]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/vevo.py
388b4debee27d7331ae7dc351338e3829e539071
[youtubedl] / youtube_dl / extractor / vevo.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6 from ..compat import (
7 compat_etree_fromstring,
8 compat_str,
9 compat_urlparse,
10 )
11 from ..utils import (
12 ExtractorError,
13 int_or_none,
14 sanitized_Request,
15 parse_iso8601,
16 )
17
18
19 class VevoBaseIE(InfoExtractor):
20 def _extract_json(self, webpage, video_id, item):
21 return self._parse_json(
22 self._search_regex(
23 r'window\.__INITIAL_STORE__\s*=\s*({.+?});\s*</script>',
24 webpage, 'initial store'),
25 video_id)['default'][item]
26
27
28 class VevoIE(VevoBaseIE):
29 '''
30 Accepts urls from vevo.com or in the format 'vevo:{id}'
31 (currently used by MTVIE and MySpaceIE)
32 '''
33 _VALID_URL = r'''(?x)
34 (?:https?://www\.vevo\.com/watch/(?!playlist|genre)(?:[^/]+/(?:[^/]+/)?)?|
35 https?://cache\.vevo\.com/m/html/embed\.html\?video=|
36 https?://videoplayer\.vevo\.com/embed/embedded\?videoId=|
37 vevo:)
38 (?P<id>[^&?#]+)'''
39
40 _TESTS = [{
41 'url': 'http://www.vevo.com/watch/hurts/somebody-to-die-for/GB1101300280',
42 'md5': '95ee28ee45e70130e3ab02b0f579ae23',
43 'info_dict': {
44 'id': 'GB1101300280',
45 'ext': 'mp4',
46 'title': 'Hurts - Somebody to Die For',
47 'timestamp': 1372057200,
48 'upload_date': '20130624',
49 'uploader': 'Hurts',
50 'track': 'Somebody to Die For',
51 'artist': 'Hurts',
52 'genre': 'Pop',
53 },
54 'expected_warnings': ['Unable to download SMIL file'],
55 }, {
56 'note': 'v3 SMIL format',
57 'url': 'http://www.vevo.com/watch/cassadee-pope/i-wish-i-could-break-your-heart/USUV71302923',
58 'md5': 'f6ab09b034f8c22969020b042e5ac7fc',
59 'info_dict': {
60 'id': 'USUV71302923',
61 'ext': 'mp4',
62 'title': 'Cassadee Pope - I Wish I Could Break Your Heart',
63 'timestamp': 1392796919,
64 'upload_date': '20140219',
65 'uploader': 'Cassadee Pope',
66 'track': 'I Wish I Could Break Your Heart',
67 'artist': 'Cassadee Pope',
68 'genre': 'Country',
69 },
70 'expected_warnings': ['Unable to download SMIL file'],
71 }, {
72 'note': 'Age-limited video',
73 'url': 'https://www.vevo.com/watch/justin-timberlake/tunnel-vision-explicit/USRV81300282',
74 'info_dict': {
75 'id': 'USRV81300282',
76 'ext': 'mp4',
77 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
78 'age_limit': 18,
79 'timestamp': 1372888800,
80 'upload_date': '20130703',
81 'uploader': 'Justin Timberlake',
82 'track': 'Tunnel Vision (Explicit)',
83 'artist': 'Justin Timberlake',
84 'genre': 'Pop',
85 },
86 'expected_warnings': ['Unable to download SMIL file'],
87 }, {
88 'note': 'No video_info',
89 'url': 'http://www.vevo.com/watch/k-camp-1/Till-I-Die/USUV71503000',
90 'md5': '8b83cc492d72fc9cf74a02acee7dc1b0',
91 'info_dict': {
92 'id': 'USUV71503000',
93 'ext': 'mp4',
94 'title': 'K Camp - Till I Die',
95 'age_limit': 18,
96 'timestamp': 1449468000,
97 'upload_date': '20151207',
98 'uploader': 'K Camp',
99 'track': 'Till I Die',
100 'artist': 'K Camp',
101 'genre': 'Rap/Hip-Hop',
102 },
103 }, {
104 'note': 'Only available via webpage',
105 'url': 'http://www.vevo.com/watch/GBUV71600656',
106 'md5': '67e79210613865b66a47c33baa5e37fe',
107 'info_dict': {
108 'id': 'GBUV71600656',
109 'ext': 'mp4',
110 'title': 'ABC - Viva Love',
111 'age_limit': 0,
112 'timestamp': 1461830400,
113 'upload_date': '20160428',
114 'uploader': 'ABC',
115 'track': 'Viva Love',
116 'artist': 'ABC',
117 'genre': 'Pop',
118 },
119 'expected_warnings': ['Failed to download video versions info'],
120 }, {
121 # no genres available
122 'url': 'http://www.vevo.com/watch/INS171400764',
123 'only_matching': True,
124 }]
125 _SMIL_BASE_URL = 'http://smil.lvl3.vevo.com'
126 _SOURCE_TYPES = {
127 0: 'youtube',
128 1: 'brightcove',
129 2: 'http',
130 3: 'hls_ios',
131 4: 'hls',
132 5: 'smil', # http
133 7: 'f4m_cc',
134 8: 'f4m_ak',
135 9: 'f4m_l3',
136 10: 'ism',
137 13: 'smil', # rtmp
138 18: 'dash',
139 }
140 _VERSIONS = {
141 0: 'youtube', # only in AuthenticateVideo videoVersions
142 1: 'level3',
143 2: 'akamai',
144 3: 'level3',
145 4: 'amazon',
146 }
147
148 def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None):
149 formats = []
150 els = smil.findall('.//{http://www.w3.org/2001/SMIL20/Language}video')
151 for el in els:
152 src = el.attrib['src']
153 m = re.match(r'''(?xi)
154 (?P<ext>[a-z0-9]+):
155 (?P<path>
156 [/a-z0-9]+ # The directory and main part of the URL
157 _(?P<tbr>[0-9]+)k
158 _(?P<width>[0-9]+)x(?P<height>[0-9]+)
159 _(?P<vcodec>[a-z0-9]+)
160 _(?P<vbr>[0-9]+)
161 _(?P<acodec>[a-z0-9]+)
162 _(?P<abr>[0-9]+)
163 \.[a-z0-9]+ # File extension
164 )''', src)
165 if not m:
166 continue
167
168 format_url = self._SMIL_BASE_URL + m.group('path')
169 formats.append({
170 'url': format_url,
171 'format_id': 'smil_' + m.group('tbr'),
172 'vcodec': m.group('vcodec'),
173 'acodec': m.group('acodec'),
174 'tbr': int(m.group('tbr')),
175 'vbr': int(m.group('vbr')),
176 'abr': int(m.group('abr')),
177 'ext': m.group('ext'),
178 'width': int(m.group('width')),
179 'height': int(m.group('height')),
180 })
181 return formats
182
183 def _initialize_api(self, video_id):
184 req = sanitized_Request(
185 'http://www.vevo.com/auth', data=b'')
186 webpage = self._download_webpage(
187 req, None,
188 note='Retrieving oauth token',
189 errnote='Unable to retrieve oauth token')
190
191 if 'THIS PAGE IS CURRENTLY UNAVAILABLE IN YOUR REGION' in webpage:
192 self.raise_geo_restricted(
193 '%s said: This page is currently unavailable in your region' % self.IE_NAME)
194
195 auth_info = self._parse_json(webpage, video_id)
196 self._api_url_template = self.http_scheme() + '//apiv2.vevo.com/%s?token=' + auth_info['access_token']
197
198 def _call_api(self, path, *args, **kwargs):
199 return self._download_json(self._api_url_template % path, *args, **kwargs)
200
201 def _real_extract(self, url):
202 video_id = self._match_id(url)
203
204 json_url = 'http://api.vevo.com/VideoService/AuthenticateVideo?isrc=%s' % video_id
205 response = self._download_json(
206 json_url, video_id, 'Downloading video info',
207 'Unable to download info', fatal=False) or {}
208 video_info = response.get('video') or {}
209 artist = None
210 featured_artist = None
211 uploader = None
212 view_count = None
213 formats = []
214
215 if not video_info:
216 try:
217 self._initialize_api(video_id)
218 except ExtractorError:
219 ytid = response.get('errorInfo', {}).get('ytid')
220 if ytid:
221 self.report_warning(
222 'Video is geoblocked, trying with the YouTube video %s' % ytid)
223 return self.url_result(ytid, 'Youtube', ytid)
224
225 raise
226
227 video_info = self._call_api(
228 'video/%s' % video_id, video_id, 'Downloading api video info',
229 'Failed to download video info')
230
231 video_versions = self._call_api(
232 'video/%s/streams' % video_id, video_id,
233 'Downloading video versions info',
234 'Failed to download video versions info',
235 fatal=False)
236
237 # Some videos are only available via webpage (e.g.
238 # https://github.com/rg3/youtube-dl/issues/9366)
239 if not video_versions:
240 webpage = self._download_webpage(url, video_id)
241 video_versions = self._extract_json(webpage, video_id, 'streams')[video_id][0]
242
243 timestamp = parse_iso8601(video_info.get('releaseDate'))
244 artists = video_info.get('artists')
245 if artists:
246 artist = uploader = artists[0]['name']
247 view_count = int_or_none(video_info.get('views', {}).get('total'))
248
249 for video_version in video_versions:
250 version = self._VERSIONS.get(video_version['version'])
251 version_url = video_version.get('url')
252 if not version_url:
253 continue
254
255 if '.ism' in version_url:
256 continue
257 elif '.mpd' in version_url:
258 formats.extend(self._extract_mpd_formats(
259 version_url, video_id, mpd_id='dash-%s' % version,
260 note='Downloading %s MPD information' % version,
261 errnote='Failed to download %s MPD information' % version,
262 fatal=False))
263 elif '.m3u8' in version_url:
264 formats.extend(self._extract_m3u8_formats(
265 version_url, video_id, 'mp4', 'm3u8_native',
266 m3u8_id='hls-%s' % version,
267 note='Downloading %s m3u8 information' % version,
268 errnote='Failed to download %s m3u8 information' % version,
269 fatal=False))
270 else:
271 m = re.search(r'''(?xi)
272 _(?P<width>[0-9]+)x(?P<height>[0-9]+)
273 _(?P<vcodec>[a-z0-9]+)
274 _(?P<vbr>[0-9]+)
275 _(?P<acodec>[a-z0-9]+)
276 _(?P<abr>[0-9]+)
277 \.(?P<ext>[a-z0-9]+)''', version_url)
278 if not m:
279 continue
280
281 formats.append({
282 'url': version_url,
283 'format_id': 'http-%s-%s' % (version, video_version['quality']),
284 'vcodec': m.group('vcodec'),
285 'acodec': m.group('acodec'),
286 'vbr': int(m.group('vbr')),
287 'abr': int(m.group('abr')),
288 'ext': m.group('ext'),
289 'width': int(m.group('width')),
290 'height': int(m.group('height')),
291 })
292 else:
293 timestamp = int_or_none(self._search_regex(
294 r'/Date\((\d+)\)/',
295 video_info['releaseDate'], 'release date', fatal=False),
296 scale=1000)
297 artists = video_info.get('mainArtists')
298 if artists:
299 artist = uploader = artists[0]['artistName']
300
301 featured_artists = video_info.get('featuredArtists')
302 if featured_artists:
303 featured_artist = featured_artists[0]['artistName']
304
305 smil_parsed = False
306 for video_version in video_info['videoVersions']:
307 version = self._VERSIONS.get(video_version['version'])
308 if version == 'youtube':
309 continue
310 else:
311 source_type = self._SOURCE_TYPES.get(video_version['sourceType'])
312 renditions = compat_etree_fromstring(video_version['data'])
313 if source_type == 'http':
314 for rend in renditions.findall('rendition'):
315 attr = rend.attrib
316 formats.append({
317 'url': attr['url'],
318 'format_id': 'http-%s-%s' % (version, attr['name']),
319 'height': int_or_none(attr.get('frameheight')),
320 'width': int_or_none(attr.get('frameWidth')),
321 'tbr': int_or_none(attr.get('totalBitrate')),
322 'vbr': int_or_none(attr.get('videoBitrate')),
323 'abr': int_or_none(attr.get('audioBitrate')),
324 'vcodec': attr.get('videoCodec'),
325 'acodec': attr.get('audioCodec'),
326 })
327 elif source_type == 'hls':
328 formats.extend(self._extract_m3u8_formats(
329 renditions.find('rendition').attrib['url'], video_id,
330 'mp4', 'm3u8_native', m3u8_id='hls-%s' % version,
331 note='Downloading %s m3u8 information' % version,
332 errnote='Failed to download %s m3u8 information' % version,
333 fatal=False))
334 elif source_type == 'smil' and version == 'level3' and not smil_parsed:
335 formats.extend(self._extract_smil_formats(
336 renditions.find('rendition').attrib['url'], video_id, False))
337 smil_parsed = True
338 self._sort_formats(formats)
339
340 track = video_info['title']
341 if featured_artist:
342 artist = '%s ft. %s' % (artist, featured_artist)
343 title = '%s - %s' % (artist, track) if artist else track
344
345 genres = video_info.get('genres')
346 genre = (
347 genres[0] if genres and isinstance(genres, list) and
348 isinstance(genres[0], compat_str) else None)
349
350 is_explicit = video_info.get('isExplicit')
351 if is_explicit is True:
352 age_limit = 18
353 elif is_explicit is False:
354 age_limit = 0
355 else:
356 age_limit = None
357
358 duration = video_info.get('duration')
359
360 return {
361 'id': video_id,
362 'title': title,
363 'formats': formats,
364 'thumbnail': video_info.get('imageUrl') or video_info.get('thumbnailUrl'),
365 'timestamp': timestamp,
366 'uploader': uploader,
367 'duration': duration,
368 'view_count': view_count,
369 'age_limit': age_limit,
370 'track': track,
371 'artist': uploader,
372 'genre': genre,
373 }
374
375
376 class VevoPlaylistIE(VevoBaseIE):
377 _VALID_URL = r'https?://www\.vevo\.com/watch/(?P<kind>playlist|genre)/(?P<id>[^/?#&]+)'
378
379 _TESTS = [{
380 'url': 'http://www.vevo.com/watch/playlist/dadbf4e7-b99f-4184-9670-6f0e547b6a29',
381 'info_dict': {
382 'id': 'dadbf4e7-b99f-4184-9670-6f0e547b6a29',
383 'title': 'Best-Of: Birdman',
384 },
385 'playlist_count': 10,
386 }, {
387 'url': 'http://www.vevo.com/watch/genre/rock',
388 'info_dict': {
389 'id': 'rock',
390 'title': 'Rock',
391 },
392 'playlist_count': 20,
393 }, {
394 'url': 'http://www.vevo.com/watch/playlist/dadbf4e7-b99f-4184-9670-6f0e547b6a29?index=0',
395 'md5': '32dcdfddddf9ec6917fc88ca26d36282',
396 'info_dict': {
397 'id': 'USCMV1100073',
398 'ext': 'mp4',
399 'title': 'Birdman - Y.U. MAD',
400 'timestamp': 1323417600,
401 'upload_date': '20111209',
402 'uploader': 'Birdman',
403 'track': 'Y.U. MAD',
404 'artist': 'Birdman',
405 'genre': 'Rap/Hip-Hop',
406 },
407 'expected_warnings': ['Unable to download SMIL file'],
408 }, {
409 'url': 'http://www.vevo.com/watch/genre/rock?index=0',
410 'only_matching': True,
411 }]
412
413 def _real_extract(self, url):
414 mobj = re.match(self._VALID_URL, url)
415 playlist_id = mobj.group('id')
416 playlist_kind = mobj.group('kind')
417
418 webpage = self._download_webpage(url, playlist_id)
419
420 qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
421 index = qs.get('index', [None])[0]
422
423 if index:
424 video_id = self._search_regex(
425 r'<meta[^>]+content=(["\'])vevo://video/(?P<id>.+?)\1[^>]*>',
426 webpage, 'video id', default=None, group='id')
427 if video_id:
428 return self.url_result('vevo:%s' % video_id, VevoIE.ie_key())
429
430 playlists = self._extract_json(webpage, playlist_id, '%ss' % playlist_kind)
431
432 playlist = (list(playlists.values())[0]
433 if playlist_kind == 'playlist' else playlists[playlist_id])
434
435 entries = [
436 self.url_result('vevo:%s' % src, VevoIE.ie_key())
437 for src in playlist['isrcs']]
438
439 return self.playlist_result(
440 entries, playlist.get('playlistId') or playlist_id,
441 playlist.get('name'), playlist.get('description'))