Skip to content

Commit c0c5134

Browse files
remitamineremitamine
committed
[curiositystream] fix format extraction(closes ytdl-org#26845, closes ytdl-org#28668)
1 parent 72a2c0a commit c0c5134

File tree

1 file changed

+58
-45
lines changed

1 file changed

+58
-45
lines changed

‎youtube_dl/extractor/curiositystream.py

Lines changed: 58 additions & 45 deletions
Original file line numberDiff line numberDiff line change
@@ -25,12 +25,12 @@ def _handle_errors(self, result):
2525
raise ExtractorError(
2626
'%s said: %s' % (self.IE_NAME, error), expected=True)
2727

28-
def _call_api(self, path, video_id):
28+
def _call_api(self, path, video_id, query=None):
2929
headers = {}
3030
if self._auth_token:
3131
headers['X-Auth-Token'] = self._auth_token
3232
result = self._download_json(
33-
self._API_BASE_URL + path, video_id, headers=headers)
33+
self._API_BASE_URL + path, video_id, headers=headers, query=query)
3434
self._handle_errors(result)
3535
return result['data']
3636

@@ -52,62 +52,75 @@ class CuriosityStreamIE(CuriosityStreamBaseIE):
5252
_VALID_URL = r'https?://(?:app\.)?curiositystream\.com/video/(?P<id>\d+)'
5353
_TEST = {
5454
'url': 'https://app.curiositystream.com/video/2',
55-
'md5': '262bb2f257ff301115f1973540de8983',
5655
'info_dict': {
5756
'id': '2',
5857
'ext': 'mp4',
5958
'title': 'How Did You Develop The Internet?',
6059
'description': 'Vint Cerf, Google\'s Chief Internet Evangelist, describes how he and Bob Kahn created the internet.',
61-
}
60+
},
61+
'params': {
62+
'format': 'bestvideo',
63+
# m3u8 download
64+
'skip_download': True,
65+
},
6266
}
6367

6468
def _real_extract(self, url):
6569
video_id = self._match_id(url)
66-
media = self._call_api('media/' + video_id, video_id)
67-
title = media['title']
6870

6971
formats = []
70-
for encoding in media.get('encodings', []):
71-
m3u8_url = encoding.get('master_playlist_url')
72-
if m3u8_url:
73-
formats.extend(self._extract_m3u8_formats(
74-
m3u8_url, video_id, 'mp4', 'm3u8_native',
75-
m3u8_id='hls', fatal=False))
76-
encoding_url = encoding.get('url')
77-
file_url = encoding.get('file_url')
78-
if not encoding_url and not file_url:
79-
continue
80-
f = {
81-
'width': int_or_none(encoding.get('width')),
82-
'height': int_or_none(encoding.get('height')),
83-
'vbr': int_or_none(encoding.get('video_bitrate')),
84-
'abr': int_or_none(encoding.get('audio_bitrate')),
85-
'filesize': int_or_none(encoding.get('size_in_bytes')),
86-
'vcodec': encoding.get('video_codec'),
87-
'acodec': encoding.get('audio_codec'),
88-
'container': encoding.get('container_type'),
89-
}
90-
for f_url in (encoding_url, file_url):
91-
if not f_url:
72+
for encoding_format in ('m3u8', 'mpd'):
73+
media = self._call_api('media/' + video_id, video_id, query={
74+
'encodingsNew': 'true',
75+
'encodingsFormat': encoding_format,
76+
})
77+
for encoding in media.get('encodings', []):
78+
playlist_url = encoding.get('master_playlist_url')
79+
if encoding_format == 'm3u8':
80+
# use `m3u8` entry_protocol until EXT-X-MAP is properly supported by `m3u8_native` entry_protocol
81+
formats.extend(self._extract_m3u8_formats(
82+
playlist_url, video_id, 'mp4',
83+
m3u8_id='hls', fatal=False))
84+
elif encoding_format == 'mpd':
85+
formats.extend(self._extract_mpd_formats(
86+
playlist_url, video_id, mpd_id='dash', fatal=False))
87+
encoding_url = encoding.get('url')
88+
file_url = encoding.get('file_url')
89+
if not encoding_url and not file_url:
9290
continue
93-
fmt = f.copy()
94-
rtmp = re.search(r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+))/(?P<playpath>mp[34]:.+)$', f_url)
95-
if rtmp:
96-
fmt.update({
97-
'url': rtmp.group('url'),
98-
'play_path': rtmp.group('playpath'),
99-
'app': rtmp.group('app'),
100-
'ext': 'flv',
101-
'format_id': 'rtmp',
102-
})
103-
else:
104-
fmt.update({
105-
'url': f_url,
106-
'format_id': 'http',
107-
})
108-
formats.append(fmt)
91+
f = {
92+
'width': int_or_none(encoding.get('width')),
93+
'height': int_or_none(encoding.get('height')),
94+
'vbr': int_or_none(encoding.get('video_bitrate')),
95+
'abr': int_or_none(encoding.get('audio_bitrate')),
96+
'filesize': int_or_none(encoding.get('size_in_bytes')),
97+
'vcodec': encoding.get('video_codec'),
98+
'acodec': encoding.get('audio_codec'),
99+
'container': encoding.get('container_type'),
100+
}
101+
for f_url in (encoding_url, file_url):
102+
if not f_url:
103+
continue
104+
fmt = f.copy()
105+
rtmp = re.search(r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+))/(?P<playpath>mp[34]:.+)$', f_url)
106+
if rtmp:
107+
fmt.update({
108+
'url': rtmp.group('url'),
109+
'play_path': rtmp.group('playpath'),
110+
'app': rtmp.group('app'),
111+
'ext': 'flv',
112+
'format_id': 'rtmp',
113+
})
114+
else:
115+
fmt.update({
116+
'url': f_url,
117+
'format_id': 'http',
118+
})
119+
formats.append(fmt)
109120
self._sort_formats(formats)
110121

122+
title = media['title']
123+
111124
subtitles = {}
112125
for closed_caption in media.get('closed_captions', []):
113126
sub_url = closed_caption.get('file')
@@ -140,7 +153,7 @@ class CuriosityStreamCollectionIE(CuriosityStreamBaseIE):
140153
'title': 'Curious Minds: The Internet',
141154
'description': 'How is the internet shaping our lives in the 21st Century?',
142155
},
143-
'playlist_mincount': 17,
156+
'playlist_mincount': 16,
144157
}, {
145158
'url': 'https://curiositystream.com/series/2',
146159
'only_matching': True,

0 commit comments

Comments
 (0)