googlevideo.py 文件源码

def _parse_gdocs(html):
    urls = []
    for match in re.finditer('\[\s*"([^"]+)"\s*,\s*"([^"]+)"\s*\]', html):
        key, value = match.groups()
        if key == 'fmt_stream_map':
            items = value.split(',')
            for item in items:
                _source_itag, source_url = item.split('|')
                if isinstance(source_url, unicode):
                    source_url = source_url.encode('utf-8')

                source_url = source_url.decode('unicode_escape')
                quality = itag_map.get(_source_itag, 'Unknown Quality [%s]' % _source_itag)
                source_url = urllib2.unquote(source_url)
                urls.append([quality, source_url])
            return urls

    return urls