+ stream_url = asset.get('stream_url') or asset.get('streamUrl')
+ if stream_url:
+ youtube_url = self._search_regex(
+ r'(https?://www\.youtube\.com/watch\?v=.*)', stream_url, 'youtube URL', default=None)
+ if youtube_url:
+ return self.url_result(youtube_url, 'Youtube')
+
+ video_id = compat_str(asset['id'])
+ thumbnail = asset.get('thumbnail_url') or asset.get('thumbnailUrl')
+ duration = float_or_none(asset.get('data', {}).get('duration'))
+
+ formats = []
+
+ def extract_output_format(src):
+ return {
+ 'url': src['url'],
+ 'format_id': '%sp' % (src.get('height') or format_id),
+ 'width': int_or_none(src.get('width')),
+ 'height': int_or_none(src.get('height')),
+ 'vbr': int_or_none(src.get('video_bitrate_in_kbps')),
+ 'vcodec': src.get('video_codec'),
+ 'fps': int_or_none(src.get('frame_rate')),
+ 'abr': int_or_none(src.get('audio_bitrate_in_kbps')),
+ 'acodec': src.get('audio_codec'),
+ 'asr': int_or_none(src.get('audio_sample_rate')),
+ 'tbr': int_or_none(src.get('total_bitrate_in_kbps')),
+ 'filesize': int_or_none(src.get('file_size_in_bytes')),
+ }
+
+ outputs = asset.get('data', {}).get('outputs')
+ if not isinstance(outputs, dict):
+ outputs = {}
+
+ def add_output_format_meta(f, key):
+ output = outputs.get(key)
+ if isinstance(output, dict):
+ output_format = extract_output_format(output)
+ output_format.update(f)
+ return output_format
+ return f
+
+ download_urls = asset.get('download_urls')
+ if isinstance(download_urls, dict):
+ video = download_urls.get('Video')
+ if isinstance(video, list):
+ for format_ in video:
+ video_url = format_.get('file')
+ if not video_url:
+ continue
+ format_id = format_.get('label')
+ f = {
+ 'url': format_['file'],
+ 'format_id': '%sp' % format_id,
+ 'height': int_or_none(format_id),
+ }
+ if format_id:
+ # Some videos contain additional metadata (e.g.
+ # https://www.udemy.com/ios9-swift/learn/#/lecture/3383208)
+ f = add_output_format_meta(f, format_id)
+ formats.append(f)
+
+ view_html = lecture.get('view_html')
+ if view_html:
+ view_html_urls = set()
+ for source in re.findall(r'<source[^>]+>', view_html):
+ attributes = extract_attributes(source)
+ src = attributes.get('src')
+ if not src:
+ continue
+ res = attributes.get('data-res')
+ height = int_or_none(res)
+ if src in view_html_urls:
+ continue
+ view_html_urls.add(src)
+ if attributes.get('type') == 'application/x-mpegURL' or determine_ext(src) == 'm3u8':
+ m3u8_formats = self._extract_m3u8_formats(
+ src, video_id, 'mp4', entry_protocol='m3u8_native',
+ m3u8_id='hls', fatal=False)
+ for f in m3u8_formats:
+ m = re.search(r'/hls_(?P<height>\d{3,4})_(?P<tbr>\d{2,})/', f['url'])
+ if m:
+ if not f.get('height'):
+ f['height'] = int(m.group('height'))
+ if not f.get('tbr'):
+ f['tbr'] = int(m.group('tbr'))
+ formats.extend(m3u8_formats)
+ else:
+ formats.append(add_output_format_meta({
+ 'url': src,
+ 'format_id': '%dp' % height if height else None,
+ 'height': height,
+ }, res))
+
+ self._sort_formats(formats, field_preference=('height', 'width', 'tbr', 'format_id'))
Rapsys Git / youtubedl / blobdiff