web-scraping-with-python VS puppeteer

import asyncio import pytest from pyppeteer.errors import PageError from urllib.parse import quote import json import os import sys from os import environ from pyppeteer import connect, launch exec_platform = os.getenv('EXEC_PLATFORM') # Get username and access key of the LambdaTest Platform username = environ.get('LT_USERNAME', None) access_key = environ.get('LT_ACCESS_KEY', None) test1_url = 'https://ecommerce-playground.lambdatest.io/' test2_url = 'https://scrapingclub.com/exercise/list_infinite_scroll/' # Usecase - 1 # loc_ecomm_1 = ".order-1.col-lg-6 div:nth-of-type(1) > div:nth-of-type(1) > div:nth-of-type(1) > div:nth-of-type(1) > div:nth-of-type(1) div:nth-of-type(1) > img:nth-of-type(1)" loc_ecomm_1 = "[aria-label='1 / 2'] div:nth-of-type(1) > [alt='Nikon D300']" target_url_1 = "https://ecommerce-playground.lambdatest.io/index.php?route=product/product&product_id=63" # Usecase - 2 (Click on e-commerce sliding banner) loc_ecomm_2 = "[alt='Canon DSLR camera']" target_url_2 = "https://ecommerce-playground.lambdatest.io/index.php?route=product/product&product_id=30" # Usecase - 3 Automating interactions on https://scrapingclub.com/exercise/list_infinite_scroll/ loc_infinite_src_prod1 = ".grid .p-4 [href='/exercise/list_basic_detail/93926-C/']" target_url_3 = "https://scrapingclub.com/exercise/list_basic_detail/93926-C/" # Usecase - 4 Automating interactions on https://scrapingclub.com/exercise/list_infinite_scroll/ # when the images are lazy loaded loc_infinite_src_prod2 = "div:nth-of-type(31) > .p-4 [href='/exercise/list_basic_detail/94967-A/']" target_url_4 = "https://scrapingclub.com/exercise/list_basic_detail/94967-A/" # Set timeout in ms timeOut = 60000 async def scroll_to_element(page, selector): # Scroll until the element is detected await page.evaluateHandle( '''async (selector) => { const element = document.querySelector(selector); if (element) { element.scrollIntoView(); } }''', selector ) return selector async def scroll_carousel(page, scr_count): for scr in range(1, scr_count): elem_next_button = "#mz-carousel-213240 > ul li:nth-child(" + str(scr) + ")" await asyncio.sleep(1) elem_next_button = await page.querySelector(elem_next_button) await elem_next_button.click() # Replica of https://github.com/hjsblogger/web-scraping-with-python/blob/ # main/tests/beautiful-soup/test_infinite_scraping.py#L67C5-L80C18 async def scroll_end_of_page(page): start_height = await page.evaluate('document.documentElement.scrollHeight') while True: # Scroll to the bottom of the page await page.evaluate(f'window.scrollTo(0, {start_height})') # Wait for the content to load await asyncio.sleep(1) # Get the new scroll height scroll_height = await page.evaluate('document.documentElement.scrollHeight') if scroll_height == start_height: # If heights are the same, we reached the end of the page break # Add an additional wait await asyncio.sleep(2) start_height = scroll_height # Additional wait after scrolling await asyncio.sleep(2) @pytest.mark.asyncio @pytest.mark.order(1) async def test_lazy_load_ecomm_1(page): # The time out can be set using the setDefaultNavigationTimeout # It is primarily used for overriding the default page timeout of 30 seconds page.setDefaultNavigationTimeout(timeOut) await page.goto(test1_url, {'waitUntil': 'load', 'timeout': timeOut}) # Set the viewport - Apple MacBook Air 13-inch # Reference - https://codekbyte.com/devices-viewport-sizes/ # await page.setViewport({'width': 1440, 'height': 770}) await asyncio.sleep(2) if exec_platform == 'local': # Scroll until the element is detected elem_button = await scroll_to_element(page, loc_ecomm_1) # await page.click(elem_button) # Wait until the page is loaded # https://miyakogi.github.io/pyppeteer/reference.html#pyppeteer.page.Page.waitForNavigation navigationPromise = asyncio.ensure_future(page.waitForNavigation()) await page.click(elem_button) await navigationPromise elif exec_platform == 'cloud': elem_button = await page.waitForSelector(loc_ecomm_1, {'visible': True}) await asyncio.gather( elem_button.click(), page.waitForNavigation({'waitUntil': 'networkidle2', 'timeout': 30000}), ) # Assert if required, since the test is a simple one; we leave as is :D current_url = page.url print('Current URL is: ' + current_url) try: assert current_url == target_url_1 print("Test Success: Product checkout successful") except PageError as e: print("Test Failure: Could not checkout Product") print("Error Code" + str(e)) @pytest.mark.asyncio @pytest.mark.order(2) async def test_lazy_load_ecomm_2(page): carousel_len = 4 # The time out can be set using the setDefaultNavigationTimeout # It is primarily used for overriding the default page timeout of 30 seconds page.setDefaultNavigationTimeout(timeOut) await page.goto(test1_url, {'waitUntil': 'load', 'timeout': timeOut}) # Set the viewport - Apple MacBook Air 13-inch # Reference - https://codekbyte.com/devices-viewport-sizes/ # await page.setViewport({'width': 1440, 'height': 770}) await asyncio.sleep(2) # Approach 1: Directly click on the third button on the carousel # elem_carousel_banner = await page.querySelector("#mz-carousel-213240 > ul li:nth-child(3)") # await asyncio.sleep(1) # await elem_carousel_banner.click() # Approach 2 (Only for demo): Serially click on every button on carousel await scroll_carousel(page, carousel_len) await asyncio.sleep(1) # elem_prod_1 = await page.querySelector(loc_ecomm_2) elem_prod_1 = await page.waitForSelector(loc_ecomm_2, {'visible': True}) await asyncio.gather( elem_prod_1.click(), page.waitForNavigation({'waitUntil': 'networkidle2', 'timeout': 60000}), ) # Assert if required, since the test is a simple one; we leave as is :D current_url = page.url print('Current URL is: ' + current_url) try: assert current_url == target_url_2 print("Test Success: Product checkout successful") except PageError as e: print("Test Failure: Could not checkout Product") print("Error Code" + str(e)) @pytest.mark.asyncio @pytest.mark.order(3) async def test_lazy_load_infinite_scroll_1(page): # The time out can be set using the setDefaultNavigationTimeout # It is primarily used for overriding the default page timeout of 30 seconds page.setDefaultNavigationTimeout(timeOut) await page.goto(test2_url, {'waitUntil': 'load', 'timeout': timeOut}) # Set the viewport - Apple MacBook Air 13-inch # Reference - https://codekbyte.com/devices-viewport-sizes/ # await page.setViewport({'width': 1440, 'height': 770}) await asyncio.sleep(1) elem_prod1 = await page.querySelector(loc_infinite_src_prod1) await asyncio.gather( elem_prod1.click(), page.waitForNavigation({'waitUntil': 'networkidle2', 'timeout': 60000}), ) # await asyncio.sleep(1) # await elem_carousel_banner.click() # elem_button = scroll_to_element(page, loc_infinite_src_prod1) # print(elem_button) # await asyncio.sleep(2) # await elem_button.click() # Assert if required, since the test is a simple one; we leave as is :D current_url = page.url print('Current URL is: ' + current_url) try: assert current_url == target_url_3 print("Test Success: Product checkout successful") except PageError as e: print("Test Failure: Could not checkout Product") print("Error Code" + str(e)) @pytest.mark.asyncio @pytest.mark.order(4) async def test_lazy_load_infinite_scroll_2(page): # The time out can be set using the setDefaultNavigationTimeout # It is primarily used for overriding the default page timeout of 30 seconds page.setDefaultNavigationTimeout(timeOut) # Tested navigation using LambdaTest YouTube channel # await page.goto("https://www.youtube.com/@LambdaTest/videos", await page.goto(test2_url, {'waitUntil': 'load', 'timeout': timeOut}) # Set the viewport - Apple MacBook Air 13-inch # Reference - https://codekbyte.com/devices-viewport-sizes/ # await page.setViewport({'width': 1440, 'height': 770}) await asyncio.sleep(1) await scroll_end_of_page(page) await page.evaluate('window.scrollTo(0, 0)') await asyncio.sleep(1) # elem_prod = await page.querySelector(loc_infinite_src_prod2) # asyncio.sleep(1) # await asyncio.gather( # elem_prod.click(), # page.waitForNavigation({'waitUntil': 'load', 'timeout': 60000}), # ) elem_button = await scroll_to_element(page, loc_infinite_src_prod2) await asyncio.sleep(1) # await page.click(elem_button) await asyncio.gather( page.click(elem_button), page.waitForNavigation({'waitUntil': 'networkidle2', 'timeout': 60000}), ) # Assert if required, since the test is a simple one; we leave as is :D current_url = page.url print('Current URL is: ' + current_url) try: assert current_url == target_url_4 print("Test Success: Product checkout successful") except PageError as e: print("Test Failure: Could not checkout Product") print("Error Code" + str(e))

The helpers.py file contains the core implementation for scraping the LambdaTest YouTube channel.

Use Selenium or Third-Party APIs: Utilize tools like Selenium / Puppeteer or third-party APIs such as the Crawlbase Crawling API to load the webpage. These tools can handle JavaScript rendering, ensuring all dynamic content is displayed.

This project tests how the browser language can be changed with Puppeteer. It implements multiple options to set the language of Chrome and checks each option against BrowserLeaks to see how it affected the JavaScript proeprties and HTTP headers available by the browser. For more information, see my article The Puppeteer Language Experiment on DEV.to.

In Crawlee, you can scrape JavaScript rendered websites using the built-in headless Puppeteer and Playwright browsers. It is important to note that, by default, Crawlee scrapes in headless mode. If you don't want headless, then just set headless: false.

Some quick docs fumbling and you get a working script for loading the page and grabbing the innerHTML of the listing element like so:

I am not in any way associated with the developers at puppeteer, but if you are looking for a way to contribute, they are open source

With the advent of tools like Puppeteer and now Playwright, end-to-end testing has become much easier and more reliable. For anyone who's used Selenium in the past, you know what I'm talking about. Puppeteer has opened the way in terms of E2E tooling, but Playwright has taken it to the next level and made it easier to await for certain selectors or conditions to be fulfilled (via locators), thus making tests more reliable and less flaky. Also, it's a game changer that it introduced a test-runner - this made the integration between the headless browser and the actual test code much smoother.

1.Puppeteer: Puppeteer is a Node library that provides a high-level API to control headless Chrome or Chromium using the DevTools Protocol. Key Features: More control over Chrome. Enables web scraping. Allows taking screenshots and generating PDFs for UI testing. Measures load times through the Chrome Performance Analysis tool

HTML to PDF conversion is a common requirement in modern web applications. It allows users to save web pages, reports, and other content in a format that is easy to share and print. There are many libraries and services available for converting HTML to PDF, each with its own strengths and weaknesses. In this article, we will compare some of the most popular HTML to PDF renderers in Node.js, including Puppeteer, Playwright, node-html-pdf, and Onedoc.

Playwright seems to be a superior library for working with headless browsers than Puppeteer, but I will go with Puppeteer.

Puppeteer is a Node library that provides a high-level API to control headless Chrome or Chromium. It's primarily used for browser automation, making it a powerful tool for end-to-end testing of web applications, taking screenshots, and generating pre-rendered content from web pages.