html_to_telegraph.py 文件源码

def preprocess_media_tags(element):
    if isinstance(element, html.HtmlElement):
        if element.tag in ['ol', 'ul']:
            # ignore any spaces between <ul> and <li>
            element.text = ''
        elif element.tag == 'li':
            # ignore spaces after </li>
            element.tail = ''
        elif element.tag == 'iframe':
            iframe_src = element.get('src')

            youtube = re.match(youtube_re, iframe_src)
            vimeo = re.match(vimeo_re, iframe_src)
            if youtube or vimeo:
                element.text = ''  # ignore any legacy text
                if youtube:
                    yt_id = urlparse(iframe_src).path.replace('/embed/', '')
                    element.set('src', '/embed/youtube?url=' + quote_plus('https://www.youtube.com/watch?v=' + yt_id))
                elif vimeo:
                    element.set('src', '/embed/vimeo?url=' + quote_plus('https://vimeo.com/' + vimeo.group(2)))

                if not len(element.xpath('./ancestor::figure')):
                    _wrap_figure(element)
            else:
                element.drop_tag()

        elif element.tag == 'blockquote' and element.get('class') == 'twitter-tweet':
            twitter_links = element.xpath('.//a[@href]')
            for tw_link in twitter_links:
                if twitter_re.match(tw_link.get('href')):
                    twitter_frame = html.HtmlElement()
                    twitter_frame.tag = 'iframe'
                    twitter_frame.set('src', '/embed/twitter?url=' + quote_plus(tw_link.get('href')))
                    element.addprevious(twitter_frame)
                    _wrap_figure(twitter_frame)
                    element.drop_tree()