Upwork Job Scraper

# Dockerfile

FROM apify/actor-node-puppeteer-chrome:20

RUN npm ls crawlee apify puppeteer playwright

COPY --chown=myuser package*.json ./

RUN npm --quiet set progress=false \
    && npm install --omit=dev --omit=optional \
    && npm install puppeteer-extra puppeteer-extra-plugin-stealth puppeteer-extra-plugin-recaptcha random-useragent \
    && echo "Installed NPM packages:" \
    && (npm list --omit=dev --all || true) \
    && echo "Node.js version:" \
    && node --version \
    && echo "NPM version:" \
    && npm --version \
    && rm -r ~/.npm


COPY --chown=myuser . ./

CMD ./start_xvfb_and_run_cmd.sh && npm start --silent

1

{
    "actorSpecification": 1,
    "name": "upwork-job-scraper",
    "title": "Project Puppeteer Crawler JavaScript",
    "description": "Crawlee and Puppeteer project in JavaScript.",
    "version": "0.0",
    "meta": {
        "templateId": "js-crawlee-puppeteer-chrome"
    },
    "input": "./input_schema.json",
    "dockerfile": "./Dockerfile"
}

{
    "title": "Amazon Product Search",
    "type": "object",
    "schemaVersion": 1,
    "properties": {
        "searchQuery": {
            "title": "Search Query",
            "type": "string",
            "description": "Search query",
            "editor": "textfield",
            "default": "designer"
        }
    }
}

1/**************************************************************************
2 * main.js - Single-file Apify Actor scraping Indeed using puppeteer-extra
3 *           and puppeteer-extra-plugin-recaptcha for reCAPTCHA solving,
4 *           with added logs for page titles.
5 **************************************************************************/
6
7import { Actor } from 'apify';
8import { PuppeteerCrawler, Dataset } from 'crawlee';
9import puppeteer from 'puppeteer-extra';
10import StealthPlugin from 'puppeteer-extra-plugin-stealth';
11import RecaptchaPlugin from 'puppeteer-extra-plugin-recaptcha';
12import randomUseragent from 'random-useragent';
13
14// 1) Use stealth plugin to mask many typical bot signatures
15puppeteer.use(StealthPlugin());
16
17// 2) Configure the recaptcha plugin (requires a solver, e.g. 2captcha)
18puppeteer.use(
19    RecaptchaPlugin({
20        provider: {
21            // Example: Using 2captcha (https://2captcha.com/).
22            // Store your 2captcha API key in an env var: process.env.CAPTCHA_API_KEY
23            id: '2captcha',
24            token: 'a8586dee497a9987d81ba6aa3f89a38c', // e.g., 'abc123...'
25        },
26        visualFeedback: false, // set to true to see boxes around captchas in headful mode
27    })
28);
29
30/**
31 * Helper to sleep a random time between minMs and maxMs
32 * to appear more human and reduce detection.
33 */
34function randomSleep(minMs, maxMs) {
35    const delay = Math.floor(Math.random() * (maxMs - minMs + 1)) + minMs;
36    return new Promise((resolve) => setTimeout(resolve, delay));
37}
38
39// The main request handler for each page (listings or detail)
40async function handleRequest({ request, page, log, session, crawler }) {
41    log.info(`Scraping: ${request.url}`);
42
43    try {
44        // ----- Random user agent -----
45        const userAgent = randomUseragent.getRandom();
46        if (userAgent) {
47            await page.setUserAgent(userAgent);
48        }
49
50        // ----- Block images/fonts/CSS to reduce overhead -----
51        await page.setRequestInterception(true);
52        page.on('request', (req) => {
53            const type = req.resourceType();
54            if (['image', 'stylesheet', 'font'].includes(type)) req.abort();
55            else req.continue();
56        });
57
58        // ----- Random sleep before navigation -----
59        await randomSleep(3000, 6000);
60
61        // ----- Navigate to the page -----
62        const response = await page.goto(request.url, { waitUntil: 'domcontentloaded', timeout: 90000 });
63        if (!response || response.status() === 403) {
64            log.warning('Received 403 or missing response. Marking session bad.');
65            session.markBad();
66            throw new Error(`Request blocked or invalid response at ${request.url}`);
67        }
68
69        // ← Added this: Log the page title (for main listing pages).
70        const listingTitle = await page.title();
71        log.info(`Listing page title: ${listingTitle}`);
72
73        // ----- Attempt to solve any reCAPTCHAs on the page -----
74        const { solved, error } = await page.solveRecaptchas();
75        if (error) {
76            log.error(`Captcha solve error: ${error.message}`);
77        } else if (solved?.length) {
78            log.info(`Captcha solved: ${solved.length} reCAPTCHAs found and solved.`);
79        }
80
81        // ----- Wait for main container with job postings -----
82        await page.waitForSelector('.cardOutline.tapItem', { timeout: 60000 });
83        log.info('Indeed listings found. Extracting...');
84
85        // ----- Extract job listings from this page -----
86        const jobs = await page.evaluate(() => {
87            const results = [];
88            document.querySelectorAll('.cardOutline.tapItem').forEach((job) => {
89                const titleElement = job.querySelector('h2.jobTitle > a');
90                const companyElement = job.querySelector('[data-testid="company-name"]');
91                const locationElement = job.querySelector('[data-testid="text-location"]');
92
93                results.push({
94                    title: titleElement?.textContent.trim() || null,
95                    company: companyElement?.textContent.trim() || null,
96                    location: locationElement?.textContent.trim() || null,
97                    jobLink: titleElement
98                        ? new URL(titleElement.href, 'https://www.indeed.com').href
99                        : null,
100                });
101            });
102            return results;
103        });
104
105        log.info(`Found ${jobs.length} job postings on this page.`);
106
107        // ----- For each job, navigate to the detail page (optional) -----
108        for (const job of jobs) {
109            if (!job.jobLink) continue;
110            log.info(`Scraping detail page: ${job.jobLink}`);
111
112            try {
113                // Random delay before detail page
114                await randomSleep(2000, 5000);
115
116                const detailResponse = await page.goto(job.jobLink, { waitUntil: 'domcontentloaded', timeout: 90000 });
117                if (!detailResponse || detailResponse.status() === 403) {
118                    log.warning(`403 on detail page. Marking session bad. Job link: ${job.jobLink}`);
119                    session.markBad();
120                    continue;
121                }
122
123                // ← Added this: Log the page title (for job detail pages).
124                const detailTitle = await page.title();
125                log.info(`Detail page title: ${detailTitle}`);
126
127                // Attempt to solve captchas on detail page, if any
128                const { solved: detailSolved, error: detailError } = await page.solveRecaptchas();
129                if (detailError) {
130                    log.error(`Captcha solve error on detail: ${detailError.message}`);
131                } else if (detailSolved?.length) {
132                    log.info(`Captcha solved on detail page: ${detailSolved.length}`);
133                }
134
135                // Extract additional details from the job page
136                const detailData = await page.evaluate(() => {
137                    const descriptionElement = document.querySelector('#jobDescriptionText, .jobsearch-jobDescriptionText');
138                    const salaryInfoElement = document.querySelector('#salaryInfoAndJobType');
139                    const additionalDetails = {};
140
141                    let salary = null;
142                    let jobType = null;
143
144                    if (salaryInfoElement) {
145                        const salaryElement = salaryInfoElement.querySelector('.css-19j1a75');
146                        const jobTypeElement = salaryInfoElement.querySelector('.css-k5flys');
147                        salary = salaryElement?.textContent.trim() || null;
148                        jobType = jobTypeElement?.textContent.trim() || null;
149                    }
150
151                    const description = descriptionElement?.textContent.trim() || null;
152
153                    // Additional metadata
154                    document.querySelectorAll('.jobsearch-JobInfoHeader-meta > div').forEach((section) => {
155                        const key = section.querySelector('h3')?.textContent?.trim();
156                        const value = section.querySelector('div')?.textContent?.trim();
157                        if (key && value) {
158                            additionalDetails[key] = value;
159                        }
160                    });
161
162                    return { description, salary, jobType, additionalDetails };
163                });
164
165                // Attach detail info
166                job.description = detailData.description || 'N/A';
167                job.salary = detailData.salary || 'N/A';
168                job.jobType = detailData.jobType || 'N/A';
169                job.additionalDetails = detailData.additionalDetails;
170
171                // Push to default Apify Dataset
172                await Dataset.pushData(job);
173
174            } catch (detailErr) {
175                log.error(`Error scraping job detail for ${job.jobLink}: ${detailErr.message}`);
176            }
177        }
178
179        // ----- Pagination -----
180        if (request.userData.pageCount < request.userData.paginationLimit - 1) {
181            const nextHref = await page.evaluate(() => {
182                const nextButton = document.querySelector('[data-testid="pagination-page-next"]');
183                return nextButton ? nextButton.href : null;
184            });
185
186            if (nextHref) {
187                log.info(`Enqueuing next page: ${nextHref}`);
188                await crawler.addRequests([
189                    {
190                        url: nextHref,
191                        userData: {
192                            ...request.userData,
193                            pageCount: request.userData.pageCount + 1,
194                        },
195                    },
196                ]);
197            } else {
198                log.info('No more pages found.');
199            }
200        } else {
201            log.info(`Reached pagination limit of ${request.userData.paginationLimit} pages.`);
202        }
203
204    } catch (err) {
205        log.error(`Error scraping ${request.url}: ${err.message}`);
206        session.markBad();
207    }
208}
209
210Actor.main(async () => {
211    // Initialize the Apify Actor environment
212    await Actor.init();
213
214    // ----- Retrieve input or set defaults -----
215    const input = await Actor.getInput() || {};
216    const {
217        searchQuery = 'developer',
218        location = 'New York',
219        indeedUrl = 'www.indeed.com',
220        paginationLimit = 1,
221    } = input;
222
223    // Build the Indeed start URL
224    const startUrl = `https://${indeedUrl}/jobs?q=${encodeURIComponent(searchQuery)}&l=${encodeURIComponent(location)}`;
225
226    // (Optional) residential proxy usage
227    const proxyConfiguration = await Actor.createProxyConfiguration({
228        groups: ['RESIDENTIAL'],
229    });
230
231    // Create PuppeteerCrawler with puppeteer-extra (stealth + recaptcha plugin)
232    const crawler = new PuppeteerCrawler({
233        requestHandler: handleRequest,
234        maxConcurrency: 1, // keep low concurrency to avoid detection
235        maxRequestRetries: 3,
236        requestHandlerTimeoutSecs: 180,
237        proxyConfiguration,
238
239        // Force crawlee to use our puppeteer instance with plugins
240        launchContext: {
241            launcher: puppeteer,
242            launchOptions: {
243                headless: false,
244                args: [
245                    '--disable-gpu',
246                    '--no-sandbox',
247                    '--disable-setuid-sandbox',
248                    '--disable-blink-features=AutomationControlled',
249                ],
250            },
251        },
252    });
253
254    // Start the crawler
255    await crawler.run([
256        {
257            url: startUrl,
258            userData: {
259                pageCount: 0,
260                paginationLimit,
261            },
262        },
263    ]);
264
265    // Exit the Actor
266    await Actor.exit();
267});

1// routes.js
2import { Dataset } from 'apify';
3
4export async function handleRequest({ request, page, log, crawler, session }) {
5    log.info(`Scraping page: ${request.url}`);
6
7    try {
8        // Optionally block images/fonts for performance
9        await page.setRequestInterception(true);
10        page.on('request', (req) => {
11            const resourceType = req.resourceType();
12            if (['image', 'font', 'stylesheet'].includes(resourceType)) {
13                req.abort();
14            } else {
15                req.continue();
16            }
17        });
18
19        // Navigate to the page
20        const response = await page.goto(request.url, { waitUntil: 'domcontentloaded', timeout: 90000 });
21        if (!response || response.status() === 403) {
22            log.warning(`Blocked or invalid response (403) at ${request.url}. Marking session as bad.`);
23            session.markBad();
24            throw new Error(`Request blocked or invalid response for ${request.url}`);
25        }
26
27        // Wait for job postings to appear
28        await page.waitForSelector('.cardOutline.tapItem', { timeout: 60000 });
29        log.info('Job listings container found. Extracting...');
30
31        // Extract job listings
32        const jobs = await page.evaluate(() => {
33            const results = [];
34            document.querySelectorAll('.cardOutline.tapItem').forEach((job) => {
35                const titleElement = job.querySelector('h2.jobTitle > a');
36                const companyElement = job.querySelector('[data-testid="company-name"]');
37                const locationElement = job.querySelector('[data-testid="text-location"]');
38
39                results.push({
40                    title: titleElement?.textContent?.trim() || null,
41                    company: companyElement?.textContent?.trim() || null,
42                    location: locationElement?.textContent?.trim() || null,
43                    jobLink: titleElement
44                        ? new URL(titleElement.href, 'https://www.indeed.com').href
45                        : null,
46                });
47            });
48            return results;
49        });
50
51        log.info(`Found ${jobs.length} job postings on this page.`);
52
53        // Optionally navigate to each job's detail page
54        for (const job of jobs) {
55            if (!job.jobLink) continue;
56            log.info(`Scraping detail page: ${job.jobLink}`);
57
58            try {
59                const detailResponse = await page.goto(job.jobLink, { waitUntil: 'domcontentloaded', timeout: 90000 });
60                if (!detailResponse || detailResponse.status() === 403) {
61                    log.warning(`403 on detail page ${job.jobLink}. Marking session as bad.`);
62                    session.markBad();
63                    continue;
64                }
65
66                const jobDetails = await page.evaluate(() => {
67                    const descriptionElement = document.querySelector('#jobDescriptionText, .jobsearch-jobDescriptionText');
68                    const salaryInfoElement = document.querySelector('#salaryInfoAndJobType');
69                    const additionalDetails = {};
70
71                    let salary = null;
72                    let jobType = null;
73
74                    if (salaryInfoElement) {
75                        const salaryElement = salaryInfoElement.querySelector('.css-19j1a75'); // Salary text
76                        const jobTypeElement = salaryInfoElement.querySelector('.css-k5flys'); // Full-time, etc.
77                        salary = salaryElement?.textContent.trim() || null;
78                        jobType = jobTypeElement?.textContent.trim() || null;
79                    }
80
81                    const description = descriptionElement?.textContent.trim() || null;
82
83                    // Gather any extra metadata
84                    document.querySelectorAll('.jobsearch-JobInfoHeader-meta > div').forEach((section) => {
85                        const key = section.querySelector('h3')?.textContent?.trim();
86                        const value = section.querySelector('div')?.textContent?.trim();
87                        if (key && value) {
88                            additionalDetails[key] = value;
89                        }
90                    });
91
92                    return { description, salary, jobType, additionalDetails };
93                });
94
95                // Merge detail info into the job object
96                job.description = jobDetails.description || 'N/A';
97                job.salary = jobDetails.salary || 'N/A';
98                job.jobType = jobDetails.jobType || 'N/A';
99                job.additionalDetails = jobDetails.additionalDetails;
100
101                // Push the final job data into Apify dataset
102                await Dataset.pushData(job);
103
104            } catch (detailError) {
105                log.error(`Error extracting detail from ${job.jobLink}: ${detailError.message}`);
106            }
107        }
108
109        // Handle pagination if not at the limit
110        if (request.userData.pageCount < request.userData.paginationLimit - 1) {
111            const nextPageLink = await page.evaluate(() => {
112                const nextButton = document.querySelector('[data-testid="pagination-page-next"]');
113                return nextButton ? nextButton.href : null;
114            });
115
116            if (nextPageLink) {
117                const nextPageCount = request.userData.pageCount + 1;
118                log.info(`Queueing next page: ${nextPageLink}`);
119                await crawler.addRequests([
120                    {
121                        url: nextPageLink,
122                        userData: {
123                            ...request.userData,
124                            pageCount: nextPageCount,
125                        },
126                    },
127                ]);
128            } else {
129                log.info('No more pages found.');
130            }
131        } else {
132            log.info(`Pagination limit reached: ${request.userData.paginationLimit} page(s).`);
133        }
134
135    } catch (error) {
136        log.error(`Error scraping page ${request.url}: ${error.message}`);
137        session.markBad();
138    }
139}

# configurations
.idea

# crawlee and apify storage folders
apify_storage
crawlee_storage
storage

# installed files
node_modules

# git folder
.git

root = true

[*]
indent_style = space
indent_size = 4
charset = utf-8
trim_trailing_whitespace = true
insert_final_newline = true
end_of_line = lf

{
    "extends": "@apify",
    "root": true
}

# This file tells Git which files shouldn't be added to source control

.DS_Store
.idea
dist
node_modules
apify_storage
storage

{
    "name": "crawlee-puppeteer-javascript",
    "version": "0.0.1",
    "type": "module",
    "description": "This is an example of an Apify actor.",
    "dependencies": {
        "apify": "^3.2.6",
        "crawlee": "^3.11.5",
        "puppeteer": "*"
    },
    "devDependencies": {
        "@apify/eslint-config": "^0.4.0",
        "eslint": "^8.50.0"
    },
    "scripts": {
        "start": "node src/main.js",
        "test": "echo \"Error: oops, the actor has no tests yet, sad!\" && exit 1"
    },
    "author": "It's not you it's me",
    "license": "ISC"
}