README.MD

clean_html_for_llm

This library provides a method to clean HTML content by removing specified tags and attributes while keeping specified attributes. It is particularly useful for preprocessing HTML data to remove noisy tags, making it easier for language models (LLMs) to understand the HTML and generate accurate responses. This is helpful if you are querying LLMs with your HTML data.

Installation

You can install the clean_html_for_llm library using pip:

pip install clean-html-for-llm

Usage

from clean_html_for_llm import clean_html

html_content = '<div id="main" style="color:red">Hello <script>alert("World")</script></div>'
cleaned_html = clean_html(html_content, tags_to_remove=['script'], attributes_to_keep=['id'])
print(cleaned_html)
# Output: '<div id="main">Hello </div>'

The clean_html function takes the following arguments:

• html_to_clean (str): The HTML content to clean.
• tags_to_remove (List[str]): List of tags to remove from the HTML content. Default is ['style', 'svg', 'script'].
• attributes_to_keep (List[str]): List of attributes to keep in the HTML tags. Default is ['id', 'href'].

You can customize the tags and attributes to remove or keep based on your requirements.

Examples

Example 1:

html_content = '<div id="content" class="main">This is a <span style="font-size: 18px;">paragraph</span>.</div>'
cleaned_html = clean_html(html_content)
print(cleaned_html)
# Output: '<div id="content">This is a <span>paragraph</span>.</div>'

Example 2:

html_content = '<p class="content">Click <a href="https://example.com">here</a> for more information.</p>'
cleaned_html = clean_html(html_content, tags_to_remove=['a'], attributes_to_keep=['class'])
print(cleaned_html)
# Output: '<p class="content">Click </p>'

License

This library is released under the MIT License. See LICENSE for details.