]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/vimeo.py
06b0bed41e68401a8667cbabdca0d9796ea8ca3d
[youtubedl] / youtube_dl / extractor / vimeo.py
1 # encoding: utf-8
2 from __future__ import unicode_literals
3
4 import json
5 import re
6 import itertools
7
8 from .common import InfoExtractor
9 from .subtitles import SubtitlesInfoExtractor
10 from ..compat import (
11 compat_HTTPError,
12 compat_urllib_parse,
13 compat_urllib_request,
14 compat_urlparse,
15 )
16 from ..utils import (
17 ExtractorError,
18 InAdvancePagedList,
19 int_or_none,
20 RegexNotFoundError,
21 std_headers,
22 unsmuggle_url,
23 urlencode_postdata,
24 )
25
26
27 class VimeoBaseInfoExtractor(InfoExtractor):
28 _NETRC_MACHINE = 'vimeo'
29 _LOGIN_REQUIRED = False
30
31 def _login(self):
32 (username, password) = self._get_login_info()
33 if username is None:
34 if self._LOGIN_REQUIRED:
35 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
36 return
37 self.report_login()
38 login_url = 'https://vimeo.com/log_in'
39 webpage = self._download_webpage(login_url, None, False)
40 token = self._search_regex(r'xsrft: \'(.*?)\'', webpage, 'login token')
41 data = urlencode_postdata({
42 'email': username,
43 'password': password,
44 'action': 'login',
45 'service': 'vimeo',
46 'token': token,
47 })
48 login_request = compat_urllib_request.Request(login_url, data)
49 login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
50 login_request.add_header('Cookie', 'xsrft=%s' % token)
51 self._download_webpage(login_request, None, False, 'Wrong login info')
52
53
54 class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
55 """Information extractor for vimeo.com."""
56
57 # _VALID_URL matches Vimeo URLs
58 _VALID_URL = r'''(?x)
59 https?://
60 (?:(?:www|(?P<player>player))\.)?
61 vimeo(?P<pro>pro)?\.com/
62 (?!channels/[^/?#]+/?(?:$|[?#])|album/)
63 (?:.*?/)?
64 (?:(?:play_redirect_hls|moogaloop\.swf)\?clip_id=)?
65 (?:videos?/)?
66 (?P<id>[0-9]+)
67 /?(?:[?&].*)?(?:[#].*)?$'''
68 IE_NAME = 'vimeo'
69 _TESTS = [
70 {
71 'url': 'http://vimeo.com/56015672#at=0',
72 'md5': '8879b6cc097e987f02484baf890129e5',
73 'info_dict': {
74 'id': '56015672',
75 'ext': 'mp4',
76 "upload_date": "20121220",
77 "description": "This is a test case for youtube-dl.\nFor more information, see github.com/rg3/youtube-dl\nTest chars: \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
78 "uploader_id": "user7108434",
79 "uploader": "Filippo Valsorda",
80 "title": "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
81 "duration": 10,
82 },
83 },
84 {
85 'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
86 'md5': '3b5ca6aa22b60dfeeadf50b72e44ed82',
87 'note': 'Vimeo Pro video (#1197)',
88 'info_dict': {
89 'id': '68093876',
90 'ext': 'mp4',
91 'uploader_id': 'openstreetmapus',
92 'uploader': 'OpenStreetMap US',
93 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
94 'description': 'md5:380943ec71b89736ff4bf27183233d09',
95 'duration': 1595,
96 },
97 },
98 {
99 'url': 'http://player.vimeo.com/video/54469442',
100 'md5': '619b811a4417aa4abe78dc653becf511',
101 'note': 'Videos that embed the url in the player page',
102 'info_dict': {
103 'id': '54469442',
104 'ext': 'mp4',
105 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012',
106 'uploader': 'The BLN & Business of Software',
107 'uploader_id': 'theblnbusinessofsoftware',
108 'duration': 3610,
109 'description': None,
110 },
111 },
112 {
113 'url': 'http://vimeo.com/68375962',
114 'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
115 'note': 'Video protected with password',
116 'info_dict': {
117 'id': '68375962',
118 'ext': 'mp4',
119 'title': 'youtube-dl password protected test video',
120 'upload_date': '20130614',
121 'uploader_id': 'user18948128',
122 'uploader': 'Jaime Marquínez Ferrándiz',
123 'duration': 10,
124 'description': 'This is "youtube-dl password protected test video" by Jaime Marquínez Ferrándiz on Vimeo, the home for high quality videos and the people who love them.',
125 },
126 'params': {
127 'videopassword': 'youtube-dl',
128 },
129 },
130 {
131 'url': 'http://vimeo.com/channels/keypeele/75629013',
132 'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
133 'note': 'Video is freely available via original URL '
134 'and protected with password when accessed via http://vimeo.com/75629013',
135 'info_dict': {
136 'id': '75629013',
137 'ext': 'mp4',
138 'title': 'Key & Peele: Terrorist Interrogation',
139 'description': 'md5:8678b246399b070816b12313e8b4eb5c',
140 'uploader_id': 'atencio',
141 'uploader': 'Peter Atencio',
142 'duration': 187,
143 },
144 },
145 {
146 'url': 'http://vimeo.com/76979871',
147 'md5': '3363dd6ffebe3784d56f4132317fd446',
148 'note': 'Video with subtitles',
149 'info_dict': {
150 'id': '76979871',
151 'ext': 'mp4',
152 'title': 'The New Vimeo Player (You Know, For Videos)',
153 'description': 'md5:2ec900bf97c3f389378a96aee11260ea',
154 'upload_date': '20131015',
155 'uploader_id': 'staff',
156 'uploader': 'Vimeo Staff',
157 'duration': 62,
158 }
159 },
160 {
161 # from https://www.ouya.tv/game/Pier-Solar-and-the-Great-Architects/
162 'url': 'https://player.vimeo.com/video/98044508',
163 'note': 'The js code contains assignments to the same variable as the config',
164 'info_dict': {
165 'id': '98044508',
166 'ext': 'mp4',
167 'title': 'Pier Solar OUYA Official Trailer',
168 'uploader': 'Tulio Gonçalves',
169 'uploader_id': 'user28849593',
170 },
171 },
172 ]
173
174 def _verify_video_password(self, url, video_id, webpage):
175 password = self._downloader.params.get('videopassword', None)
176 if password is None:
177 raise ExtractorError('This video is protected by a password, use the --video-password option')
178 token = self._search_regex(r'xsrft: \'(.*?)\'', webpage, 'login token')
179 data = compat_urllib_parse.urlencode({
180 'password': password,
181 'token': token,
182 })
183 # I didn't manage to use the password with https
184 if url.startswith('https'):
185 pass_url = url.replace('https', 'http')
186 else:
187 pass_url = url
188 password_request = compat_urllib_request.Request(pass_url + '/password', data)
189 password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
190 password_request.add_header('Cookie', 'xsrft=%s' % token)
191 self._download_webpage(password_request, video_id,
192 'Verifying the password',
193 'Wrong password')
194
195 def _verify_player_video_password(self, url, video_id):
196 password = self._downloader.params.get('videopassword', None)
197 if password is None:
198 raise ExtractorError('This video is protected by a password, use the --video-password option')
199 data = compat_urllib_parse.urlencode({'password': password})
200 pass_url = url + '/check-password'
201 password_request = compat_urllib_request.Request(pass_url, data)
202 password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
203 return self._download_json(
204 password_request, video_id,
205 'Verifying the password',
206 'Wrong password')
207
208 def _real_initialize(self):
209 self._login()
210
211 def _real_extract(self, url):
212 url, data = unsmuggle_url(url)
213 headers = std_headers
214 if data is not None:
215 headers = headers.copy()
216 headers.update(data)
217 if 'Referer' not in headers:
218 headers['Referer'] = url
219
220 # Extract ID from URL
221 mobj = re.match(self._VALID_URL, url)
222 video_id = mobj.group('id')
223 orig_url = url
224 if mobj.group('pro') or mobj.group('player'):
225 url = 'http://player.vimeo.com/video/' + video_id
226
227 # Retrieve video webpage to extract further information
228 request = compat_urllib_request.Request(url, None, headers)
229 try:
230 webpage = self._download_webpage(request, video_id)
231 except ExtractorError as ee:
232 if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
233 errmsg = ee.cause.read()
234 if b'Because of its privacy settings, this video cannot be played here' in errmsg:
235 raise ExtractorError(
236 'Cannot download embed-only video without embedding '
237 'URL. Please call youtube-dl with the URL of the page '
238 'that embeds this video.',
239 expected=True)
240 raise
241
242 # Now we begin extracting as much information as we can from what we
243 # retrieved. First we extract the information common to all extractors,
244 # and latter we extract those that are Vimeo specific.
245 self.report_extraction(video_id)
246
247 # Extract the config JSON
248 try:
249 try:
250 config_url = self._html_search_regex(
251 r' data-config-url="(.+?)"', webpage, 'config URL')
252 config_json = self._download_webpage(config_url, video_id)
253 config = json.loads(config_json)
254 except RegexNotFoundError:
255 # For pro videos or player.vimeo.com urls
256 # We try to find out to which variable is assigned the config dic
257 m_variable_name = re.search('(\w)\.video\.id', webpage)
258 if m_variable_name is not None:
259 config_re = r'%s=({[^}].+?});' % re.escape(m_variable_name.group(1))
260 else:
261 config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
262 config = self._search_regex(config_re, webpage, 'info section',
263 flags=re.DOTALL)
264 config = json.loads(config)
265 except Exception as e:
266 if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
267 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
268
269 if re.search('<form[^>]+?id="pw_form"', webpage) is not None:
270 self._verify_video_password(url, video_id, webpage)
271 return self._real_extract(url)
272 else:
273 raise ExtractorError('Unable to extract info section',
274 cause=e)
275 else:
276 if config.get('view') == 4:
277 config = self._verify_player_video_password(url, video_id)
278
279 # Extract title
280 video_title = config["video"]["title"]
281
282 # Extract uploader and uploader_id
283 video_uploader = config["video"]["owner"]["name"]
284 video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] if config["video"]["owner"]["url"] else None
285
286 # Extract video thumbnail
287 video_thumbnail = config["video"].get("thumbnail")
288 if video_thumbnail is None:
289 video_thumbs = config["video"].get("thumbs")
290 if video_thumbs and isinstance(video_thumbs, dict):
291 _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1]
292
293 # Extract video description
294
295 video_description = self._html_search_regex(
296 r'(?s)<div\s+class="[^"]*description[^"]*"[^>]*>(.*?)</div>',
297 webpage, 'description', default=None)
298 if not video_description:
299 video_description = self._html_search_meta(
300 'description', webpage, default=None)
301 if not video_description and mobj.group('pro'):
302 orig_webpage = self._download_webpage(
303 orig_url, video_id,
304 note='Downloading webpage for description',
305 fatal=False)
306 if orig_webpage:
307 video_description = self._html_search_meta(
308 'description', orig_webpage, default=None)
309 if not video_description and not mobj.group('player'):
310 self._downloader.report_warning('Cannot find video description')
311
312 # Extract video duration
313 video_duration = int_or_none(config["video"].get("duration"))
314
315 # Extract upload date
316 video_upload_date = None
317 mobj = re.search(r'<meta itemprop="dateCreated" content="(\d{4})-(\d{2})-(\d{2})T', webpage)
318 if mobj is not None:
319 video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3)
320
321 try:
322 view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
323 like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
324 comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
325 except RegexNotFoundError:
326 # This info is only available in vimeo.com/{id} urls
327 view_count = None
328 like_count = None
329 comment_count = None
330
331 # Vimeo specific: extract request signature and timestamp
332 sig = config['request']['signature']
333 timestamp = config['request']['timestamp']
334
335 # Vimeo specific: extract video codec and quality information
336 # First consider quality, then codecs, then take everything
337 codecs = [('vp6', 'flv'), ('vp8', 'flv'), ('h264', 'mp4')]
338 files = {'hd': [], 'sd': [], 'other': []}
339 config_files = config["video"].get("files") or config["request"].get("files")
340 for codec_name, codec_extension in codecs:
341 for quality in config_files.get(codec_name, []):
342 format_id = '-'.join((codec_name, quality)).lower()
343 key = quality if quality in files else 'other'
344 video_url = None
345 if isinstance(config_files[codec_name], dict):
346 file_info = config_files[codec_name][quality]
347 video_url = file_info.get('url')
348 else:
349 file_info = {}
350 if video_url is None:
351 video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
352 % (video_id, sig, timestamp, quality, codec_name.upper())
353
354 files[key].append({
355 'ext': codec_extension,
356 'url': video_url,
357 'format_id': format_id,
358 'width': file_info.get('width'),
359 'height': file_info.get('height'),
360 })
361 formats = []
362 for key in ('other', 'sd', 'hd'):
363 formats += files[key]
364 if len(formats) == 0:
365 raise ExtractorError('No known codec found')
366
367 subtitles = {}
368 text_tracks = config['request'].get('text_tracks')
369 if text_tracks:
370 for tt in text_tracks:
371 subtitles[tt['lang']] = 'http://vimeo.com' + tt['url']
372
373 video_subtitles = self.extract_subtitles(video_id, subtitles)
374 if self._downloader.params.get('listsubtitles', False):
375 self._list_available_subtitles(video_id, subtitles)
376 return
377
378 return {
379 'id': video_id,
380 'uploader': video_uploader,
381 'uploader_id': video_uploader_id,
382 'upload_date': video_upload_date,
383 'title': video_title,
384 'thumbnail': video_thumbnail,
385 'description': video_description,
386 'duration': video_duration,
387 'formats': formats,
388 'webpage_url': url,
389 'view_count': view_count,
390 'like_count': like_count,
391 'comment_count': comment_count,
392 'subtitles': video_subtitles,
393 }
394
395
396 class VimeoChannelIE(InfoExtractor):
397 IE_NAME = 'vimeo:channel'
398 _VALID_URL = r'https?://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])'
399 _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
400 _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
401 _TESTS = [{
402 'url': 'http://vimeo.com/channels/tributes',
403 'info_dict': {
404 'title': 'Vimeo Tributes',
405 },
406 'playlist_mincount': 25,
407 }]
408
409 def _page_url(self, base_url, pagenum):
410 return '%s/videos/page:%d/' % (base_url, pagenum)
411
412 def _extract_list_title(self, webpage):
413 return self._html_search_regex(self._TITLE_RE, webpage, 'list title')
414
415 def _extract_videos(self, list_id, base_url):
416 video_ids = []
417 for pagenum in itertools.count(1):
418 webpage = self._download_webpage(
419 self._page_url(base_url, pagenum), list_id,
420 'Downloading page %s' % pagenum)
421 video_ids.extend(re.findall(r'id="clip_(\d+?)"', webpage))
422 if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
423 break
424
425 entries = [self.url_result('http://vimeo.com/%s' % video_id, 'Vimeo')
426 for video_id in video_ids]
427 return {'_type': 'playlist',
428 'id': list_id,
429 'title': self._extract_list_title(webpage),
430 'entries': entries,
431 }
432
433 def _real_extract(self, url):
434 mobj = re.match(self._VALID_URL, url)
435 channel_id = mobj.group('id')
436 return self._extract_videos(channel_id, 'http://vimeo.com/channels/%s' % channel_id)
437
438
439 class VimeoUserIE(VimeoChannelIE):
440 IE_NAME = 'vimeo:user'
441 _VALID_URL = r'https?://vimeo\.com/(?![0-9]+(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
442 _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
443 _TESTS = [{
444 'url': 'http://vimeo.com/nkistudio/videos',
445 'info_dict': {
446 'title': 'Nki',
447 },
448 'playlist_mincount': 66,
449 }]
450
451 def _real_extract(self, url):
452 mobj = re.match(self._VALID_URL, url)
453 name = mobj.group('name')
454 return self._extract_videos(name, 'http://vimeo.com/%s' % name)
455
456
457 class VimeoAlbumIE(VimeoChannelIE):
458 IE_NAME = 'vimeo:album'
459 _VALID_URL = r'https?://vimeo\.com/album/(?P<id>\d+)'
460 _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
461 _TESTS = [{
462 'url': 'http://vimeo.com/album/2632481',
463 'info_dict': {
464 'title': 'Staff Favorites: November 2013',
465 },
466 'playlist_mincount': 13,
467 }]
468
469 def _page_url(self, base_url, pagenum):
470 return '%s/page:%d/' % (base_url, pagenum)
471
472 def _real_extract(self, url):
473 mobj = re.match(self._VALID_URL, url)
474 album_id = mobj.group('id')
475 return self._extract_videos(album_id, 'http://vimeo.com/album/%s' % album_id)
476
477
478 class VimeoGroupsIE(VimeoAlbumIE):
479 IE_NAME = 'vimeo:group'
480 _VALID_URL = r'(?:https?://)?vimeo\.com/groups/(?P<name>[^/]+)'
481 _TESTS = [{
482 'url': 'http://vimeo.com/groups/rolexawards',
483 'info_dict': {
484 'title': 'Rolex Awards for Enterprise',
485 },
486 'playlist_mincount': 73,
487 }]
488
489 def _extract_list_title(self, webpage):
490 return self._og_search_title(webpage)
491
492 def _real_extract(self, url):
493 mobj = re.match(self._VALID_URL, url)
494 name = mobj.group('name')
495 return self._extract_videos(name, 'http://vimeo.com/groups/%s' % name)
496
497
498 class VimeoReviewIE(InfoExtractor):
499 IE_NAME = 'vimeo:review'
500 IE_DESC = 'Review pages on vimeo'
501 _VALID_URL = r'https?://vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
502 _TESTS = [{
503 'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
504 'file': '75524534.mp4',
505 'md5': 'c507a72f780cacc12b2248bb4006d253',
506 'info_dict': {
507 'title': "DICK HARDWICK 'Comedian'",
508 'uploader': 'Richard Hardwick',
509 }
510 }, {
511 'note': 'video player needs Referer',
512 'url': 'http://vimeo.com/user22258446/review/91613211/13f927e053',
513 'md5': '6295fdab8f4bf6a002d058b2c6dce276',
514 'info_dict': {
515 'id': '91613211',
516 'ext': 'mp4',
517 'title': 're:(?i)^Death by dogma versus assembling agile . Sander Hoogendoorn',
518 'uploader': 'DevWeek Events',
519 'duration': 2773,
520 'thumbnail': 're:^https?://.*\.jpg$',
521 }
522 }]
523
524 def _real_extract(self, url):
525 mobj = re.match(self._VALID_URL, url)
526 video_id = mobj.group('id')
527 player_url = 'https://player.vimeo.com/player/' + video_id
528 return self.url_result(player_url, 'Vimeo', video_id)
529
530
531 class VimeoWatchLaterIE(VimeoBaseInfoExtractor, VimeoChannelIE):
532 IE_NAME = 'vimeo:watchlater'
533 IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)'
534 _VALID_URL = r'https?://vimeo\.com/home/watchlater|:vimeowatchlater'
535 _LOGIN_REQUIRED = True
536 _TITLE_RE = r'href="/home/watchlater".*?>(.*?)<'
537 _TESTS = [{
538 'url': 'http://vimeo.com/home/watchlater',
539 'only_matching': True,
540 }]
541
542 def _real_initialize(self):
543 self._login()
544
545 def _page_url(self, base_url, pagenum):
546 url = '%s/page:%d/' % (base_url, pagenum)
547 request = compat_urllib_request.Request(url)
548 # Set the header to get a partial html page with the ids,
549 # the normal page doesn't contain them.
550 request.add_header('X-Requested-With', 'XMLHttpRequest')
551 return request
552
553 def _real_extract(self, url):
554 return self._extract_videos('watchlater', 'https://vimeo.com/home/watchlater')
555
556
557 class VimeoLikesIE(InfoExtractor):
558 _VALID_URL = r'https?://(?:www\.)?vimeo\.com/user(?P<id>[0-9]+)/likes/?(?:$|[?#]|sort:)'
559 IE_NAME = 'vimeo:likes'
560 IE_DESC = 'Vimeo user likes'
561 _TEST = {
562 'url': 'https://vimeo.com/user755559/likes/',
563 'playlist_mincount': 293,
564 "info_dict": {
565 "description": "See all the videos urza likes",
566 "title": 'Videos urza likes',
567 },
568 }
569
570 def _real_extract(self, url):
571 user_id = self._match_id(url)
572 webpage = self._download_webpage(url, user_id)
573 page_count = self._int(
574 self._search_regex(
575 r'''(?x)<li><a\s+href="[^"]+"\s+data-page="([0-9]+)">
576 .*?</a></li>\s*<li\s+class="pagination_next">
577 ''', webpage, 'page count'),
578 'page count', fatal=True)
579 PAGE_SIZE = 12
580 title = self._html_search_regex(
581 r'(?s)<h1>(.+?)</h1>', webpage, 'title', fatal=False)
582 description = self._html_search_meta('description', webpage)
583
584 def _get_page(idx):
585 page_url = '%s//vimeo.com/user%s/likes/page:%d/sort:date' % (
586 self.http_scheme(), user_id, idx + 1)
587 webpage = self._download_webpage(
588 page_url, user_id,
589 note='Downloading page %d/%d' % (idx + 1, page_count))
590 video_list = self._search_regex(
591 r'(?s)<ol class="js-browse_list[^"]+"[^>]*>(.*?)</ol>',
592 webpage, 'video content')
593 paths = re.findall(
594 r'<li[^>]*>\s*<a\s+href="([^"]+)"', video_list)
595 for path in paths:
596 yield {
597 '_type': 'url',
598 'url': compat_urlparse.urljoin(page_url, path),
599 }
600
601 pl = InAdvancePagedList(_get_page, page_count, PAGE_SIZE)
602
603 return {
604 '_type': 'playlist',
605 'id': 'user%s_likes' % user_id,
606 'title': title,
607 'description': description,
608 'entries': pl,
609 }