mirrors/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-30 12:01:28 +00:00
Code Issues Packages Projects Releases Wiki Activity

[common] extract formats from #EXT-X-MEDIA tags

Browse source
This commit is contained in:
Remita Amine 2016-08-25 08:39:23 +01:00
parent 07ea9c9b05
commit f39ffc5877
1 changed files with 22 additions and 10 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

32
youtube_dl/extractor/common.py
View file

@ -1202,30 +1202,45 @@ def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None,
'preference': preference,
}]
last_info = None
last_media = None
for line in m3u8_doc.splitlines():
if line.startswith('#EXT-X-STREAM-INF:'):
last_info = parse_m3u8_attributes(line)
elif line.startswith('#EXT-X-MEDIA:'):
last_media = parse_m3u8_attributes(line)
media = parse_m3u8_attributes(line)
media_type = media.get('TYPE')
if media_type in ('VIDEO', 'AUDIO'):
media_url = media.get('URI')
if media_url:
format_id = []
for v in (media.get('GROUP-ID'), media.get('NAME')):
if v:
format_id.append(v)
formats.append({
'format_id': '-'.join(format_id),
'url': format_url(media_url),
'language': media.get('LANGUAGE'),
'vcodec': 'none' if media_type == 'AUDIO' else None,
'ext': ext,
'protocol': entry_protocol,
'preference': preference,
})
elif line.startswith('#') or not line.strip():
continue
else:
if last_info is None:
formats.append({'url': format_url(line)})
continue
tbr = int_or_none(last_info.get('BANDWIDTH'), scale=1000)
tbr = int_or_none(last_info.get('AVERAGE-BANDWIDTH') or last_info.get('BANDWIDTH'), scale=1000)
format_id = []
if m3u8_id:
format_id.append(m3u8_id)
last_media_name = last_media.get('NAME') if last_media and last_media.get('TYPE') not in ('SUBTITLES', 'CLOSED-CAPTIONS') else None
# Despite specification does not mention NAME attribute for
# EXT-X-STREAM-INF it still sometimes may be present
stream_name = last_info.get('NAME') or last_media_name
# Bandwidth of live streams may differ over time thus making
# format_id unpredictable. So it's better to keep provided
# format_id intact.
if not live:
# Despite specification does not mention NAME attribute for
# EXT-X-STREAM-INF it still sometimes may be present
stream_name = last_info.get('NAME')
format_id.append(stream_name if stream_name else '%d' % (tbr if tbr else len(formats)))
f = {
'format_id': '-'.join(format_id),
@ -1252,9 +1267,6 @@ def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None,
'abr': abr,
})
f.update(parse_codecs(last_info.get('CODECS')))
if last_media is not None:
f['m3u8_media'] = last_media
last_media = None
formats.append(f)
last_info = {}
return formats