INNER CODE UNIT · Python

remove_blacklisted_attributes

handrew/browserpilot · browserpilot/agents/gpt_selenium_agent.py:215

        def remove_blacklisted_attributes(tag, blacklisted_attributes):
            for attr in tag.attrs.copy():
                for pattern in blacklisted_attributes:
                    if re.match(pattern, attr):
                        del tag[attr]

        for tag in soup.find_all(True):
            remove_blacklisted_attributes(tag, blacklisted_attributes)

        return soup

    def __get_html_elements_for_llm(self):
        """Returns list of BeautifulSoup elements for use in GPT Index.

        First removes blacklisted elements and attributes, then removes any
        children of elements. Finally, removes any elements with no attrs.
        """
        soup = self._remove_blacklisted_elements_and_attributes()

View source record →