message: error.message,
});
}
logger.error('Receipt extract failed', {
operation: 'ocr.controller.extractReceipt.error',
userId,
error: error.message,
});
return reply.code(500).send({
error: 'Internal Server Error',
message: 'Receipt extraction failed',
});
}
}
/**
* POST /api/ocr/jobs
* Submit an async OCR job for large files.