Python Selenium scraper

1"""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
9from __future__ import annotations
10
11import asyncio
12from urllib.parse import urljoin
13
14from apify import Actor, Request
15from selenium import webdriver
16from selenium.webdriver.chrome.options import Options as ChromeOptions
17from selenium.webdriver.common.by import By
18
19# To run this Actor locally, you need to have the Selenium Chromedriver installed.
20# Follow the installation guide at:
21# https://www.selenium.dev/documentation/webdriver/getting_started/install_drivers/
22# When running on the Apify platform, the Chromedriver is already included
23# in the Actor's Docker image.
24
25# Limit the crawl to max requests. Increase it to crawl more links.
26MAX_REQUESTS_PER_CRAWL = 10
27
28# Matches Playwright's default navigation timeout, which the sibling templates rely on.
29PAGE_LOAD_TIMEOUT_SECS = 30
30
31
32async def main() -> None:
33    """Define a main entry point for the Apify Actor.
34
35    This coroutine is executed using `asyncio.run()`, so it must remain an asynchronous function for proper execution.
36    Asynchronous execution is required for communication with Apify platform, and it also enhances performance in
37    the field of web scraping significantly.
38    """
39    # Enter the context of the Actor.
40    async with Actor:
41        # Retrieve the Actor input, and use default values if not provided.
42        actor_input = await Actor.get_input() or {}
43        start_urls = actor_input.get('start_urls', [{'url': 'https://apify.com'}])
44        max_depth = actor_input.get('max_depth', 1)
45
46        # Exit if no start URLs are provided.
47        if not start_urls:
48            Actor.log.info('No start URLs specified in Actor input, exiting...')
49            await Actor.exit()
50
51        # Open the default request queue for handling URLs to be processed.
52        request_queue = await Actor.open_request_queue()
53
54        # Enqueue the start URLs with an initial crawl depth of 0.
55        for start_url in start_urls:
56            url = start_url.get('url')
57            Actor.log.info(f'Enqueuing {url} ...')
58            new_request = Request.from_url(url, user_data={'depth': 0})
59            await request_queue.add_request(new_request)
60
61        # Launch a new Selenium Chrome WebDriver and configure it.
62        Actor.log.info('Launching Chrome WebDriver...')
63        chrome_options = ChromeOptions()
64
65        if Actor.configuration.headless:
66            chrome_options.add_argument('--headless')
67
68        chrome_options.add_argument('--no-sandbox')
69        chrome_options.add_argument('--disable-dev-shm-usage')
70        driver = webdriver.Chrome(options=chrome_options)
71
72        # Without this a page that never finishes loading blocks for 120s, the client timeout.
73        driver.set_page_load_timeout(PAGE_LOAD_TIMEOUT_SECS)
74
75        # Test WebDriver setup by navigating to an example page.
76        driver.get('http://www.example.com')
77        if driver.title != 'Example Domain':
78            raise ValueError('Failed to open example page.')
79
80        handled_requests = 0
81
82        # Process the URLs from the request queue.
83        while handled_requests < MAX_REQUESTS_PER_CRAWL and (request := await request_queue.fetch_next_request()):
84            url = request.url
85
86            if not isinstance(request.user_data['depth'], (str, int)):
87                raise TypeError('Request.depth is an enexpected type.')
88
89            depth = int(request.user_data['depth'])
90            Actor.log.info(f'Scraping {url} (depth={depth}) ...')
91
92            try:
93                # Navigate to the URL using Selenium WebDriver. Use asyncio.to_thread
94                # for non-blocking execution.
95                await asyncio.to_thread(driver.get, url)
96
97                # If the current depth is less than max_depth, find nested links
98                # and enqueue them.
99                if depth < max_depth:
100                    for link in driver.find_elements(By.TAG_NAME, 'a'):
101                        link_href = link.get_attribute('href')
102                        link_url = urljoin(url, link_href)
103
104                        if link_url.startswith(('http://', 'https://')):
105                            Actor.log.info(f'Enqueuing {link_url} ...')
106                            new_request = Request.from_url(
107                                link_url,
108                                user_data={'depth': depth + 1},
109                            )
110                            await request_queue.add_request(new_request)
111
112                # Extract the desired data.
113                data = {
114                    'url': url,
115                    'title': driver.title,
116                }
117
118                # Store the extracted data to the default dataset.
119                await Actor.push_data(data)
120
121            except Exception:
122                Actor.log.exception(f'Cannot extract data from {url}.')
123
124            finally:
125                # Mark the request as handled to ensure it is not processed again.
126                await request_queue.mark_request_as_handled(request)
127                handled_requests += 1
128
129        driver.quit()