+
+ resources_ = player_talk.get('resources') or talk_info.get('resources')
+
+ http_url = None
+ for format_id, resources in resources_.items():
+ if not isinstance(resources, dict):
+ continue
+ if format_id == 'h264':
+ for resource in resources:
+ h264_url = resource.get('file')
+ if not h264_url:
+ continue
+ bitrate = int_or_none(resource.get('bitrate'))
+ formats.append({
+ 'url': h264_url,
+ 'format_id': '%s-%sk' % (format_id, bitrate),
+ 'tbr': bitrate,
+ })
+ if re.search(r'\d+k', h264_url):
+ http_url = h264_url
+ elif format_id == 'rtmp':
+ streamer = talk_info.get('streamer')
+ if not streamer:
+ continue
+ for resource in resources:
+ formats.append({
+ 'format_id': '%s-%s' % (format_id, resource.get('name')),
+ 'url': streamer,
+ 'play_path': resource['file'],
+ 'ext': 'flv',
+ 'width': int_or_none(resource.get('width')),
+ 'height': int_or_none(resource.get('height')),
+ 'tbr': int_or_none(resource.get('bitrate')),
+ })
+ elif format_id == 'hls':
+ stream_url = url_or_none(resources.get('stream'))
+ if not stream_url:
+ continue
+ formats.extend(self._extract_m3u8_formats(
+ stream_url, video_name, 'mp4', m3u8_id=format_id,
+ fatal=False))
+
+ m3u8_formats = list(filter(
+ lambda f: f.get('protocol') == 'm3u8' and f.get('vcodec') != 'none',
+ formats))
+ if http_url:
+ for m3u8_format in m3u8_formats:
+ bitrate = self._search_regex(r'(\d+k)', m3u8_format['url'], 'bitrate', default=None)
+ if not bitrate:
+ continue
+ bitrate_url = re.sub(r'\d+k', bitrate, http_url)
+ if not self._is_valid_url(
+ bitrate_url, video_name, '%s bitrate' % bitrate):
+ continue
+ f = m3u8_format.copy()
+ f.update({
+ 'url': bitrate_url,
+ 'format_id': m3u8_format['format_id'].replace('hls', 'http'),
+ 'protocol': 'http',
+ })
+ formats.append(f)
+
+ audio_download = talk_info.get('audioDownload')
+ if audio_download:
+ formats.append({
+ 'url': audio_download,
+ 'format_id': 'audio',
+ 'vcodec': 'none',
+ })
+
+ self._sort_formats(formats)
+
+ video_id = compat_str(talk_info['id'])
+
+ return {
+ 'id': video_id,
+ 'title': title,
+ 'uploader': player_talk.get('speaker') or talk_info.get('speaker'),
+ 'thumbnail': player_talk.get('thumb') or talk_info.get('thumb'),
+ 'description': self._og_search_description(webpage),
+ 'subtitles': self._get_subtitles(video_id, talk_info),
+ 'formats': formats,
+ 'duration': float_or_none(talk_info.get('duration')),
+ 'view_count': int_or_none(data.get('viewed_count')),
+ 'comment_count': int_or_none(
+ try_get(data, lambda x: x['comments']['count'])),
+ 'tags': try_get(talk_info, lambda x: x['tags'], list),
+ }
+
+ def _get_subtitles(self, video_id, talk_info):
+ sub_lang_list = {}
+ for language in try_get(
+ talk_info,
+ (lambda x: x['downloads']['languages'],
+ lambda x: x['languages']), list):
+ lang_code = language.get('languageCode') or language.get('ianaCode')
+ if not lang_code:
+ continue
+ sub_lang_list[lang_code] = [
+ {
+ 'url': 'http://www.ted.com/talks/subtitles/id/%s/lang/%s/format/%s' % (video_id, lang_code, ext),
+ 'ext': ext,
+ }
+ for ext in ['ted', 'srt']
+ ]
+ return sub_lang_list
+
+ def _watch_info(self, url, name):
+ webpage = self._download_webpage(url, name)
+
+ config_json = self._html_search_regex(
+ r'"pages\.jwplayer"\s*,\s*({.+?})\s*\)\s*</script>',
+ webpage, 'config', default=None)
+ if not config_json:
+ embed_url = self._search_regex(
+ r"<iframe[^>]+class='pages-video-embed__video__object'[^>]+src='([^']+)'", webpage, 'embed url')
+ return self.url_result(self._proto_relative_url(embed_url))
+ config = json.loads(config_json)['config']
+ video_url = config['video']['url']
+ thumbnail = config.get('image', {}).get('url')
+
+ title = self._html_search_regex(
+ r"(?s)<h1(?:\s+class='[^']+')?>(.+?)</h1>", webpage, 'title')
+ description = self._html_search_regex(
+ [
+ r'(?s)<h4 class="[^"]+" id="h3--about-this-talk">.*?</h4>(.*?)</div>',
+ r'(?s)<p><strong>About this talk:</strong>\s+(.*?)</p>',
+ ],
+ webpage, 'description', fatal=False)
+
+ return {
+ 'id': name,
+ 'url': video_url,
+ 'title': title,
+ 'thumbnail': thumbnail,
+ 'description': description,
+ }
Rapsys Git / youtubedl / blobdiff