albert/dailytrends
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

WebScraper

Browse Source
This commit is contained in:
albert
2025-07-29 12:45:28 +02:00
parent 891b1e478d
commit d8381c893d
2 changed files with 353 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

210
src/__tests__/WebScraper.test.ts Normal file
View File

@@ -0,0 +1,210 @@
import { WebScraper } from '../utils/WebScraper';
import { NewsSource } from '../types/Feed';
import { Logger } from '../utils/logger';
// Mock the Logger
jest.mock('../utils/logger', () => ({
Logger: {
error: jest.fn(),
warn: jest.fn(),
info: jest.fn(),
debug: jest.fn()
}
}));
// Mock fetch
global.fetch = jest.fn();
describe('WebScraper', () => {
let webScraper: WebScraper;
const mockFetch = fetch as jest.MockedFunction<typeof fetch>;
beforeEach(() => {
webScraper = new WebScraper();
jest.clearAllMocks();
});
describe('scrapeUrl', () => {
test('should successfully scrape a URL with complete metadata', async () => {
const mockHtml = `
<html>
<head>
<title>Test News Article</title>
<meta property="og:title" content="Test News Article">
<meta property="og:description" content="This is a test news article description">
<meta property="article:published_time" content="2024-01-15T10:30:00Z">
</head>
<body>
<h1>Test News Article</h1>
<p>Article content here...</p>
</body>
</html>
`;
mockFetch.mockResolvedValue({
ok: true,
text: () => Promise.resolve(mockHtml)
} as Response);
const result = await webScraper.scrapeUrl('https://example.com/news');
expect(result).toEqual({
title: 'Test News Article',
description: 'This is a test news article description',
url: 'https://example.com/news',
publishedAt: new Date('2024-01-15T10:30:00Z')
});
expect(mockFetch).toHaveBeenCalledWith('https://example.com/news', {
headers: {
'User-Agent': 'Mozilla/5.0 (compatible; DailyTrends/1.0)',
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8'
}
});
});
test('should handle HTTP errors gracefully', async () => {
mockFetch.mockResolvedValue({
ok: false,
status: 404,
statusText: 'Not Found'
} as Response);
const result = await webScraper.scrapeUrl('https://example.com/not-found');
expect(result).toBeNull();
expect(Logger.error).toHaveBeenCalledWith(
'Failed to fetch https://example.com/not-found: 404 Not Found'
);
});
test('should handle network errors gracefully', async () => {
mockFetch.mockRejectedValue(new Error('Network error'));
const result = await webScraper.scrapeUrl('https://example.com/error');
expect(result).toBeNull();
expect(Logger.error).toHaveBeenCalledWith(
'Error scraping https://example.com/error:',
expect.any(Error)
);
});
test('should return null when no title is found', async () => {
const mockHtml = `
<html>
<head>
<meta property="og:description" content="Description without title">
</head>
<body>
<p>Content without title</p>
</body>
</html>
`;
mockFetch.mockResolvedValue({
ok: true,
text: () => Promise.resolve(mockHtml)
} as Response);
const result = await webScraper.scrapeUrl('https://example.com/no-title');
expect(result).toBeNull();
expect(Logger.warn).toHaveBeenCalledWith('No title found for https://example.com/no-title');
});
test('should return null when no description is found', async () => {
const mockHtml = `
<html>
<head>
<title>Title Only</title>
</head>
<body>
<p>Content without description meta</p>
</body>
</html>
`;
mockFetch.mockResolvedValue({
ok: true,
text: () => Promise.resolve(mockHtml)
} as Response);
const result = await webScraper.scrapeUrl('https://example.com/no-description');
expect(result).toBeNull();
expect(Logger.warn).toHaveBeenCalledWith('No description found for https://example.com/no-description');
});
test('should use current date when no published date is found', async () => {
const mockHtml = `
<html>
<head>
<title>Test Article</title>
<meta property="og:description" content="Test description">
</head>
</html>
`;
mockFetch.mockResolvedValue({
ok: true,
text: () => Promise.resolve(mockHtml)
} as Response);
const beforeScrape = new Date();
const result = await webScraper.scrapeUrl('https://example.com/no-date');
const afterScrape = new Date();
expect(result).not.toBeNull();
expect(result!.publishedAt.getTime()).toBeGreaterThanOrEqual(beforeScrape.getTime());
expect(result!.publishedAt.getTime()).toBeLessThanOrEqual(afterScrape.getTime());
});
});
describe('convertToFeedData', () => {
test('should convert scraped data to feed format', () => {
const scrapedData = {
title: 'Test News',
description: 'Test description',
url: 'https://example.com/news',
publishedAt: new Date('2024-01-15T10:00:00Z')
};
const feedData = webScraper.convertToFeedData(scrapedData, NewsSource.EL_PAIS);
expect(feedData).toEqual({
title: 'Test News',
description: 'Test description',
url: 'https://example.com/news',
source: NewsSource.EL_PAIS,
publishedAt: new Date('2024-01-15T10:00:00Z'),
isManual: false
});
});
test('should handle HTML with special characters and entities', async () => {
const htmlWithEntities = `
<html>
<head>
<title>News &amp; Updates - El Pa&iacute;s</title>
<meta name="description" content="Breaking news &quot;today&quot; &amp; analysis">
</head>
</html>
`;
global.fetch = jest.fn().mockResolvedValue({
ok: true,
text: () => Promise.resolve(htmlWithEntities)
});
const result = await webScraper.scrapeUrl('https://example.com/news');
expect(result).toEqual({
title: 'News &amp; Updates - El Pa&iacute;s',
description: 'Breaking news &quot;today&quot; &amp; analysis',
url: 'https://example.com/news',
publishedAt: expect.any(Date)
});
});
});
});

143
src/utils/WebScraper.ts Normal file
View File

@@ -0,0 +1,143 @@
import { IFeed, NewsSource } from '../types/Feed.js';
import { Logger } from './logger.js';
interface ScrapedData {
title: string;
description: string;
url: string;
publishedAt: Date;
}
export class WebScraper {
private userAgent = 'Mozilla/5.0 (compatible; DailyTrends/1.0)';
async scrapeUrl(url: string): Promise<ScrapedData | null> {
try {
const response = await fetch(url, {
headers: {
'User-Agent': this.userAgent,
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8'
}
});
if (!response.ok) {
Logger.error(`Failed to fetch ${url}: ${response.status} ${response.statusText}`);
return null;
}
const html = await response.text();
return this.parseHtml(html, url);
} catch (error) {
Logger.error(`Error scraping ${url}:`, error);
return null;
}
}
private parseHtml(html: string, url: string): ScrapedData | null {
try {
// Extract title from <title> tag or Open Graph
const title = this.extractTitle(html);
if (!title) {
Logger.warn(`No title found for ${url}`);
return null;
}
// Extract description from meta tags
const description = this.extractDescription(html);
if (!description) {
Logger.warn(`No description found for ${url}`);
return null;
}
// Extract published date
const publishedAt = this.extractPublishedDate(html);
return {
title: title.trim(),
description: description.trim(),
url,
publishedAt
};
} catch (error) {
Logger.error(`Error parsing HTML for ${url}:`, error);
return null;
}
}
private extractTitle(html: string): string | null {
// Try Open Graph title first
const ogTitleMatch = html.match(/<meta\s+property=["']og:title["']\s+content=["']([^"']+)["']/i);
if (ogTitleMatch) {
return ogTitleMatch[1];
}
// Try Twitter title
const twitterTitleMatch = html.match(/<meta\s+name=["']twitter:title["']\s+content=["']([^"']+)["']/i);
if (twitterTitleMatch) {
return twitterTitleMatch[1];
}
// Fall back to <title> tag
const titleMatch = html.match(/<title[^>]*>([^<]+)<\/title>/i);
if (titleMatch) {
return titleMatch[1];
}
return null;
}
private extractDescription(html: string): string | null {
// Try Open Graph description first
const ogDescMatch = html.match(/<meta\s+property=["']og:description["']\s+content=["']([^"']+)["']/i);
if (ogDescMatch) {
return ogDescMatch[1];
}
// Try Twitter description
const twitterDescMatch = html.match(/<meta\s+name=["']twitter:description["']\s+content=["']([^"']+)["']/i);
if (twitterDescMatch) {
return twitterDescMatch[1];
}
// Try meta description
const metaDescMatch = html.match(/<meta\s+name=["']description["']\s+content=["']([^"']+)["']/i);
if (metaDescMatch) {
return metaDescMatch[1];
}
return null;
}
private extractPublishedDate(html: string): Date {
// Try various date formats
const datePatterns = [
/<meta\s+property=["']article:published_time["']\s+content=["']([^"']+)["']/i,
/<meta\s+name=["']pubdate["']\s+content=["']([^"']+)["']/i,
/<time[^>]+datetime=["']([^"']+)["']/i
];
for (const pattern of datePatterns) {
const match = html.match(pattern);
if (match) {
const date = new Date(match[1]);
if (!isNaN(date.getTime())) {
return date;
}
}
}
// Default to current date if no published date found
return new Date();
}
convertToFeedData(scrapedData: ScrapedData, source: NewsSource): Omit<IFeed, '_id' | 'createdAt' | 'updatedAt'> {
return {
title: scrapedData.title,
description: scrapedData.description,
url: scrapedData.url,
source,
publishedAt: scrapedData.publishedAt,
isManual: false
};
}
}