]>
jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/dvtv.py
2 from __future__
import unicode_literals
6 from .common
import InfoExtractor
20 class DVTVIE(InfoExtractor
):
22 IE_DESC
= 'http://video.aktualne.cz/'
23 _VALID_URL
= r
'https?://video\.aktualne\.cz/(?:[^/]+/)+r~(?P<id>[0-9a-f]{32})'
25 'url': 'http://video.aktualne.cz/dvtv/vondra-o-ceskem-stoleti-pri-pohledu-na-havla-mi-bylo-trapne/r~e5efe9ca855511e4833a0025900fea04/',
26 'md5': '67cb83e4a955d36e1b5d31993134a0c2',
28 'id': 'dc0768de855511e49e4b0025900fea04',
30 'title': 'Vondra o Českém století: Při pohledu na Havla mi bylo trapně',
32 'upload_date': '20141217',
33 'timestamp': 1418792400,
36 'url': 'http://video.aktualne.cz/dvtv/dvtv-16-12-2014-utok-talibanu-boj-o-kliniku-uprchlici/r~973eb3bc854e11e498be002590604f2e/',
38 'title': r
'DVTV 16. 12. 2014: útok Talibanu, boj o kliniku, uprchlíci',
39 'id': '973eb3bc854e11e498be002590604f2e',
42 'md5': 'da7ca6be4935532241fa9520b3ad91e4',
44 'id': 'b0b40906854d11e4bdad0025900fea04',
46 'title': 'Drtinová Veselovský TV 16. 12. 2014: Témata dne',
47 'description': 'md5:0916925dea8e30fe84222582280b47a0',
48 'timestamp': 1418760010,
49 'upload_date': '20141216',
52 'md5': '5f7652a08b05009c1292317b449ffea2',
54 'id': '420ad9ec854a11e4bdad0025900fea04',
56 'title': 'Školní masakr možná změní boj s Talibanem, říká novinářka',
57 'description': 'md5:ff2f9f6de73c73d7cef4f756c1c1af42',
58 'timestamp': 1418760010,
59 'upload_date': '20141216',
62 'md5': '498eb9dfa97169f409126c617e2a3d64',
64 'id': '95d35580846a11e4b6d20025900fea04',
66 'title': 'Boj o kliniku: Veřejný zájem, nebo právo na majetek?',
67 'description': 'md5:889fe610a70fee5511dc3326a089188e',
68 'timestamp': 1418760010,
69 'upload_date': '20141216',
72 'md5': 'b8dc6b744844032dab6ba3781a7274b9',
74 'id': '6fe14d66853511e4833a0025900fea04',
76 'title': 'Pánek: Odmítání syrských uprchlíků je ostudou české vlády',
77 'description': 'md5:544f86de6d20c4815bea11bf2ac3004f',
78 'timestamp': 1418760010,
79 'upload_date': '20141216',
83 'url': 'https://video.aktualne.cz/dvtv/zeman-si-jen-leci-mindraky-sobotku-nenavidi-a-babis-se-mu-te/r~960cdb3a365a11e7a83b0025900fea04/',
84 'md5': 'f8efe9656017da948369aa099788c8ea',
86 'id': '3c496fec365911e7a6500025900fea04',
88 'title': 'Zeman si jen léčí mindráky, Sobotku nenávidí a Babiš se mu teď hodí, tvrdí Kmenta',
90 'upload_date': '20170511',
91 'timestamp': 1494514200,
94 'skip_download': True,
97 'url': 'http://video.aktualne.cz/v-cechach-poprve-zazni-zelenkova-zrestaurovana-mse/r~45b4b00483ec11e4883b002590604f2e/',
98 'only_matching': True,
100 # Test live stream video (liveStarter) parsing
101 'url': 'https://video.aktualne.cz/dvtv/zive-mistryne-sveta-eva-samkova-po-navratu-ze-sampionatu/r~182654c2288811e990fd0cc47ab5f122/',
102 'md5': '2e552e483f2414851ca50467054f9d5d',
104 'id': '8d116360288011e98c840cc47ab5f122',
106 'title': 'Živě: Mistryně světa Eva Samková po návratu ze šampionátu',
107 'upload_date': '20190204',
108 'timestamp': 1549289591,
111 # Video content is no longer available
112 'skip_download': True,
116 def _parse_video_metadata(self
, js
, video_id
, timestamp
):
117 data
= self
._parse
_json
(js
, video_id
, transform_source
=js_to_json
)
118 title
= unescapeHTML(data
['title'])
120 live_starter
= try_get(data
, lambda x
: x
['plugins']['liveStarter'], dict)
122 data
.update(live_starter
)
125 for tracks
in data
.get('tracks', {}).values():
127 video_url
= video
.get('src')
130 video_type
= video
.get('type')
131 ext
= determine_ext(video_url
, mimetype2ext(video_type
))
132 if video_type
== 'application/vnd.apple.mpegurl' or ext
== 'm3u8':
133 formats
.extend(self
._extract
_m
3u8_formats
(
134 video_url
, video_id
, 'mp4', entry_protocol
='m3u8_native',
135 m3u8_id
='hls', fatal
=False))
136 elif video_type
== 'application/dash+xml' or ext
== 'mpd':
137 formats
.extend(self
._extract
_mpd
_formats
(
138 video_url
, video_id
, mpd_id
='dash', fatal
=False))
140 label
= video
.get('label')
141 height
= self
._search
_regex
(
142 r
'^(\d+)[pP]', label
or '', 'height', default
=None)
145 'format_id': join_nonempty('http', ext
, label
),
146 'height': int_or_none(height
),
148 self
._sort
_formats
(formats
)
151 'id': data
.get('mediaid') or video_id
,
153 'description': data
.get('description'),
154 'thumbnail': data
.get('image'),
155 'duration': int_or_none(data
.get('duration')),
156 'timestamp': int_or_none(timestamp
),
160 def _real_extract(self
, url
):
161 video_id
= self
._match
_id
(url
)
162 webpage
= self
._download
_webpage
(url
, video_id
)
163 timestamp
= parse_iso8601(self
._html
_search
_meta
(
164 'article:published_time', webpage
, 'published time', default
=None))
166 items
= re
.findall(r
'(?s)playlist\.push\(({.+?})\);', webpage
)
168 return self
.playlist_result(
169 [self
._parse
_video
_metadata
(i
, video_id
, timestamp
) for i
in items
],
170 video_id
, self
._html
_search
_meta
('twitter:title', webpage
))
172 item
= self
._search
_regex
(
173 r
'(?s)BBXPlayer\.setup\((.+?)\);',
174 webpage
, 'video', default
=None)
176 # remove function calls (ex. htmldeentitize)
177 # TODO this should be fixed in a general way in the js_to_json
178 item
= re
.sub(r
'\w+?\((.+)\)', r
'\1', item
)
179 return self
._parse
_video
_metadata
(item
, video_id
, timestamp
)
181 raise ExtractorError('Could not find neither video nor playlist')