+
+ talk_info = self._extract_info(webpage)['talks'][0]
+
+ external = talk_info.get('external')
+ if external:
+ service = external['service']
+ self.to_screen('Found video from %s' % service)
+ ext_url = None
+ if service.lower() == 'youtube':
+ ext_url = external.get('code')
+ return {
+ '_type': 'url',
+ 'url': ext_url or external['uri'],
+ }
+
+ formats = [{
+ 'url': format_url,
+ 'format_id': format_id,
+ 'format': format_id,
+ } for (format_id, format_url) in talk_info['nativeDownloads'].items() if format_url is not None]
+ if formats:
+ for f in formats:
+ finfo = self._NATIVE_FORMATS.get(f['format_id'])
+ if finfo:
+ f.update(finfo)
+
+ http_url = None
+ for format_id, resources in talk_info['resources'].items():
+ 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':
+ formats.extend(self._extract_m3u8_formats(
+ resources.get('stream'), video_name, 'mp4', m3u8_id=format_id, fatal=False))
+
+ m3u8_formats = list(filter(
+ lambda f: f.get('protocol') == 'm3u8' and f.get('vcodec') != 'none' and f.get('resolution') != 'multiple',
+ 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
+ f = m3u8_format.copy()
+ f.update({
+ 'url': re.sub(r'\d+k', bitrate, http_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'])
+
+ thumbnail = talk_info['thumb']
+ if not thumbnail.startswith('http'):
+ thumbnail = 'http://' + thumbnail
+ return {
+ 'id': video_id,
+ 'title': talk_info['title'].strip(),
+ 'uploader': talk_info['speaker'],
+ 'thumbnail': thumbnail,
+ 'description': self._og_search_description(webpage),
+ 'subtitles': self._get_subtitles(video_id, talk_info),
+ 'formats': formats,
+ 'duration': talk_info.get('duration'),
+ }
+
+ def _get_subtitles(self, video_id, talk_info):
+ languages = [lang['languageCode'] for lang in talk_info.get('languages', [])]
+ if languages:
+ sub_lang_list = {}
+ for l in languages:
+ sub_lang_list[l] = [
+ {
+ 'url': 'http://www.ted.com/talks/subtitles/id/%s/lang/%s/format/%s' % (video_id, l, ext),
+ 'ext': ext,
+ }
+ for ext in ['ted', 'srt']
+ ]
+ return sub_lang_list
+ else:
+ return {}
+
+ 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