html_to_telegraph.py 文件源码

def convert_html_to_telegraph_format(html_string, clean_html=True, output_format="json_string"):
    if clean_html:
        html_string = clean_article_html(html_string)

        body = preprocess_fragments(
            _fragments_from_string(html_string)
        )
        if body is not None:
            desc = [x for x in body.iterdescendants()]
            for tag in desc:
                preprocess_media_tags(tag)
            move_to_top(body)
            post_process(body)
    else:
        fragments = _fragments_from_string(html_string)
        body = fragments[0].getparent() if len(fragments) else None

    content = []
    if body is not None:
        content = [_recursive_convert(x) for x in body.iterchildren()]

    if output_format == 'json_string':
        return json.dumps(content, ensure_ascii=False)
    elif output_format == 'python_list':
        return content
    elif output_format == 'html_string':
        return html.tostring(body, encoding='unicode')