Compare commits

...

4 Commits

Author SHA1 Message Date
Gitea Actions
ecec686347 ci: Bump version to 0.7.5 [skip ci] 2025-12-31 22:27:56 +05:00
86de680080 flyer processing fixes
All checks were successful
Deploy to Test Environment / deploy-to-test (push) Successful in 16m36s
2025-12-31 09:27:06 -08:00
Gitea Actions
0371947065 ci: Bump version to 0.7.4 [skip ci] 2025-12-31 22:03:02 +05:00
296698758c flyer upload (anon) issues
All checks were successful
Deploy to Test Environment / deploy-to-test (push) Successful in 19m20s
2025-12-31 09:02:09 -08:00
8 changed files with 246 additions and 64 deletions

4
package-lock.json generated
View File

@@ -1,12 +1,12 @@
{
"name": "flyer-crawler",
"version": "0.7.3",
"version": "0.7.5",
"lockfileVersion": 3,
"requires": true,
"packages": {
"": {
"name": "flyer-crawler",
"version": "0.7.3",
"version": "0.7.5",
"dependencies": {
"@bull-board/api": "^6.14.2",
"@bull-board/express": "^6.14.2",

View File

@@ -1,7 +1,7 @@
{
"name": "flyer-crawler",
"private": true,
"version": "0.7.3",
"version": "0.7.5",
"type": "module",
"scripts": {
"dev": "concurrently \"npm:start:dev\" \"vite\"",

View File

@@ -63,7 +63,8 @@ describe('FlyerAiProcessor', () => {
};
vi.mocked(mockAiService.extractCoreDataFromFlyerImage).mockResolvedValue(mockAiResponse);
const result = await service.extractAndValidateData([], jobData, logger);
const imagePaths = [{ path: 'page1.jpg', mimetype: 'image/jpeg' }];
const result = await service.extractAndValidateData(imagePaths, jobData, logger);
expect(mockAiService.extractCoreDataFromFlyerImage).toHaveBeenCalledTimes(1);
expect(mockPersonalizationRepo.getAllMasterItems).toHaveBeenCalledTimes(1);
@@ -83,7 +84,8 @@ describe('FlyerAiProcessor', () => {
};
vi.mocked(mockAiService.extractCoreDataFromFlyerImage).mockResolvedValue(invalidResponse as any);
await expect(service.extractAndValidateData([], jobData, logger)).rejects.toThrow(
const imagePaths = [{ path: 'page1.jpg', mimetype: 'image/jpeg' }];
await expect(service.extractAndValidateData(imagePaths, jobData, logger)).rejects.toThrow(
AiDataValidationError,
);
});
@@ -101,7 +103,8 @@ describe('FlyerAiProcessor', () => {
vi.mocked(mockAiService.extractCoreDataFromFlyerImage).mockResolvedValue(mockAiResponse as any);
const { logger } = await import('./logger.server');
const result = await service.extractAndValidateData([], jobData, logger);
const imagePaths = [{ path: 'page1.jpg', mimetype: 'image/jpeg' }];
const result = await service.extractAndValidateData(imagePaths, jobData, logger);
// It should not throw, but return the data and log a warning.
expect(result.data).toEqual(mockAiResponse);
@@ -122,7 +125,8 @@ describe('FlyerAiProcessor', () => {
vi.mocked(mockAiService.extractCoreDataFromFlyerImage).mockResolvedValue(mockAiResponse);
const { logger } = await import('./logger.server');
const result = await service.extractAndValidateData([], jobData, logger);
const imagePaths = [{ path: 'page1.jpg', mimetype: 'image/jpeg' }];
const result = await service.extractAndValidateData(imagePaths, jobData, logger);
expect(result.data).toEqual(mockAiResponse);
expect(result.needsReview).toBe(true);
expect(logger.warn).toHaveBeenCalledWith(expect.any(Object), expect.stringContaining('contains no items. The flyer will be saved with an item_count of 0. Flagging for review.'));

View File

@@ -6,6 +6,7 @@ import type { FlyerInsert, FlyerItemInsert } from '../types';
import type { AiProcessorResult } from './flyerAiProcessor.server'; // Keep this import for AiProcessorResult
import { AiFlyerDataSchema } from '../types/ai'; // Import consolidated schema
import { generateFlyerIcon } from '../utils/imageProcessor';
import { TransformationError } from './processingErrors';
/**
* This class is responsible for transforming the validated data from the AI service
@@ -57,41 +58,47 @@ export class FlyerDataTransformer {
): Promise<{ flyerData: FlyerInsert; itemsForDb: FlyerItemInsert[] }> {
logger.info('Starting data transformation from AI output to database format.');
const { data: extractedData, needsReview } = aiResult;
try {
const { data: extractedData, needsReview } = aiResult;
const firstImage = imagePaths[0].path;
const iconFileName = await generateFlyerIcon(
firstImage,
path.join(path.dirname(firstImage), 'icons'),
logger,
);
const firstImage = imagePaths[0].path;
const iconFileName = await generateFlyerIcon(
firstImage,
path.join(path.dirname(firstImage), 'icons'),
logger,
);
const itemsForDb: FlyerItemInsert[] = extractedData.items.map((item) => this._normalizeItem(item));
const itemsForDb: FlyerItemInsert[] = extractedData.items.map((item) => this._normalizeItem(item));
const storeName = extractedData.store_name || 'Unknown Store (auto)';
if (!extractedData.store_name) {
logger.warn('AI did not return a store name. Using fallback "Unknown Store (auto)".');
const storeName = extractedData.store_name || 'Unknown Store (auto)';
if (!extractedData.store_name) {
logger.warn('AI did not return a store name. Using fallback "Unknown Store (auto)".');
}
const flyerData: FlyerInsert = {
file_name: originalFileName,
image_url: `/flyer-images/${path.basename(firstImage)}`,
icon_url: `/flyer-images/icons/${iconFileName}`,
checksum,
store_name: storeName,
valid_from: extractedData.valid_from,
valid_to: extractedData.valid_to,
store_address: extractedData.store_address, // The number of items is now calculated directly from the transformed data.
item_count: itemsForDb.length,
uploaded_by: userId,
status: needsReview ? 'needs_review' : 'processed',
};
logger.info(
{ itemCount: itemsForDb.length, storeName: flyerData.store_name },
'Data transformation complete.',
);
return { flyerData, itemsForDb };
} catch (err) {
logger.error({ err }, 'Transformation process failed');
// Wrap and rethrow with the new error class
throw new TransformationError('Flyer Data Transformation Failed');
}
const flyerData: FlyerInsert = {
file_name: originalFileName,
image_url: `/flyer-images/${path.basename(firstImage)}`,
icon_url: `/flyer-images/icons/${iconFileName}`,
checksum,
store_name: storeName,
valid_from: extractedData.valid_from,
valid_to: extractedData.valid_to,
store_address: extractedData.store_address, // The number of items is now calculated directly from the transformed data.
item_count: itemsForDb.length,
uploaded_by: userId,
status: needsReview ? 'needs_review' : 'processed',
};
logger.info(
{ itemCount: itemsForDb.length, storeName: flyerData.store_name },
'Data transformation complete.',
);
return { flyerData, itemsForDb };
}
}

View File

@@ -1,12 +1,8 @@
// src/services/flyerProcessingService.server.test.ts
import { describe, it, expect, vi, beforeEach, type Mocked } from 'vitest';
import sharp from 'sharp';
import { Job, UnrecoverableError } from 'bullmq';
import type { Dirent } from 'node:fs';
import type { Logger } from 'pino';
import { z } from 'zod';
import { AiFlyerDataSchema } from './flyerAiProcessor.server';
import type { Flyer, FlyerInsert, FlyerItemInsert } from '../types';
import { AiFlyerDataSchema } from '../types/ai';
import type { FlyerInsert } from '../types';
import type { CleanupJobData, FlyerJobData } from '../types/job-data';
// 1. Create hoisted mocks FIRST

View File

@@ -203,6 +203,14 @@ export class FlyerProcessingService {
logger: Logger,
initialStages: ProcessingStage[],
): Promise<never> {
// Map specific error codes to their corresponding processing stage names.
// This is more maintainable than a long if/else if chain.
const errorCodeToStageMap = new Map<string, string>([
['PDF_CONVERSION_FAILED', 'Preparing Inputs'],
['UNSUPPORTED_FILE_TYPE', 'Preparing Inputs'],
['AI_VALIDATION_FAILED', 'Extracting Data with AI'],
['TRANSFORMATION_FAILED', 'Transforming AI Data'], // Add new mapping
]);
const normalizedError = error instanceof Error ? error : new Error(String(error));
let errorPayload: { errorCode: string; message: string; [key: string]: any };
let stagesToReport: ProcessingStage[] = [...initialStages]; // Create a mutable copy
@@ -215,16 +223,15 @@ export class FlyerProcessingService {
}
// Determine which stage failed
let errorStageIndex = -1;
const failedStageName = errorCodeToStageMap.get(errorPayload.errorCode);
let errorStageIndex = failedStageName ? stagesToReport.findIndex(s => s.name === failedStageName) : -1;
// 1. Try to map specific error codes/messages to stages
if (errorPayload.errorCode === 'PDF_CONVERSION_FAILED' || errorPayload.errorCode === 'UNSUPPORTED_FILE_TYPE') {
errorStageIndex = stagesToReport.findIndex(s => s.name === 'Preparing Inputs');
} else if (errorPayload.errorCode === 'AI_VALIDATION_FAILED') {
errorStageIndex = stagesToReport.findIndex(s => s.name === 'Extracting Data with AI');
} else if (errorPayload.message.includes('Icon generation failed')) {
// Fallback for generic errors not in the map. This is less robust and relies on string matching.
// A future improvement would be to wrap these in specific FlyerProcessingError subclasses.
if (errorStageIndex === -1 && errorPayload.message.includes('Icon generation failed')) {
errorStageIndex = stagesToReport.findIndex(s => s.name === 'Transforming AI Data');
} else if (errorPayload.message.includes('Database transaction failed')) {
}
if (errorStageIndex === -1 && errorPayload.message.includes('Database transaction failed')) {
errorStageIndex = stagesToReport.findIndex(s => s.name === 'Saving to Database');
}
@@ -260,24 +267,16 @@ export class FlyerProcessingService {
// Logging logic
if (normalizedError instanceof FlyerProcessingError) {
const logDetails: Record<string, any> = { err: normalizedError };
// Simplify log object creation
const logDetails: Record<string, any> = { ...errorPayload, err: normalizedError };
if (normalizedError instanceof AiDataValidationError) {
logDetails.validationErrors = normalizedError.validationErrors;
logDetails.rawData = normalizedError.rawData;
}
// Also include stderr for PdfConversionError in logs
if (normalizedError instanceof PdfConversionError) {
logDetails.stderr = normalizedError.stderr;
}
// Include the errorPayload details in the log, but avoid duplicating err, validationErrors, rawData
Object.assign(logDetails, errorPayload);
// Remove the duplicated err property if it was assigned by Object.assign
if ('err' in logDetails && logDetails.err === normalizedError) {
// This check prevents accidental deletion if 'err' was a legitimate property of errorPayload
delete logDetails.err;
}
// Ensure the original error object is always passed as 'err' for consistency in logging
logDetails.err = normalizedError;
logger.error(logDetails, `A known processing error occurred: ${normalizedError.name}`);
} else {

View File

@@ -62,6 +62,18 @@ export class AiDataValidationError extends FlyerProcessingError {
}
}
/**
* Error thrown when a transformation step fails.
*/
export class TransformationError extends FlyerProcessingError {
constructor(message: string) {
super(
message,
'TRANSFORMATION_FAILED',
'There was a problem transforming the flyer data. Please check the input.',
);
}
}
/**
* Error thrown when an image conversion fails (e.g., using sharp).
*/

View File

@@ -0,0 +1,164 @@
// src/tests/e2e/auth.e2e.test.ts
import { describe, it, expect, afterAll, beforeAll } from 'vitest';
import * as apiClient from '../../services/apiClient';
import { cleanupDb } from '../utils/cleanup';
import { createAndLoginUser, TEST_PASSWORD } from '../utils/testHelpers';
import type { UserProfile } from '../../types';
/**
* @vitest-environment node
*/
describe('Authentication E2E Flow', () => {
let testUser: UserProfile;
const createdUserIds: string[] = [];
beforeAll(async () => {
// Create a user that can be used for login-related tests in this suite.
const { user } = await createAndLoginUser({
email: `e2e-login-user-${Date.now()}@example.com`,
fullName: 'E2E Login User',
// E2E tests use apiClient which doesn't need the `request` object.
});
testUser = user;
createdUserIds.push(user.user.user_id);
});
afterAll(async () => {
if (createdUserIds.length > 0) {
await cleanupDb({ userIds: createdUserIds });
}
});
describe('Registration Flow', () => {
it('should successfully register a new user', async () => {
const email = `e2e-register-success-${Date.now()}@example.com`;
const fullName = 'E2E Register User';
// Act
const response = await apiClient.registerUser(email, TEST_PASSWORD, fullName);
const data = await response.json();
// Assert
expect(response.status).toBe(201);
expect(data.message).toBe('User registered successfully!');
expect(data.userprofile).toBeDefined();
expect(data.userprofile.user.email).toBe(email);
expect(data.token).toBeTypeOf('string');
// Add to cleanup
createdUserIds.push(data.userprofile.user.user_id);
});
it('should fail to register a user with a weak password', async () => {
const email = `e2e-register-weakpass-${Date.now()}@example.com`;
const weakPassword = '123';
// Act
const response = await apiClient.registerUser(email, weakPassword, 'Weak Pass User');
const errorData = await response.json();
// Assert
expect(response.status).toBe(400);
expect(errorData.errors[0].message).toContain('Password must be at least 8 characters long.');
});
it('should fail to register a user with a duplicate email', async () => {
const email = `e2e-register-duplicate-${Date.now()}@example.com`;
// Act 1: Register the user successfully
const firstResponse = await apiClient.registerUser(email, TEST_PASSWORD, 'Duplicate User');
const firstData = await firstResponse.json();
expect(firstResponse.status).toBe(201);
createdUserIds.push(firstData.userprofile.user.user_id); // Add for cleanup
// Act 2: Attempt to register the same user again
const secondResponse = await apiClient.registerUser(email, TEST_PASSWORD, 'Duplicate User');
const errorData = await secondResponse.json();
// Assert
expect(secondResponse.status).toBe(409); // Conflict
expect(errorData.message).toContain('A user with this email address already exists.');
});
});
describe('Login Flow', () => {
it('should successfully log in a registered user', async () => {
// Act: Attempt to log in with the user created in beforeAll
const response = await apiClient.loginUser(testUser.user.email, TEST_PASSWORD, false);
const data = await response.json();
// Assert
expect(response.status).toBe(200);
expect(data.userprofile).toBeDefined();
expect(data.userprofile.user.email).toBe(testUser.user.email);
expect(data.token).toBeTypeOf('string');
});
it('should fail to log in with an incorrect password', async () => {
// Act: Attempt to log in with the wrong password
const response = await apiClient.loginUser(testUser.user.email, 'wrong-password', false);
const errorData = await response.json();
// Assert
expect(response.status).toBe(401);
expect(errorData.message).toBe('Incorrect email or password.');
});
it('should fail to log in with a non-existent email', async () => {
const response = await apiClient.loginUser('no-one-here@example.com', TEST_PASSWORD, false);
const errorData = await response.json();
expect(response.status).toBe(401);
expect(errorData.message).toBe('Incorrect email or password.');
});
});
describe('Forgot/Reset Password Flow', () => {
it('should allow a user to reset their password and log in with the new one', async () => {
// Arrange: Create a user to reset the password for
const email = `e2e-reset-pass-${Date.now()}@example.com`;
const registerResponse = await apiClient.registerUser(email, TEST_PASSWORD, 'Reset Pass User');
const registerData = await registerResponse.json();
expect(registerResponse.status).toBe(201);
createdUserIds.push(registerData.userprofile.user.user_id);
// Act 1: Request a password reset.
// The test environment returns the token directly in the response for E2E testing.
const forgotResponse = await apiClient.requestPasswordReset(email);
const forgotData = await forgotResponse.json();
const resetToken = forgotData.token;
// Assert 1: Check that we received a token.
expect(forgotResponse.status).toBe(200);
expect(resetToken).toBeDefined();
expect(resetToken).toBeTypeOf('string');
// Act 2: Use the token to set a new password.
const newPassword = 'my-new-e2e-password-!@#$';
const resetResponse = await apiClient.resetPassword(resetToken, newPassword);
const resetData = await resetResponse.json();
// Assert 2: Check for a successful password reset message.
expect(resetResponse.status).toBe(200);
expect(resetData.message).toBe('Password has been reset successfully.');
// Act 3 & Assert 3 (Verification): Log in with the NEW password to confirm the change.
const loginResponse = await apiClient.loginUser(email, newPassword, false);
const loginData = await loginResponse.json();
expect(loginResponse.status).toBe(200);
expect(loginData.userprofile).toBeDefined();
expect(loginData.userprofile.user.email).toBe(email);
});
it('should return a generic success message for a non-existent email to prevent enumeration', async () => {
const nonExistentEmail = `non-existent-e2e-${Date.now()}@example.com`;
const response = await apiClient.requestPasswordReset(nonExistentEmail);
const data = await response.json();
expect(response.status).toBe(200);
expect(data.message).toBe('If an account with that email exists, a password reset link has been sent.');
expect(data.token).toBeUndefined();
});
});
});