feat: 添加华能集团电子商务平台爬虫功能

- 新增 ChngCrawler 爬虫类,支持华能集团电子商务平台招标信息抓取
- 更新 BidCrawlerService,集成 ChngCrawler 到爬虫任务中
- 添加代理配置支持,从环境变量读取代理设置
- 优化爬虫逻辑,支持多个爬虫并行执行
- 新增 ChngCrawler 的单元测试文件
- 改进错误处理,单个爬虫失败不影响其他爬虫执行
- 更新 chdtp_target.ts,添加代理认证支持
This commit is contained in:
dmy
2026-01-11 18:20:43 +08:00
parent d9105797f4
commit 07a7301968
5 changed files with 350 additions and 15 deletions

View File

@@ -1,7 +1,9 @@
import { Injectable, Logger } from '@nestjs/common';
import { ConfigService } from '@nestjs/config';
import * as puppeteer from 'puppeteer';
import { BidsService } from '../../bids/services/bid.service';
import { ChdtpCrawler } from './chdtp_target';
import { ChngCrawler } from './chng_target';
@Injectable()
export class BidCrawlerService {
@@ -9,33 +11,64 @@ export class BidCrawlerService {
constructor(
private bidsService: BidsService,
private configService: ConfigService,
) {}
async crawlAll() {
this.logger.log('Starting crawl task with Puppeteer...');
// 从环境变量读取代理配置
const proxyHost = this.configService.get<string>('PROXY_HOST');
const proxyPort = this.configService.get<string>('PROXY_PORT');
const proxyUsername = this.configService.get<string>('PROXY_USERNAME');
const proxyPassword = this.configService.get<string>('PROXY_PASSWORD');
// 构建代理参数
const args = [
'--no-sandbox',
'--disable-setuid-sandbox',
'--disable-blink-features=AutomationControlled',
'--disable-infobars',
'--window-position=0,0',
'--ignore-certifcate-errors',
'--ignore-certifcate-errors-spki-list',
];
if (proxyHost && proxyPort) {
const proxyUrl = proxyUsername && proxyPassword
? `http://${proxyUsername}:${proxyPassword}@${proxyHost}:${proxyPort}`
: `http://${proxyHost}:${proxyPort}`;
args.push(`--proxy-server=${proxyUrl}`);
this.logger.log(`Using proxy: ${proxyHost}:${proxyPort}`);
}
const browser = await puppeteer.launch({
headless: true,
args: ['--no-sandbox', '--disable-setuid-sandbox'],
args,
});
const crawlers = [ChdtpCrawler, ChngCrawler];
try {
// Currently only supports ChdtpCrawler, but can be extended to a list of crawlers
const crawler = ChdtpCrawler;
this.logger.log(`Crawling: ${crawler.name}`);
const results = await crawler.crawl(browser);
this.logger.log(`Extracted ${results.length} items from ${crawler.name}`);
for (const crawler of crawlers) {
this.logger.log(`Crawling: ${crawler.name}`);
try {
const results = await crawler.crawl(browser);
this.logger.log(`Extracted ${results.length} items from ${crawler.name}`);
for (const item of results) {
await this.bidsService.createOrUpdate({
title,
url: itemUrl,
publishDate,
source: type || 'Unknown',
});
for (const item of results) {
await this.bidsService.createOrUpdate({
title: item.title,
url: item.url,
publishDate: item.publishDate,
source: crawler.name,
unit: '',
});
}
} catch (err) {
this.logger.error(`Error crawling ${crawler.name}: ${err.message}`);
}
}
} catch (error) {
this.logger.error(`Crawl task failed: ${error.message}`);
} finally {

View File

@@ -15,6 +15,13 @@ export const ChdtpCrawler = {
async crawl(browser: puppeteer.Browser): Promise<ChdtpResult[]> {
const logger = new Logger('ChdtpCrawler');
const page = await browser.newPage();
const username = process.env.PROXY_USERNAME;
const password = process.env.PROXY_PASSWORD;
if (username && password) {
await page.authenticate({ username, password });
}
await page.setUserAgent('Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/110.0.0.0 Safari/537.36');
const allResults: ChdtpResult[] = [];

View File

@@ -0,0 +1,60 @@
import { ChngCrawler } from './chng_target';
import * as puppeteer from 'puppeteer';
// Increase timeout to 120 seconds for manual inspection and slow sites
jest.setTimeout(120000);
describe('ChngCrawler Real Site Test', () => {
let browser: puppeteer.Browser;
beforeAll(async () => {
browser = await puppeteer.launch({
headless: false, // Run in non-headless mode
args: [
'--no-sandbox',
'--disable-setuid-sandbox',
'--disable-blink-features=AutomationControlled',
'--window-size=1920,1080'
],
defaultViewport: null
});
});
afterAll(async () => {
if (browser) {
// Keep open for a few seconds after test to see result
await new Promise(r => setTimeout(r, 50000));
// await browser.close();
}
});
it('should visit the website and list all found bid information', async () => {
console.log(`
Starting crawl for: ${ChngCrawler.name}`);
console.log(`Target URL: ${ChngCrawler.url}`);
const results = await ChngCrawler.crawl(browser);
console.log(`
Successfully found ${results.length} items:
`);
console.log('----------------------------------------');
results.forEach((item, index) => {
console.log(`${index + 1}. [${item.publishDate.toLocaleDateString()}] ${item.title}`);
console.log(` Link: ${item.url}`);
console.log('----------------------------------------');
});
expect(results).toBeDefined();
expect(Array.isArray(results)).toBeTruthy();
if (results.length === 0) {
console.warn('Warning: No items found. Observe the browser window to see if content is loading or if there is a verification challenge.');
} else {
const firstItem = results[0];
expect(firstItem.title).toBeTruthy();
expect(firstItem.url).toMatch(/^https?:\/\//);
expect(firstItem.publishDate).toBeInstanceOf(Date);
}
});
});

View File

@@ -0,0 +1,163 @@
import * as puppeteer from 'puppeteer';
import { Logger } from '@nestjs/common';
import { ChdtpResult } from './chdtp_target';
export const ChngCrawler = {
name: '华能集团电子商务平台',
url: 'https://ec.chng.com.cn/ecmall/index.html#/purchase/home?top=0',
baseUrl: 'https://ec.chng.com.cn/ecmall/index.html',
async crawl(browser: puppeteer.Browser): Promise<ChdtpResult[]> {
const logger = new Logger('ChngCrawler');
let page = await browser.newPage();
const username = process.env.PROXY_USERNAME;
const password = process.env.PROXY_PASSWORD;
if (username && password) {
await page.authenticate({ username, password });
}
await page.evaluateOnNewDocument(() => {
Object.defineProperty(navigator, 'webdriver', { get: () => false });
});
await page.setUserAgent('Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/143.0.0.0 Safari/537.36');
await page.setViewport({ width: 1920, height: 1080 });
const allResults: ChdtpResult[] = [];
let currentPage = 1;
const maxPages = 5;
try {
logger.log('Navigating to Bing...');
await page.goto('https://cn.bing.com', { waitUntil: 'networkidle2' });
logger.log('Searching for target site...');
const searchBoxSelector = 'input[name="q"]';
await page.waitForSelector(searchBoxSelector);
await page.type(searchBoxSelector, 'https://ec.chng.com.cn/');
await page.keyboard.press('Enter');
await page.waitForNavigation({ waitUntil: 'networkidle2' });
logger.log('Clicking search result...');
const firstResultSelector = '#b_results .b_algo h2 a';
await page.waitForSelector(firstResultSelector);
const newTargetPromise = browser.waitForTarget(target => target.opener() === page.target());
await page.click(firstResultSelector);
const newTarget = await newTargetPromise;
const newPage = await newTarget.page();
if (newPage) {
await page.close();
page = newPage;
if (username && password) {
await page.authenticate({ username, password });
}
}
await page.waitForNavigation({ waitUntil: 'domcontentloaded' }).catch(() => {});
// PAUSE 15 SECONDS as requested
logger.log('Pausing 15 seconds before looking for "采购专栏"...');
await new Promise(r => setTimeout(r, 15000));
logger.log('Looking for "采购专栏" link...');
await page.waitForFunction(() => {
const divs = Array.from(document.querySelectorAll('div.text'));
return divs.some(div => div.textContent && div.textContent.includes('采购专栏'));
}, { timeout: 60000 });
const purchaseTargetPromise = browser.waitForTarget(target => target.opener() === page.target(), { timeout: 15000 }).catch(() => null);
await page.evaluate(() => {
const divs = Array.from(document.querySelectorAll('div.text'));
const target = divs.find(div => div.textContent && div.textContent.includes('采购专栏')) as HTMLElement;
if (target) target.click();
});
const purchaseTarget = await purchaseTargetPromise;
if (purchaseTarget) {
const pPage = await purchaseTarget.page();
if (pPage) {
logger.log('Switched to Purchase Page tab.');
page = pPage;
if (username && password) {
await page.authenticate({ username, password });
}
await new Promise(r => setTimeout(r, 5000));
}
}
logger.log(`Active URL: ${page.url()}`);
while (currentPage <= maxPages) {
logger.log(`Processing page ${currentPage}...`);
// Wait for table rows to load
await page.waitForFunction(() => {
return document.querySelectorAll('tr.ant-table-row').length > 0;
}, { timeout: 60000 }).catch(() => logger.warn('Content not found. Site might be slow.'));
const pageResults = await page.evaluate((baseUrl) => {
// Extract from table rows
const items = Array.from(document.querySelectorAll('tr.ant-table-row'));
return items.map(item => {
const titleSpan = item.querySelector('span.list-text');
const dateCell = item.querySelector('td.ant-table-row-cell-break-word p');
if (titleSpan && dateCell) {
const title = titleSpan.textContent?.trim() || '';
const dateStr = dateCell.textContent?.trim() || '';
if (title.length < 5) return null; // Filter noise
// URL is not directly available in the table, need to construct from data-row-key
const rowKey = item.getAttribute('data-row-key');
const url = rowKey ? `${baseUrl}#/purchase/detail?id=${rowKey}` : '';
return {
title,
dateStr,
url
};
}
return null;
}).filter(i => i !== null);
}, this.baseUrl);
if (pageResults.length === 0) {
logger.warn(`No results found on page ${currentPage}. Extraction failed.`);
break;
}
allResults.push(...pageResults.map(r => ({
title: r!.title,
publishDate: new Date(r!.dateStr),
url: r!.url
})));
logger.log(`Extracted ${pageResults.length} items.`);
// Pagination: look for the "right" icon SVG
const nextButton = await page.$('svg[data-icon="right"]');
if (!nextButton) break;
await nextButton.click();
await new Promise(r => setTimeout(r, 5000));
currentPage++;
}
return allResults;
} catch (error) {
logger.error(`Crawl failed: ${error.message}`);
return allResults;
} finally {
if (page) await page.close();
}
},
extract() { return []; }
};

View File

@@ -0,0 +1,72 @@
import { chromium } from 'playwright';
import { ChngCrawler } from './chng_target';
jest.setTimeout(120000);
describe('ChngCrawler Playwright Test', () => {
let browser;
beforeAll(async () => {
browser = await chromium.launch({
headless: false,
args: ['--no-sandbox', '--disable-setuid-sandbox']
});
});
afterAll(async () => {
if (browser) {
await browser.close();
}
});
it('should visit the website and list all found bid information', async () => {
console.log(`
Starting crawl for: ${ChngCrawler.name}`);
console.log(`Target URL: ${ChngCrawler.url}`);
const context = await browser.newContext({
viewport: { width: 1920, height: 1080 },
userAgent: 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36',
});
const page = await context.newPage();
// Add stealth scripts if needed, but Playwright is often better at evasion
await page.addInitScript(() => {
Object.defineProperty(navigator, 'webdriver', { get: () => false });
});
await page.goto(ChngCrawler.url, { waitUntil: 'networkidle', timeout: 60000 });
// Wait for content
try {
await page.waitForSelector('.ant-table-row', { timeout: 30000 });
} catch (e) {
console.warn('Timed out waiting for .ant-table-row');
}
const content = await page.content();
// Reuse the extraction logic from the Crawler definition
const results = ChngCrawler.extract(content);
console.log(`
Successfully found ${results.length} items:
`);
console.log('----------------------------------------');
results.forEach((item, index) => {
console.log(`${index + 1}. [${item.publishDate.toLocaleDateString()}] ${item.title}`);
console.log(` Link: ${item.url}`);
console.log('----------------------------------------');
});
if (results.length === 0) {
console.warn('No items found. Debugging content length: ' + content.length);
if (content.length < 500) {
console.log('Content dump:', content);
}
}
expect(Array.isArray(results)).toBeTruthy();
});
});