YouTube Comment Scraper

YouTubeCommentScraper is a Python package designed to scrape comments from YouTube videos using Selenium. The scraper is customizable, allowing you to run the browser in headless mode, control the timeout, pause time for scrolling, and more. You can also choose whether to log actions and return the page source along with the comments.

Features

• Headless Mode: Run the browser in headless mode (optional).
• Customizable Timeouts: Set the timeout for waiting for elements to load.
• Automatic Scrolling: Automatically scrolls the page until all comments are loaded.
• Logging Support: Enable logging to a file for tracking activities.
• Return Page Source: Optionally return the page source along with the comments.
• BeautifulSoup Integration: Extract comments using BeautifulSoup for robust parsing.

Installation

To install the package, use the following command:

pip install youtube-comments-scrapper

Dependencies

This package requires the following dependencies:

• selenium
• webdriver-manager
• beautifulsoup4
• lxml (optional but recommended for faster HTML parsing)

You can install these dependencies using the following command (optional):

pip install selenium webdriver-manager beautifulsoup4 lxml

Usage

1. Basic Usage: Scraping Comments

Here's a simple example to scrape comments from a YouTube video:

from youtube_comments_scraper import YouTubeCommentScraper

scraper = YouTubeCommentScraper(headless=True, timeout=10, scroll_pause_time=1.5, enable_logging=False, return_page_source=False)
video_url = "https://www.youtube.com/watch?v=Ycg48pVp3SU"
comments = scraper.scrape_comments(video_url)

print("Comments:", comments)

2. Scraping Comments with Logging Enabled

Enable logging to track the actions performed during scraping:

from youtube_comments_scraper import YouTubeCommentScraper

scraper = YouTubeCommentScraper(headless=True, timeout=10, scroll_pause_time=1.5, enable_logging=True, return_page_source=False)
video_url = "https://www.youtube.com/watch?v=Ycg48pVp3SU"
comments = scraper.scrape_comments(video_url)

print("Comments:", comments)

This will generate a log file (youtube_scraper.log) in the current directory.

3. Returning Page Source Along with Comments

If you want to extract comments and return the page's HTML source:

from youtube_comments_scraper import YouTubeCommentScraper

scraper = YouTubeCommentScraper(headless=True, timeout=10, scroll_pause_time=1.5, enable_logging=False, return_page_source=True)
video_url = "https://www.youtube.com/watch?v=Ycg48pVp3SU"
comments, page_source = scraper.scrape_comments(video_url)

print("Comments:", comments)
print("Page Source:", page_source)

4. Custom Scroll Pause Time

You can control how long the scraper pauses between scroll actions using the scroll_pause_time parameter:

from youtube_comments_scraper import YouTubeCommentScraper

scraper = YouTubeCommentScraper(headless=True, timeout=10, scroll_pause_time=2.0, enable_logging=False, return_page_source=False)
video_url = "https://www.youtube.com/watch?v=Ycg48pVp3SU"
comments = scraper.scrape_comments(video_url)

print("Comments:", comments)

5. Scraping Comments Without Scrolling

If you only want to scrape the comments that load without scrolling:

from youtube_comments_scraper import YouTubeCommentScraper

scraper = YouTubeCommentScraper(headless=True, timeout=10, scroll_pause_time=1.5, enable_logging=False, return_page_source=False)
video_url = "https://www.youtube.com/watch?v=Ycg48pVp3SU"
comments = scraper.scrape_comments(video_url, scroll=False)

print("Comments:", comments)

6. Logging Custom Messages

You can log custom messages using the built-in log_info, log_warning, and log_error methods:

scraper.log_info("This is an info log message.")
scraper.log_warning("This is a warning message.")
scraper.log_error("This is an error message.")

Class Reference

YouTubeCommentScraper

__init__(self, headless=True, timeout=10, scroll_pause_time=1.5, enable_logging=False, return_page_source=False)

• headless (bool): Run the browser in headless mode. Default is True.
• timeout (int): The maximum time to wait for elements to load. Default is 10 seconds.
• scroll_pause_time (float): The pause time between scroll actions. Default is 1.5 seconds.
• enable_logging (bool): Whether to enable logging to a file. Default is False.
• return_page_source (bool): Whether to return the page source along with comments. Default is False.

scrape_comments(self, video_url, scroll=True)

• video_url (str): The URL of the YouTube video.
• scroll (bool): Whether to scroll the page to load all comments. Default is True.

Returns:

• A tuple (comments, page_source) if return_page_source is True, otherwise just the list of comments.

log_info(self, message)

• Logs an informational message.

log_warning(self, message)

• Logs a warning message.

log_error(self, message)

• Logs an error message.

License

This project is licensed under the MIT License.