Pricing
Pay per usage
Crawls websites with the headless Chrome and Puppeteer library using a provided server-side Node.js code. This crawler is an alternative to apify/web-scraper that gives you finer control over the process. Supports both recursive crawling and list of URLs. Supports login to website.
5.0 (6)
Pricing
Pay per usage
211
9.3K
825
Issues response
14 days
Last modified
4 days ago
{ "extends": "@apify/tsconfig", "compilerOptions": { "outDir": "dist", "module": "ESNext", "allowJs": true, "skipLibCheck": true }, "include": ["src"]}{ "title": "Puppeteer Scraper Input", "description": "Use the following form to configure Puppeteer Scraper. The Start URLs and Page function are required and all other fields are optional. To learn more about the different options, click on their title or on the nearby question marks. For details about the Page function visit the Actor's README in the ACTOR INFO tab.", "type": "object", "schemaVersion": 1, "properties": { "startUrls": { "sectionCaption": "Basic configuration", "title": "Start URLs", "type": "array", "description": "URLs to start with", "prefill": [{ "url": "https://crawlee.dev/js" }], "editor": "requestListSources" }, "globs": { "title": "Glob Patterns", "type": "array", "description": "Glob patterns to match links in the page that you want to enqueue. Combine with Link selector to tell the scraper where to find links. Omitting the Glob patterns will cause the scraper to enqueue all links matched by the Link selector.", "editor": "globs", "default": [], "prefill": [ { "glob": "https://crawlee.dev/js/*/*" } ] }, "pseudoUrls": { "title": "Pseudo-URLs", "type": "array", "description": "Pseudo-URLs to match links in the page that you want to enqueue. Combine with Link selector to tell the scraper where to find links. Omitting the Pseudo-URLs will cause the scraper to enqueue all links matched by the Link selector.", "editor": "pseudoUrls", "default": [], "prefill": [] }, "excludes": { "title": "Exclude Glob Patterns", "type": "array", "description": "Glob patterns to match links in the page that you want to exclude from being enqueued.", "editor": "globs", "default": [], "prefill": [ { "glob": "/**/*.{png,jpg,jpeg,pdf}" } ] }, "linkSelector": { "title": "Link selector", "type": "string", "description": "CSS selector matching elements with 'href' attributes that should be enqueued. To enqueue urls from <code><div class=\"my-class\" href=...></code> tags, you would enter <strong>div.my-class</strong>. Leave empty to ignore all links.", "editor": "textfield", "prefill": "a" }, "clickableElementsSelector": { "title": "Clickable elements selector", "type": "string", "description": "For pages where simple 'href' links are not available, this attribute allows you to specify a CSS selector matching elements that the scraper will mouse click after the page function finishes. Any triggered requests, navigations or open tabs will be intercepted and the target URLs will be filtered using Pseudo URLs and/or Glob patterns and subsequently added to the request queue. Leave empty to prevent the scraper from clicking in the page. Using this setting will have a performance impact.", "editor": "textfield" }, "keepUrlFragments": { "title": "Keep URL fragments", "type": "boolean", "description": "URL fragments (the parts of URL after a <code>#</code>) are not considered when the scraper determines whether a URL has already been visited. This means that when adding URLs such as <code>https://example.com/#foo</code> and <code>https://example.com/#bar</code>, only the first will be visited. Turn this option on to tell the scraper to visit both.", "default": false }, "respectRobotsTxtFile": { "title": "Respect the robots.txt file", "type": "boolean", "description": "If enabled, the crawler will consult the robots.txt file for the target website before crawling each page. At the moment, the crawler does not use any specific user agent identifier. The crawl-delay directive is also not supported yet.", "default": false, "prefill": true }, "pageFunction": { "title": "Page function", "type": "string", "description": "Function executed for each request", "prefill": "async function pageFunction(context) {\n const { page, request, log } = context;\n const title = await page.title();\n log.info(`URL: ${request.url} TITLE: ${title}`);\n return {\n url: request.url,\n title\n };\n}", "editor": "javascript" }, "proxyConfiguration": { "sectionCaption": "Proxy and browser configuration", "title": "Proxy configuration", "type": "object", "description": "Specifies proxy servers that will be used by the scraper in order to hide its origin.<br><br>For details, see <a href='https://apify.com/apify/puppeteer-scraper#proxy-configuration' target='_blank' rel='noopener'>Proxy configuration</a> in README.", "prefill": { "useApifyProxy": true }, "default": { "useApifyProxy": true }, "editor": "proxy" }, "proxyRotation": { "title": "Proxy rotation", "type": "string", "description": "This property indicates the strategy of proxy rotation and can only be used in conjunction with Apify Proxy. The recommended setting automatically picks the best proxies from your available pool and rotates them evenly, discarding proxies that become blocked or unresponsive. If this strategy does not work for you for any reason, you may configure the scraper to either use a new proxy for each request, or to use one proxy as long as possible, until the proxy fails. IMPORTANT: This setting will only use your available Apify Proxy pool, so if you don't have enough proxies for a given task, no rotation setting will produce satisfactory results.", "default": "RECOMMENDED", "editor": "select", "enum": ["RECOMMENDED", "PER_REQUEST", "UNTIL_FAILURE"], "enumTitles": [ "Use recommended settings", "Rotate proxy after each request", "Use one proxy until failure" ] }, "sessionPoolName": { "title": "Session pool name", "type": "string", "description": "<b>Use only english alphanumeric characters dashes and underscores.</b> A session is a representation of a user. It has it's own IP and cookies which are then used together to emulate a real user. Usage of the sessions is controlled by the Proxy rotation option. By providing a session pool name, you enable sharing of those sessions across multiple Actor runs. This is very useful when you need specific cookies for accessing the websites or when a lot of your proxies are already blocked. Instead of trying randomly, a list of working sessions will be saved and a new Actor run can reuse those sessions. Note that the IP lock on sessions expires after 24 hours, unless the session is used again in that window.", "editor": "textfield", "minLength": 3, "maxLength": 200, "pattern": "[0-9A-z-]" }, "initialCookies": { "title": "Initial cookies", "type": "array", "description": "The provided cookies will be pre-set to all pages the scraper opens.", "default": [], "prefill": [], "editor": "json" }, "useChrome": { "title": "Use Chrome", "type": "boolean", "description": "The scraper will use a real Chrome browser instead of a Chromium masking as Chrome. Using this option may help with bypassing certain anti-scraping protections, but risks that the scraper will be unstable or not work at all.", "default": false, "groupCaption": "Browser masking", "groupDescription": "Settings that help mask as a real user and prevent scraper detection." }, "headless": { "title": "Run browsers in headless mode", "type": "boolean", "description": "By default, browsers run in headless mode. You can toggle this off to run them in headful mode, which can help with certain rare anti-scraping protections but is slower and more costly.", "default": true }, "ignoreSslErrors": { "title": "Ignore SSL errors", "type": "boolean", "description": "Scraper will ignore SSL certificate errors.", "default": false, "groupCaption": "Security", "groupDescription": "Various options that help you disable browser security features such as HTTPS certificates and CORS." }, "ignoreCorsAndCsp": { "title": "Ignore CORS and CSP", "type": "boolean", "description": "Scraper will ignore CSP (content security policy) and CORS (cross origin resource sharing) settings of visited pages and requested domains. This enables you to freely use XHR/Fetch to make HTTP requests from the scraper.", "default": false }, "downloadMedia": { "sectionCaption": "Performance and limits", "title": "Download media", "type": "boolean", "description": "Scraper will download media such as images, fonts, videos and sounds. Disabling this may speed up the scrape, but certain websites could stop working correctly.", "default": true, "groupCaption": "Resources", "groupDescription": "Settings that help to disable downloading web page resources." }, "downloadCss": { "title": "Download CSS", "type": "boolean", "description": "Scraper will download CSS stylesheets. Disabling this may speed up the scrape, but certain websites could stop working correctly.", "default": true }, "maxRequestRetries": { "title": "Max request retries", "type": "integer", "description": "Maximum number of times the request for the page will be retried in case of an error. Setting it to 0 means that the request will be attempted once and will not be retried if it fails.", "minimum": 0, "default": 3, "unit": "retries" }, "maxPagesPerCrawl": { "title": "Max pages per run", "type": "integer", "description": "Maximum number of pages that the scraper will open. 0 means unlimited.", "minimum": 0, "default": 0, "unit": "pages" }, "maxResultsPerCrawl": { "title": "Max result records", "type": "integer", "description": "Maximum number of results that will be saved to dataset. The scraper will terminate afterwards. 0 means unlimited.", "minimum": 0, "default": 0, "unit": "results" }, "maxCrawlingDepth": { "title": "Max crawling depth", "type": "integer", "description": "Defines how many links away from the StartURLs will the scraper descend. 0 means unlimited.", "minimum": 0, "default": 0 }, "maxConcurrency": { "title": "Max concurrency", "type": "integer", "description": "Defines how many pages can be processed by the scraper in parallel. The scraper automatically increases and decreases concurrency based on available system resources. Use this option to set a hard limit.", "minimum": 1, "default": 50 }, "pageLoadTimeoutSecs": { "title": "Page load timeout", "type": "integer", "description": "Maximum time the scraper will allow a web page to load in seconds.", "minimum": 1, "default": 60, "unit": "secs" }, "pageFunctionTimeoutSecs": { "title": "Page function timeout", "type": "integer", "description": "Maximum time the scraper will wait for the page function to execute in seconds.", "minimum": 1, "default": 60, "unit": "secs" }, "waitUntil": { "title": "Navigation wait until", "type": "array", "description": "The scraper will wait until the selected events are triggered in the page before executing the page function. Available events are <code>domcontentloaded</code>, <code>load</code>, <code>networkidle2</code> and <code>networkidle0</code>. <a href=\"https://pptr.dev/#?product=Puppeteer&show=api-pagegotourl-options\" target=\"_blank\">See Puppeteer docs</a>.", "default": ["networkidle2"], "prefill": ["networkidle2"], "editor": "json" }, "preNavigationHooks": { "sectionCaption": "Advanced configuration", "title": "Pre-navigation hooks", "type": "string", "description": "Async functions that are sequentially evaluated before the navigation. Good for setting additional cookies or browser properties before navigation. The function accepts two parameters, `crawlingContext` and `gotoOptions`, which are passed to the `page.goto()` function the crawler calls to navigate.", "prefill": "// We need to return array of (possibly async) functions here.\n// The functions accept two arguments: the \"crawlingContext\" object\n// and \"gotoOptions\".\n[\n async (crawlingContext, gotoOptions) => {\n const { page } = crawlingContext;\n // ...\n },\n]", "editor": "javascript" }, "postNavigationHooks": { "title": "Post-navigation hooks", "type": "string", "description": "Async functions that are sequentially evaluated after the navigation. Good for checking if the navigation was successful. The function accepts `crawlingContext` as the only parameter.", "prefill": "// We need to return array of (possibly async) functions here.\n// The functions accept a single argument: the \"crawlingContext\" object.\n[\n async (crawlingContext) => {\n const { page } = crawlingContext;\n // ...\n },\n]", "editor": "javascript" }, "closeCookieModals": { "title": "Dismiss cookie modals", "type": "boolean", "description": "Using the [I don't care about cookies](https://addons.mozilla.org/en-US/firefox/addon/i-dont-care-about-cookies/) browser extension. When on, the crawler will automatically try to dismiss cookie consent modals. This can be useful when crawling European websites that show cookie consent modals.", "default": false }, "maxScrollHeightPixels": { "title": "Maximum scrolling distance in pixels", "type": "integer", "description": "The crawler will scroll down the page until all content is loaded or the maximum scrolling distance is reached. Setting this to `0` disables scrolling altogether.", "default": 5000 }, "debugLog": { "title": "Debug log", "type": "boolean", "description": "Debug messages will be included in the log. Use <code>context.log.debug('message')</code> to log your own debug messages.", "default": false, "groupCaption": "Enable logs", "groupDescription": "Logs settings" }, "browserLog": { "title": "Browser log", "type": "boolean", "description": "Console messages from the Browser will be included in the log. This may result in the log being flooded by error messages, warnings and other messages of little value, especially with high concurrency.", "default": false }, "customData": { "title": "Custom data", "type": "object", "description": "This object will be available on pageFunction's context as customData.", "default": {}, "prefill": {}, "editor": "json" }, "datasetName": { "title": "Dataset name", "type": "string", "description": "Name or ID of the dataset that will be used for storing results. If left empty, the default dataset of the run will be used.", "editor": "textfield" }, "keyValueStoreName": { "title": "Key-value store name", "type": "string", "description": "Name or ID of the key-value store that will be used for storing records. If left empty, the default key-value store of the run will be used.", "editor": "textfield" }, "requestQueueName": { "title": "Request queue name", "type": "string", "description": "Name of the request queue that will be used for storing requests. If left empty, the default request queue of the run will be used.", "editor": "textfield" } }, "required": ["startUrls", "pageFunction", "proxyConfiguration"]}FROM apify/actor-node-puppeteer-chrome:22 AS builder
COPY package*.json ./
RUN npm install --include=dev --audit=false
COPY . ./
RUN npm run build
FROM apify/actor-node-puppeteer-chrome:22
COPY /home/myuser/dist ./dist
COPY package*.json ./
RUN npm --quiet set progress=false \ && npm install --omit=dev --omit=optional \ && 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 . ./
ENV APIFY_DISABLE_OUTDATED_WARNING=1
CMD ./start_xvfb_and_run_cmd.sh && npm run start:prod --silent{ "name": "actor-puppeteer-scraper", "version": "3.1.0", "private": true, "description": "Crawl web pages using Apify, headless Chrome and Puppeteer", "type": "module", "dependencies": { "@apify/scraper-tools": "^1.1.4", "@crawlee/puppeteer": "^3.14.1", "apify": "^3.2.6", "idcac-playwright": "^0.1.3", "puppeteer": "*" }, "devDependencies": { "@apify/tsconfig": "^0.1.0", "@types/node": "^22.7.4", "tsx": "^4.19.1", "typescript": "~5.8.0" }, "scripts": { "start": "npm run start:dev", "start:prod": "node dist/main.js", "start:dev": "tsx src/main.ts", "build": "tsc" }, "repository": { "type": "git", "url": "https://github.com/apify/apify-sdk-js" }, "author": { "name": "Apify Technologies", "email": "support@apify.com", "url": "https://apify.com" }, "contributors": [ "Marek Trunkat <marek@apify.com>", "Ondra Urban <ondra@apify.com>" ], "license": "Apache-2.0", "homepage": "https://github.com/apify/apify-sdk-js"}1# Change Log2
3## 3.0.12 (2024-10-25)4
5- Updated Crawlee version to v3.11.5 and SDK v3.2.66- Updated Node to v227
8## 3.0.11 (2024-04-09)9
10- Updated Crawlee version to v3.8.0.11- Updated to use new request queue in scraper12
13## 3.0.8 (2023-08-22)14
15- Updated Crawlee version to v3.5.2.16- Updated Node.js version to v18.17- Added new options:18 - **Dismiss cookie modals** (`closeCookieModals`): Using the [I don't care about cookies](https://addons.mozilla.org/en-US/firefox/addon/i-dont-care-about-cookies/) browser extension. When on, the crawler will automatically try to dismiss cookie consent modals. This can be useful when crawling European websites that show cookie consent modals.19 - **Maximum scrolling distance in pixels** (`maxScrollHeightPixels`): The crawler will scroll down the page until all content is loaded or the maximum scrolling distance is reached. Setting this to `0` disables scrolling altogether.20 - **Exclude Glob Patterns** (`excludes`): Glob patterns to match links in the page that you want to exclude from being enqueued.21
22## 3.0 (`version-3`)23
24- Rewrite from Apify SDK to Crawlee, see the [v3 migration guide](https://sdk.apify.com/docs/upgrading/upgrading-to-v3) for more details.25- Proxy usage is now required.26
27## 2.0 (`version-2`)28
29Main difference between v1 and v2 of the scrapers is the upgrade of SDK to v2, which requires node v15.10+. SDK v2 uses http2 to do the requests with cheerio-scraper, and the http2 support in older node versions were too buggy, so we decided to drop support for those. If you need to run on older node version, use SDK v1.30
31Please refer to the SDK 1.0 migration guide for more details about functional changes in the SDK. SDK v2 basically only changes the required node version and has no other breaking changes.32
33- deprecated `useRequestQueue` option has been removed34 - `RequestQueue` will be always used35- deprecated `context.html` getter from the `cheerio-scraper` has been removed36 - use `context.body` instead37- deprecated prepareRequestFunction input option38 - use `pre/postNavigationHooks` instead39- removed `puppeteerPool/autoscaledPool` from the crawlingContext object40 - `puppeteerPool` was replaced by `browserPool`41 - `autoscaledPool` and `browserPool` and available on the `crawler` property of `crawlingContext` object42- custom "Key-value store name" option in Advanced configuration is now fixed, previously the default store was always used# configurations.idea
# crawlee and apify storage foldersapify_storagecrawlee_storagestorage
# installed filesnode_modules1import { runActor } from '@apify/scraper-tools';2
3import { CrawlerSetup } from './internals/crawler_setup.js';4
5runActor(CrawlerSetup);{ "actorSpecification": 1, "name": "puppeteer-scraper", "version": "0.1", "buildTag": "latest", "storages": { "dataset": { "actorSpecification": 1, "fields": {}, "views": {} } }}1import type {2 GlobInput,3 ProxyConfigurationOptions,4 PseudoUrlInput,5 RegExpInput,6 RequestOptions,7 Session,8} from '@crawlee/puppeteer';9import type { Dictionary } from '@crawlee/utils';10import type { PuppeteerLifeCycleEvent } from 'puppeteer';11
12export const enum ProxyRotation {13 Recommended = 'RECOMMENDED',14 PerRequest = 'PER_REQUEST',15 UntilFailure = 'UNTIL_FAILURE',16}17
18/**19 * Replicates the INPUT_SCHEMA with TypeScript types for quick reference20 * and IDE type check integration.21 */22export interface Input {23 startUrls: RequestOptions[];24 globs: GlobInput[];25 regexps: RegExpInput[];26 pseudoUrls: PseudoUrlInput[];27 excludes: GlobInput[];28 linkSelector?: string;29 clickableElementsSelector?: string;30 keepUrlFragments: boolean;31 respectRobotsTxtFile: boolean;32 pageFunction: string;33 preNavigationHooks?: string;34 postNavigationHooks?: string;35 proxyConfiguration: ProxyConfigurationOptions;36 proxyRotation: ProxyRotation;37 sessionPoolName?: string;38 initialCookies: Parameters<Session['setCookies']>[0];39 useChrome: boolean;40 maxScrollHeightPixels: number;41 ignoreSslErrors: boolean;42 ignoreCorsAndCsp: boolean;43 closeCookieModals: boolean;44 downloadMedia: boolean;45 downloadCss: boolean;46 maxRequestRetries: number;47 maxPagesPerCrawl: number;48 maxResultsPerCrawl: number;49 maxCrawlingDepth: number;50 maxConcurrency: number;51 pageLoadTimeoutSecs: number;52 pageFunctionTimeoutSecs: number;53 waitUntil: PuppeteerLifeCycleEvent[];54 debugLog: boolean;55 browserLog: boolean;56 customData: Dictionary;57 datasetName?: string;58 keyValueStoreName?: string;59 requestQueueName?: string;60 headless: boolean;61}1import { readFile } from 'node:fs/promises';2import { dirname } from 'node:path';3import { fileURLToPath, URL } from 'node:url';4
5import type {6 AutoscaledPool,7 EnqueueLinksByClickingElementsOptions,8 EnqueueLinksOptions,9 ProxyConfiguration,10 PuppeteerCrawlerOptions,11 PuppeteerCrawlingContext,12 Request,13} from '@crawlee/puppeteer';14import {15 Dataset,16 KeyValueStore,17 log,18 PuppeteerCrawler,19 RequestList,20 RequestQueueV2,21} from '@crawlee/puppeteer';22import type { Awaitable, Dictionary } from '@crawlee/utils';23import { sleep } from '@crawlee/utils';24import type { ApifyEnv } from 'apify';25import { Actor } from 'apify';26import { getInjectableScript } from 'idcac-playwright';27import type { HTTPResponse } from 'puppeteer';28
29import type {30 CrawlerSetupOptions,31 RequestMetadata,32} from '@apify/scraper-tools';33import {34 browserTools,35 constants as scraperToolsConstants,36 createContext,37 tools,38} from '@apify/scraper-tools';39
40import type { Input } from './consts.js';41import { ProxyRotation } from './consts.js';42
43const SESSION_STORE_NAME = 'APIFY-PUPPETEER-SCRAPER-SESSION-STORE';44const REQUEST_QUEUE_INIT_FLAG_KEY = 'REQUEST_QUEUE_INITIALIZED';45
46const {47 META_KEY,48 DEFAULT_VIEWPORT,49 DEVTOOLS_TIMEOUT_SECS,50 SESSION_MAX_USAGE_COUNTS,51} = scraperToolsConstants;52const SCHEMA = JSON.parse(53 await readFile(new URL('../../INPUT_SCHEMA.json', import.meta.url), 'utf8'),54);55
56/**57 * Holds all the information necessary for constructing a crawler58 * instance and creating a context for a pageFunction invocation.59 */60export class CrawlerSetup implements CrawlerSetupOptions {61 name = 'Puppeteer Scraper';62 rawInput: string;63 env: ApifyEnv;64 /**65 * Used to store data that persist navigations66 */67 globalStore = new Map();68 requestQueue: RequestQueueV2;69 keyValueStore: KeyValueStore;70 customData: unknown;71 input: Input;72 maxSessionUsageCount: number;73 evaledPageFunction: (...args: unknown[]) => unknown;74 evaledPreNavigationHooks: ((...args: unknown[]) => Awaitable<void>)[];75 evaledPostNavigationHooks: ((...args: unknown[]) => Awaitable<void>)[];76 blockedUrlPatterns: string[] = [];77 devtools: boolean;78 datasetName?: string;79 keyValueStoreName?: string;80 requestQueueName?: string;81
82 crawler!: PuppeteerCrawler;83 dataset!: Dataset;84 pagesOutputted!: number;85 private initPromise: Promise<void>;86
87 constructor(input: Input) {88 // Set log level early to prevent missed messages.89 if (input.debugLog) log.setLevel(log.LEVELS.DEBUG);90
91 // Keep this as string to be immutable.92 this.rawInput = JSON.stringify(input);93
94 // Attempt to load page function from disk if not present on input.95 tools.maybeLoadPageFunctionFromDisk(96 input,97 dirname(fileURLToPath(import.meta.url)),98 );99
100 // Validate INPUT if not running on Apify Cloud Platform.101 if (!Actor.isAtHome()) tools.checkInputOrThrow(input, SCHEMA);102
103 this.input = input;104 this.env = Actor.getEnv();105
106 // Validations107 this.input.pseudoUrls.forEach((purl) => {108 if (!tools.isPlainObject(purl)) {109 throw new Error(110 'The pseudoUrls Array must only contain Objects.',111 );112 }113 if (purl.userData && !tools.isPlainObject(purl.userData)) {114 throw new Error(115 'The userData property of a pseudoUrl must be an Object.',116 );117 }118 });119
120 this.input.initialCookies.forEach((cookie) => {121 if (!tools.isPlainObject(cookie)) {122 throw new Error(123 'The initialCookies Array must only contain Objects.',124 );125 }126 });127
128 this.input.waitUntil.forEach((event) => {129 if (130 !/^(domcontentloaded|load|networkidle2|networkidle0)$/.test(131 event,132 )133 ) {134 throw new Error(135 'Navigation wait until events must be valid. See tooltip.',136 );137 }138 });139
140 // solving proxy rotation settings141 this.maxSessionUsageCount =142 SESSION_MAX_USAGE_COUNTS[this.input.proxyRotation];143
144 // Functions need to be evaluated.145 this.evaledPageFunction = tools.evalFunctionOrThrow(146 this.input.pageFunction,147 );148
149 if (this.input.preNavigationHooks) {150 this.evaledPreNavigationHooks = tools.evalFunctionArrayOrThrow(151 this.input.preNavigationHooks,152 'preNavigationHooks',153 );154 } else {155 this.evaledPreNavigationHooks = [];156 }157
158 if (this.input.postNavigationHooks) {159 this.evaledPostNavigationHooks = tools.evalFunctionArrayOrThrow(160 this.input.postNavigationHooks,161 'postNavigationHooks',162 );163 } else {164 this.evaledPostNavigationHooks = [];165 }166
167 // Excluded resources168 this.blockedUrlPatterns = [];169 if (!this.input.downloadMedia) {170 this.blockedUrlPatterns = [171 '.jpg',172 '.jpeg',173 '.png',174 '.svg',175 '.gif',176 '.webp',177 '.webm',178 '.ico',179 '.woff',180 '.eot',181 ];182 }183
184 if (!this.input.downloadCss) {185 this.blockedUrlPatterns.push('.css');186 }187
188 // Start Chromium with Debugger any time the page function includes the keyword.189 this.devtools = this.input.pageFunction.includes('debugger;');190
191 // Named storages192 this.datasetName = this.input.datasetName;193 this.keyValueStoreName = this.input.keyValueStoreName;194 this.requestQueueName = this.input.requestQueueName;195
196 // Initialize async operations.197 this.crawler = null!;198 this.requestQueue = null!;199 this.dataset = null!;200 this.keyValueStore = null!;201 this.initPromise = this._initializeAsync();202 }203
204 private async _initializeAsync() {205 // RequestList206 const startUrls = this.input.startUrls.map((req) => {207 req.useExtendedUniqueKey = true;208 req.keepUrlFragment = this.input.keepUrlFragments;209 return req;210 });211
212 // KeyValueStore213 this.keyValueStore = await KeyValueStore.open(this.keyValueStoreName);214
215 // RequestQueue216 this.requestQueue = await RequestQueueV2.open(this.requestQueueName);217
218 if (219 !(await this.keyValueStore.recordExists(220 REQUEST_QUEUE_INIT_FLAG_KEY,221 ))222 ) {223 const requests: Request[] = [];224 for await (const request of await RequestList.open(225 null,226 startUrls,227 )) {228 if (229 this.input.maxResultsPerCrawl > 0 &&230 requests.length >= 1.5 * this.input.maxResultsPerCrawl231 ) {232 break;233 }234 requests.push(request);235 }236
237 const { waitForAllRequestsToBeAdded } =238 await this.requestQueue.addRequestsBatched(requests);239
240 void waitForAllRequestsToBeAdded.then(async () => {241 await this.keyValueStore.setValue(242 REQUEST_QUEUE_INIT_FLAG_KEY,243 '1',244 );245 });246 }247
248 // Dataset249 this.dataset = await Dataset.open(this.datasetName);250 const info = await this.dataset.getInfo();251 this.pagesOutputted = info?.itemCount ?? 0;252 }253
254 /**255 * Resolves to a `PuppeteerCrawler` instance.256 */257 async createCrawler() {258 await this.initPromise;259
260 const args = [];261 if (this.input.ignoreCorsAndCsp) args.push('--disable-web-security');262
263 const options: PuppeteerCrawlerOptions = {264 requestHandler: this._requestHandler.bind(this),265 requestQueue: this.requestQueue,266 requestHandlerTimeoutSecs: this.devtools267 ? DEVTOOLS_TIMEOUT_SECS268 : this.input.pageFunctionTimeoutSecs,269 preNavigationHooks: [],270 postNavigationHooks: [],271 failedRequestHandler: this._failedRequestHandler.bind(this),272 respectRobotsTxtFile: this.input.respectRobotsTxtFile,273 maxConcurrency: this.input.maxConcurrency,274 maxRequestRetries: this.input.maxRequestRetries,275 maxRequestsPerCrawl:276 this.input.maxPagesPerCrawl === 0277 ? undefined278 : this.input.maxPagesPerCrawl,279 proxyConfiguration: (await Actor.createProxyConfiguration(280 this.input.proxyConfiguration,281 )) as any as ProxyConfiguration,282 launchContext: {283 useChrome: this.input.useChrome,284 launchOptions: {285 acceptInsecureCerts: this.input.ignoreSslErrors,286 defaultViewport: DEFAULT_VIEWPORT,287 devtools: this.devtools,288 args,289 headless: this.input.headless,290 },291 },292 useSessionPool: true,293 persistCookiesPerSession: true,294 sessionPoolOptions: {295 persistStateKeyValueStoreId: this.input.sessionPoolName296 ? SESSION_STORE_NAME297 : undefined,298 persistStateKey: this.input.sessionPoolName,299 sessionOptions: {300 maxUsageCount: this.maxSessionUsageCount,301 },302 },303 experiments: {304 requestLocking: true,305 },306 };307
308 this._createNavigationHooks(options);309
310 if (this.input.proxyRotation === ProxyRotation.UntilFailure) {311 options.sessionPoolOptions!.maxPoolSize = 1;312 }313
314 this.crawler = new PuppeteerCrawler(options);315
316 return this.crawler;317 }318
319 private _createNavigationHooks(options: PuppeteerCrawlerOptions) {320 options.preNavigationHooks!.push(321 async ({ request, page, session, blockRequests }, gotoOptions) => {322 // Attach a console listener to get all logs from Browser context.323 if (this.input.browserLog) browserTools.dumpConsole(page);324
325 // Prevent download of stylesheets and media, unless selected otherwise326 if (this.blockedUrlPatterns.length) {327 await blockRequests({328 urlPatterns: this.blockedUrlPatterns,329 });330 }331
332 // Add initial cookies, if any.333 if (334 this.input.initialCookies &&335 this.input.initialCookies.length336 ) {337 const cookiesToSet = session338 ? tools.getMissingCookiesFromSession(339 session,340 this.input.initialCookies,341 request.url,342 )343 : this.input.initialCookies;344
345 if (cookiesToSet?.length) {346 // setting initial cookies that are not already in the session and page347 session?.setCookies(cookiesToSet, request.url);348 await page.setCookie(...cookiesToSet);349 }350 }351
352 // Disable content security policy.353 if (this.input.ignoreCorsAndCsp) await page.setBypassCSP(true);354
355 if (gotoOptions) {356 gotoOptions.timeout =357 (this.devtools358 ? DEVTOOLS_TIMEOUT_SECS359 : this.input.pageLoadTimeoutSecs) * 1000;360 gotoOptions.waitUntil = this.input.waitUntil;361 }362 },363 );364
365 options.preNavigationHooks!.push(366 ...this._runHookWithEnhancedContext(this.evaledPreNavigationHooks),367 );368 options.postNavigationHooks!.push(369 ...this._runHookWithEnhancedContext(this.evaledPostNavigationHooks),370 );371 }372
373 private _runHookWithEnhancedContext(374 hooks: ((...args: unknown[]) => Awaitable<void>)[],375 ) {376 return hooks.map((hook) => (ctx: Dictionary, ...args: unknown[]) => {377 const { customData } = this.input;378 return hook({ ...ctx, Apify: Actor, Actor, customData }, ...args);379 });380 }381
382 private async _failedRequestHandler({ request }: PuppeteerCrawlingContext) {383 const lastError =384 request.errorMessages[request.errorMessages.length - 1];385 const errorMessage = lastError ? lastError.split('\n')[0] : 'no error';386 log.error(387 `Request ${request.url} failed and will not be retried anymore. Marking as failed.\nLast Error Message: ${errorMessage}`,388 );389 return this._handleResult(request, undefined, undefined, true);390 }391
392 /**393 * First of all, it initializes the state that is exposed to the user via394 * `pageFunction` context.395 *396 * Then it invokes the user provided `pageFunction` with the prescribed context397 * and saves its return value.398 *399 * Finally, it makes decisions based on the current state and post-processes400 * the data returned from the `pageFunction`.401 */402 private async _requestHandler(crawlingContext: PuppeteerCrawlingContext) {403 const { request, response, crawler } = crawlingContext;404
405 /**406 * PRE-PROCESSING407 */408 // Make sure that an object containing internal metadata409 // is present on every request.410 tools.ensureMetaData(request);411
412 // Abort the crawler if the maximum number of results was reached.413 const aborted = await this._handleMaxResultsPerCrawl(414 crawler.autoscaledPool,415 );416 if (aborted) return;417
418 const pageFunctionArguments: Dictionary = {};419
420 // We must use properties and descriptors not to trigger getters / setters.421 Object.defineProperties(422 pageFunctionArguments,423 Object.getOwnPropertyDescriptors(crawlingContext),424 );425
426 pageFunctionArguments.response = {427 status: response && response.status(),428 headers: response && response.headers(),429 };430
431 // Setup and create Context.432 const contextOptions = {433 crawlerSetup: {434 rawInput: this.rawInput,435 env: this.env,436 globalStore: this.globalStore,437 requestQueue: this.requestQueue,438 keyValueStore: this.keyValueStore,439 customData: this.input.customData,440 },441 pageFunctionArguments,442 };443 const { context, state } = createContext(contextOptions);444
445 if (this.input.closeCookieModals) {446 await sleep(500);447 await crawlingContext.page.evaluate(getInjectableScript());448 await sleep(2000);449 }450
451 if (this.input.maxScrollHeightPixels > 0) {452 await crawlingContext.infiniteScroll({453 maxScrollHeight: this.input.maxScrollHeightPixels,454 });455 }456
457 /**458 * USER FUNCTION INVOCATION459 */460 const pageFunctionResult = await this.evaledPageFunction(context);461
462 /**463 * POST-PROCESSING464 */465 // Enqueue more links if Pseudo URLs and a link selector are available,466 // unless the user invoked the `skipLinks()` context function467 // or maxCrawlingDepth would be exceeded.468 if (!state.skipLinks) await this._handleLinks(crawlingContext);469
470 // Save the `pageFunction`s result to the default dataset.471 await this._handleResult(472 request,473 response,474 pageFunctionResult as Dictionary,475 );476 }477
478 private async _handleMaxResultsPerCrawl(autoscaledPool?: AutoscaledPool) {479 if (480 !this.input.maxResultsPerCrawl ||481 this.pagesOutputted < this.input.maxResultsPerCrawl482 )483 return false;484 if (!autoscaledPool) return false;485 log.info(486 `User set limit of ${this.input.maxResultsPerCrawl} results was reached. Finishing the crawl.`,487 );488 await autoscaledPool.abort();489 return true;490 }491
492 private async _handleLinks({493 request,494 enqueueLinks,495 enqueueLinksByClickingElements,496 }: PuppeteerCrawlingContext) {497 if (!this.requestQueue) return;498 const currentDepth = (request.userData![META_KEY] as RequestMetadata)499 .depth;500 const hasReachedMaxDepth =501 this.input.maxCrawlingDepth &&502 currentDepth >= this.input.maxCrawlingDepth;503 if (hasReachedMaxDepth) {504 log.debug(505 `Request ${request.url} reached the maximum crawling depth of ${currentDepth}.`,506 );507 return;508 }509
510 const enqueueOptions: EnqueueLinksOptions = {511 globs: this.input.globs,512 pseudoUrls: this.input.pseudoUrls,513 exclude: this.input.excludes,514 transformRequestFunction: (requestOptions) => {515 requestOptions.userData ??= {};516 requestOptions.userData[META_KEY] = {517 parentRequestId: request.id || request.uniqueKey,518 depth: currentDepth + 1,519 };520
521 requestOptions.useExtendedUniqueKey = true;522 requestOptions.keepUrlFragment = this.input.keepUrlFragments;523 return requestOptions;524 },525 };526
527 if (this.input.linkSelector) {528 await enqueueLinks({529 ...enqueueOptions,530 selector: this.input.linkSelector,531 });532 }533
534 if (this.input.clickableElementsSelector) {535 await enqueueLinksByClickingElements({536 ...enqueueOptions,537 selector: this.input.clickableElementsSelector,538 } as EnqueueLinksByClickingElementsOptions);539 }540 }541
542 private async _handleResult(543 request: Request,544 response?: HTTPResponse,545 pageFunctionResult?: Dictionary,546 isError?: boolean,547 ) {548 const payload = tools.createDatasetPayload(549 request,550 response,551 pageFunctionResult,552 isError,553 );554 await this.dataset.pushData(payload);555 this.pagesOutputted++;556 }557}