Skip to content

ispras/scrapy-puppeteer

Scrapy-puppeteer-client

This package aims to manage Chrome browser with Puppeteer from Scrapy spiders. This allows to scrape sites that require JS to function properly and to make the scraper more similar to humans. It is a client library for scrapy-puppeteer-service.

⚠️ This repository is under development.

This project is under development. Use it at your own risk.

Installation

Using pip (master branch):

$ pip install scrapy-puppeteer-client

Configuration

You should have scrapy-puppeteer-service started. Then add its URL to settings.py and enable puppeteer downloader middleware:

DOWNLOADER_MIDDLEWARES = {
    'scrapypuppeteer.middleware.PuppeteerServiceDownloaderMiddleware': 1042
}

PUPPETEER_SERVICE_URL = "http://localhost:3000"  # Not necessary in other execution methods

# To change the execution method, you must add the corresponding setting:
EXECUTION_METHOD = "Puppeteer"

Available methods: Puppeteer, Pyppeteer, Playwright

Pyppeteer and Playwright methods do not require a running service. They use the pyppeteer and playwright libraries for Python to interact with the browser. Actions such as CustomJsAction, RecaptchaSolver, and Har are not available when using these methods.

To use Pyppeteer or Playwright methods you need to install Chromium.

Basic usage

Use scrapypuppeteer.PuppeteerRequest instead of scrapy.Request to render URLs with Puppeteer:

import scrapy
from scrapypuppeteer import PuppeteerRequest

class MySpider(scrapy.Spider):
    ...
    def start_requests(self):
        yield PuppeteerRequest('https://exapmle.com', callback=self.parse)
    
    def parse(self, response):
        links = response.css(...)
        ...

Puppeter responses

There is a parent PuppeteerResponse class from which other response classes are inherited.

Here is a list of them all:

  • PuppeteerHtmlResponse - has html and cookies properties
  • PuppeteerScreenshotResponse - has screenshot property
  • PuppeteerHarResponse - has har property
  • PuppeteerJsonResponse - has data property and to_html() method which tries to transform itself to PuppeteerHtmlResponse
  • PuppeteerRecaptchaSolverResponse(PuppeteerJsonResponse, PuppeteerHtmlResponse) - has recaptcha_data property

Advanced usage

PuppeteerRequest's first argument is a browser action. Available actions are defined in scrapypuppeteer.actions module as subclasses of PuppeteerServiceAction. Passing a URL into request is a shortcut for GoTo(url) action.

Here is the list of available actions:

  • GoTo(url, options) - navigate to URL
  • GoForward(options) - navigate forward in history
  • GoBack(options) - navigate back in history
  • Click(selector, click_options, wait_options) - click on element on page
  • Compose(*actions) - composition of several puppeteer action
  • Scroll(selector, wait_options) - scroll page
  • Screenshot(options) - take screenshot
  • Har() - to get the HAR file, pass the har_recording=True argument to PuppeteerRequest at the start of execution.
  • FillForm(input_mapping, submit_button) - to fill out and submit forms on page.
  • RecaptchaSolver(solve_recaptcha) - find or solve recaptcha on page
  • CustomJsAction(js_function) - evaluate JS function on page

Available options essentially mirror service method parameters, which in turn mirror puppeteer API functions to some extent. See scrapypuppeteer.actions module for details.

You may pass close_page=False option to a request to retain browser tab and its state after request's completion. Then use response.follow to continue interacting with the same tab:

import scrapy
from scrapypuppeteer import PuppeteerRequest, PuppeteerHtmlResponse
from scrapypuppeteer.actions import Click

class MySpider(scrapy.Spider):
    ...
    def start_requests(self):
        yield PuppeteerRequest(
            'https://exapmle.com',  # will be transformed into GoTo action
            close_page=False,
            callback=self.parse,
        )

    def parse(self, response: PuppeteerHtmlResponse):
        ...
        # parse and yield some items
        ...
        next_page_selector = 'button.next-page-or-smth'
        if response.css(next_page_selector ):
            yield response.follow(
                Click(
                    next_page_selector,
                    wait_options={'selectorOrTimeout': 3000},  # wait 3 seconds
                ),
                close_page=False,
                callback=self.parse,
            )

You may also use follow_all method to continue interacting.

On your first request service will create new incognito browser context and new page in it. Their ids will be in returned in response object as context_id and page_id attributes. Following such response means passing context and page ids to next request. You also may specify requests context and page ids directly.

Right before your spider has done the crawling, the service middleware will take care of closing all used browser contexts with scrapypuppeteer.CloseContextRequest. It accepts a list of all browser contexts to be closed.

One may customize which PuppeteerRequest's headers will be sent to remote website by the service via include_headers attribute in request or globally with PUPPETEER_INCLUDE_HEADERS setting. Available values are True (all headers), False (no headers) or list of header names. By default, only cookies are sent.

You would also like to send meta with your request. By default, you are not allowed to do this in order to sustain backward compatibility. You can change this behaviour by setting PUPPETEER_INCLUDE_META to True.

Automatic recaptcha solving

Enable PuppeteerRecaptchaDownloaderMiddleware to automatically solve recaptcha during scraping. We do not recommend to use RecaptchaSolver action when the middleware works.

DOWNLOADER_MIDDLEWARES = {
    'scrapypuppeteer.middleware.PuppeteerRecaptchaDownloaderMiddleware': 1041,
    'scrapypuppeteer.middleware.PuppeteerServiceDownloaderMiddleware': 1042
}

Note that the number of RecaptchaMiddleware has to be lower than ServiceMiddleware's. You must provide some settings to use the middleware:

PUPPETEER_INCLUDE_META = True  # Essential to send meta

RECAPTCHA_ACTIVATION = True  # Enables the middleware
RECAPTCHA_SOLVING = False  # Automatic recaptcha solving
RECAPTCHA_SUBMIT_SELECTORS = {  # Selectors for "submit recaptcha" button
    'www.google.com/recaptcha/api2/demo': '',  # No selectors needed
}

If you set RECAPTCHA_SOLVING to False the middleware will try to find captcha and will notify you about number of found captchas on the page.

If you don't want the middleware to work on specific request you may provide special meta key: 'dont_recaptcha': True. In this case RecaptchaMiddleware will just skip the request.

TODO

  • skeleton that could handle goto, click, scroll, and actions
  • headers and cookies management
  • proxy support for puppeteer
  • error handling for requests
  • har support