Selenium + Chrome

1"""This module defines the main entry point for the Apify Actor.
2
3Feel free to modify this file to suit your specific needs.
4
5To build Apify Actors, utilize the Apify SDK toolkit, read more at the official documentation:
6https://docs.apify.com/sdk/python
7"""
8
9import asyncio
10from urllib.parse import urljoin
11
12from selenium import webdriver
13from selenium.webdriver.chrome.options import Options as ChromeOptions
14from selenium.webdriver.common.by import By
15
16from apify import Actor, Request
17
18# To run this Actor locally, you need to have the Selenium Chromedriver installed.
19# Follow the installation guide at: https://www.selenium.dev/documentation/webdriver/getting_started/install_drivers/
20# When running on the Apify platform, the Chromedriver is already included in the Actor's Docker image.
21
22
23async def main() -> None:
24    """Main entry point for the Apify Actor.
25
26    This coroutine is executed using `asyncio.run()`, so it must remain an asynchronous function for proper execution.
27    Asynchronous execution is required for communication with Apify platform, and it also enhances performance in
28    the field of web scraping significantly.
29    """
30    async with Actor:
31        # Retrieve the Actor input, and use default values if not provided.
32        actor_input = await Actor.get_input() or {}
33        start_urls = actor_input.get('start_urls', [{'url': 'https://apify.com'}])
34        max_depth = actor_input.get('max_depth', 1)
35
36        # Exit if no start URLs are provided.
37        if not start_urls:
38            Actor.log.info('No start URLs specified in actor input, exiting...')
39            await Actor.exit()
40
41        # Open the default request queue for handling URLs to be processed.
42        request_queue = await Actor.open_request_queue()
43
44        # Enqueue the start URLs with an initial crawl depth of 0.
45        for start_url in start_urls:
46            url = start_url.get('url')
47            Actor.log.info(f'Enqueuing {url} ...')
48            request = Request.from_url(url, user_data={'depth': 0})
49            await request_queue.add_request(request)
50
51        # Launch a new Selenium Chrome WebDriver and configure it.
52        Actor.log.info('Launching Chrome WebDriver...')
53        chrome_options = ChromeOptions()
54
55        if Actor.config.headless:
56            chrome_options.add_argument('--headless')
57
58        chrome_options.add_argument('--no-sandbox')
59        chrome_options.add_argument('--disable-dev-shm-usage')
60        driver = webdriver.Chrome(options=chrome_options)
61
62        # Test WebDriver setup by navigating to an example page.
63        driver.get('http://www.example.com')
64        assert driver.title == 'Example Domain'
65
66        # Process the URLs from the request queue.
67        while request := await request_queue.fetch_next_request():
68            url = request.url
69            depth = request.user_data['depth']
70            Actor.log.info(f'Scraping {url} ...')
71
72            try:
73                # Navigate to the URL using Selenium WebDriver. Use asyncio.to_thread for non-blocking execution.
74                await asyncio.to_thread(driver.get, url)
75
76                # If the current depth is less than max_depth, find nested links and enqueue them.
77                if depth < max_depth:
78                    for link in driver.find_elements(By.TAG_NAME, 'a'):
79                        link_href = link.get_attribute('href')
80                        link_url = urljoin(url, link_href)
81
82                        if link_url.startswith(('http://', 'https://')):
83                            Actor.log.info(f'Enqueuing {link_url} ...')
84                            request = Request.from_url(link_url, user_data={'depth': depth + 1})
85                            await request_queue.add_request(request)
86
87                # Extract the desired data.
88                data = {
89                    'url': url,
90                    'title': driver.title,
91                }
92
93                # Store the extracted data to the default dataset.
94                await Actor.push_data(data)
95
96            except Exception:
97                Actor.log.exception(f'Cannot extract data from {url}.')
98
99            finally:
100                # Mark the request as handled to ensure it is not processed again.
101                await request_queue.mark_request_as_handled(request)
102
103        driver.quit()