CleanMyCSS is a command line python tool that crawls an entire website and returns clean versions of the css files, removing unused rules.
It is a script I developed many many years ago but that I still use to optimize the CSS from many of my legacy websites, including large CSS files like bootstrap.
- Automatic website crawler
- Downloads HTML and JS to search for CSS rule usages using
purgecsstool - Login screen support
- URL clustering (package urlclustering), only crawls samples when there are thousands of similar URLs
Websites need to be lightning fast nowadays. The more resources a website needs to download, the slower and worse ranked in search engines like Google. Most CSS cleaners are either not complete (dont crawl entire websites) or costly. This is a simple command line script that makes all of that for you.
- Install dependencies in requirements.txt
pip install -r requirements.txt
- Install purgecss
npm i -g purgecss
- Download chromedriver
python clean.py -m 1000
-u https://www.mywebsite.com
- Login and crawl
python clean.py -m 1000
-u https://www.mywebsite.com/
--username admin
--password adminpassword
--username_selector "[name='username']"
--password_selector "[name='password']"
--login_url https://www.mywebsite.com/accounts/login/
--chrome ./chromedriver #assuming chromedriver is in the same directory
# Main options
-u, --url: Main URL to parse from
--chrome: Chrome driver path
-m, --max-urls: Number of max URLs to crawl, default is 30.", default=30, type=int)
# Login options
--login_url: Optional. If the login URL differs from the main parse URL
--login_btn_selector: Optional: The CSS or xpath selector for the login button
--username: Optional: the username for the login
--username_selector: Optional, The CSS or xpath selector for the username field
--password: Optional, the password for the login
--password_selector: Optional, The CSS or xpath selector for the password field
--extra_urls: Optional, extra seed URLs to crawl
Results are stored in runs/ directory.
CSSs files are in runs/{runid}/cssout.
There are also CSV reports of the parsed links, CSS with original size and reduction runs/{runid}/reports.