kumi/yt-dlp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

[extractor/twitter:spaces] Fix extraction (#7512)

Browse source 
Closes #7455
Authored by: bashonly
This commit is contained in:
bashonly 2023-07-04 22:05:52 -05:00 committed by GitHub
parent 3b7f5300c5
commit 1cffd621cb
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 24 additions and 14 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

38
yt_dlp/extractor/twitter.py
View file

@ -889,8 +889,10 @@ class TwitterIE(TwitterBaseIE):
'uploader_id': 'MoniqueCamarra', 'uploader_id': 'MoniqueCamarra',
'live_status': 'was_live', 'live_status': 'was_live',
'release_timestamp': 1658417414, 'release_timestamp': 1658417414,
'description': 'md5:acce559345fd49f129c20dbcda3f1201', 'description': 'md5:4dc8e972f1d8b3c6580376fabb02a3ad',
'timestamp': 1658407771464, 'timestamp': 1658407771,
'release_date': '20220721',
'upload_date': '20220721',
}, },
'add_ie': ['TwitterSpaces'], 'add_ie': ['TwitterSpaces'],
'params': {'skip_download': 'm3u8'}, 'params': {'skip_download': 'm3u8'},
@ -1436,7 +1438,10 @@ class TwitterSpacesIE(TwitterBaseIE):
'uploader': r're:Lucio Di Gaetano.*?', 'uploader': r're:Lucio Di Gaetano.*?',
'uploader_id': 'luciodigaetano', 'uploader_id': 'luciodigaetano',
'live_status': 'was_live', 'live_status': 'was_live',
'timestamp': 1659877956397, 'timestamp': 1659877956,
'upload_date': '20220807',
'release_timestamp': 1659904215,
'release_date': '20220807',
}, },
'params': {'skip_download': 'm3u8'}, 'params': {'skip_download': 'm3u8'},
}] }]
@ -1482,26 +1487,31 @@ class TwitterSpacesIE(TwitterBaseIE):
metadata = space_data['metadata'] metadata = space_data['metadata']
live_status = try_call(lambda: self.SPACE_STATUS[metadata['state'].lower()]) live_status = try_call(lambda: self.SPACE_STATUS[metadata['state'].lower()])
is_live = live_status == 'is_live'
formats = [] formats = []
if live_status == 'is_upcoming': if live_status == 'is_upcoming':
self.raise_no_formats('Twitter Space not started yet', expected=True) self.raise_no_formats('Twitter Space not started yet', expected=True)
elif live_status == 'post_live': elif not is_live and not metadata.get('is_space_available_for_replay'):
self.raise_no_formats('Twitter Space ended but not downloadable yet', expected=True) self.raise_no_formats('Twitter Space ended and replay is disabled', expected=True)
else: elif metadata.get('media_key'):
source = self._call_api( source = traverse_obj(
f'live_video_stream/status/{metadata["media_key"]}', metadata['media_key'])['source'] self._call_api(f'live_video_stream/status/{metadata["media_key"]}', metadata['media_key']),
('source', ('noRedirectPlaybackUrl', 'location'), {url_or_none}), get_all=False)
# XXX: Native downloader does not work
formats = self._extract_m3u8_formats( formats = self._extract_m3u8_formats(
traverse_obj(source, 'noRedirectPlaybackUrl', 'location'), source, metadata['media_key'], 'm4a', live=is_live, fatal=False,
metadata['media_key'], 'm4a', 'm3u8', live=live_status == 'is_live', headers={'Referer': 'https://twitter.com/'}) if source else []
headers={'Referer': 'https://twitter.com/'})
for fmt in formats: for fmt in formats:
fmt.update({'vcodec': 'none', 'acodec': 'aac'}) fmt.update({'vcodec': 'none', 'acodec': 'aac'})
if not is_live:
fmt['container'] = 'm4a_dash'
participants = ', '.join(traverse_obj( participants = ', '.join(traverse_obj(
space_data, ('participants', 'speakers', ..., 'display_name'))) or 'nobody yet' space_data, ('participants', 'speakers', ..., 'display_name'))) or 'nobody yet'
if not formats and live_status == 'post_live':
self.raise_no_formats('Twitter Space ended but not downloadable yet', expected=True)
return { return {
'id': space_id, 'id': space_id,
'title': metadata.get('title'), 'title': metadata.get('title'),
@ -1513,7 +1523,7 @@ class TwitterSpacesIE(TwitterBaseIE):
'live_status': live_status, 'live_status': live_status,
'release_timestamp': try_call( 'release_timestamp': try_call(
lambda: int_or_none(metadata['scheduled_start'], scale=1000)), lambda: int_or_none(metadata['scheduled_start'], scale=1000)),
'timestamp': metadata.get('created_at'), 'timestamp': int_or_none(metadata.get('created_at'), scale=1000),
'formats': formats, 'formats': formats,
} }