]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/vimeo.py
Merge branch 'ustream-embed-recorded2' of https://github.com/anovicecodemonkey/youtub...
[youtube-dl] / 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 ..utils import (
11     compat_HTTPError,
12     compat_urllib_parse,
13     compat_urllib_request,
14     clean_html,
15     get_element_by_attribute,
16     ExtractorError,
17     RegexNotFoundError,
18     std_headers,
19     unsmuggle_url,
20     urlencode_postdata,
21     int_or_none,
22 )
23
24
25 class VimeoBaseInfoExtractor(InfoExtractor):
26     _NETRC_MACHINE = 'vimeo'
27     _LOGIN_REQUIRED = False
28
29     def _login(self):
30         (username, password) = self._get_login_info()
31         if username is None:
32             if self._LOGIN_REQUIRED:
33                 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
34             return
35         self.report_login()
36         login_url = 'https://vimeo.com/log_in'
37         webpage = self._download_webpage(login_url, None, False)
38         token = self._search_regex(r'xsrft: \'(.*?)\'', webpage, 'login token')
39         data = urlencode_postdata({
40             'email': username,
41             'password': password,
42             'action': 'login',
43             'service': 'vimeo',
44             'token': token,
45         })
46         login_request = compat_urllib_request.Request(login_url, data)
47         login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
48         login_request.add_header('Cookie', 'xsrft=%s' % token)
49         self._download_webpage(login_request, None, False, 'Wrong login info')
50
51
52 class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
53     """Information extractor for vimeo.com."""
54
55     # _VALID_URL matches Vimeo URLs
56     _VALID_URL = r'''(?x)
57         (?P<proto>(?:https?:)?//)?
58         (?:(?:www|(?P<player>player))\.)?
59         vimeo(?P<pro>pro)?\.com/
60         (?:.*?/)?
61         (?:(?:play_redirect_hls|moogaloop\.swf)\?clip_id=)?
62         (?:videos?/)?
63         (?P<id>[0-9]+)
64         /?(?:[?&].*)?(?:[#].*)?$'''
65     IE_NAME = 'vimeo'
66     _TESTS = [
67         {
68             'url': 'http://vimeo.com/56015672#at=0',
69             'md5': '8879b6cc097e987f02484baf890129e5',
70             'info_dict': {
71                 'id': '56015672',
72                 'ext': 'mp4',
73                 "upload_date": "20121220",
74                 "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",
75                 "uploader_id": "user7108434",
76                 "uploader": "Filippo Valsorda",
77                 "title": "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
78                 "duration": 10,
79             },
80         },
81         {
82             'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
83             'md5': '3b5ca6aa22b60dfeeadf50b72e44ed82',
84             'note': 'Vimeo Pro video (#1197)',
85             'info_dict': {
86                 'id': '68093876',
87                 'ext': 'mp4',
88                 'uploader_id': 'openstreetmapus',
89                 'uploader': 'OpenStreetMap US',
90                 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
91                 'duration': 1595,
92             },
93         },
94         {
95             'url': 'http://player.vimeo.com/video/54469442',
96             'md5': '619b811a4417aa4abe78dc653becf511',
97             'note': 'Videos that embed the url in the player page',
98             'info_dict': {
99                 'id': '54469442',
100                 'ext': 'mp4',
101                 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software',
102                 'uploader': 'The BLN & Business of Software',
103                 'uploader_id': 'theblnbusinessofsoftware',
104                 'duration': 3610,
105             },
106         },
107         {
108             'url': 'http://vimeo.com/68375962',
109             'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
110             'note': 'Video protected with password',
111             'info_dict': {
112                 'id': '68375962',
113                 'ext': 'mp4',
114                 'title': 'youtube-dl password protected test video',
115                 'upload_date': '20130614',
116                 'uploader_id': 'user18948128',
117                 'uploader': 'Jaime Marquínez Ferrándiz',
118                 'duration': 10,
119             },
120             'params': {
121                 'videopassword': 'youtube-dl',
122             },
123         },
124         {
125             'url': 'http://vimeo.com/76979871',
126             'md5': '3363dd6ffebe3784d56f4132317fd446',
127             'note': 'Video with subtitles',
128             'info_dict': {
129                 'id': '76979871',
130                 'ext': 'mp4',
131                 'title': 'The New Vimeo Player (You Know, For Videos)',
132                 'description': 'md5:2ec900bf97c3f389378a96aee11260ea',
133                 'upload_date': '20131015',
134                 'uploader_id': 'staff',
135                 'uploader': 'Vimeo Staff',
136                 'duration': 62,
137             }
138         },
139     ]
140
141     @classmethod
142     def suitable(cls, url):
143         if VimeoChannelIE.suitable(url):
144             # Otherwise channel urls like http://vimeo.com/channels/31259 would
145             # match
146             return False
147         else:
148             return super(VimeoIE, cls).suitable(url)
149
150     def _verify_video_password(self, url, video_id, webpage):
151         password = self._downloader.params.get('videopassword', None)
152         if password is None:
153             raise ExtractorError('This video is protected by a password, use the --video-password option')
154         token = self._search_regex(r'xsrft: \'(.*?)\'', webpage, 'login token')
155         data = compat_urllib_parse.urlencode({
156             'password': password,
157             'token': token,
158         })
159         # I didn't manage to use the password with https
160         if url.startswith('https'):
161             pass_url = url.replace('https', 'http')
162         else:
163             pass_url = url
164         password_request = compat_urllib_request.Request(pass_url + '/password', data)
165         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
166         password_request.add_header('Cookie', 'xsrft=%s' % token)
167         self._download_webpage(password_request, video_id,
168                                'Verifying the password',
169                                'Wrong password')
170
171     def _verify_player_video_password(self, url, video_id):
172         password = self._downloader.params.get('videopassword', None)
173         if password is None:
174             raise ExtractorError('This video is protected by a password, use the --video-password option')
175         data = compat_urllib_parse.urlencode({'password': password})
176         pass_url = url + '/check-password'
177         password_request = compat_urllib_request.Request(pass_url, data)
178         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
179         return self._download_json(
180             password_request, video_id,
181             'Verifying the password',
182             'Wrong password')
183
184     def _real_initialize(self):
185         self._login()
186
187     def _real_extract(self, url):
188         url, data = unsmuggle_url(url)
189         headers = std_headers
190         if data is not None:
191             headers = headers.copy()
192             headers.update(data)
193
194         # Extract ID from URL
195         mobj = re.match(self._VALID_URL, url)
196         video_id = mobj.group('id')
197         if mobj.group('pro') or mobj.group('player'):
198             url = 'http://player.vimeo.com/video/' + video_id
199         else:
200             url = 'https://vimeo.com/' + video_id
201
202         # Retrieve video webpage to extract further information
203         request = compat_urllib_request.Request(url, None, headers)
204         try:
205             webpage = self._download_webpage(request, video_id)
206         except ExtractorError as ee:
207             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
208                 errmsg = ee.cause.read()
209                 if b'Because of its privacy settings, this video cannot be played here' in errmsg:
210                     raise ExtractorError(
211                         'Cannot download embed-only video without embedding '
212                         'URL. Please call youtube-dl with the URL of the page '
213                         'that embeds this video.',
214                         expected=True)
215             raise
216
217         # Now we begin extracting as much information as we can from what we
218         # retrieved. First we extract the information common to all extractors,
219         # and latter we extract those that are Vimeo specific.
220         self.report_extraction(video_id)
221
222         # Extract the config JSON
223         try:
224             try:
225                 config_url = self._html_search_regex(
226                     r' data-config-url="(.+?)"', webpage, 'config URL')
227                 config_json = self._download_webpage(config_url, video_id)
228                 config = json.loads(config_json)
229             except RegexNotFoundError:
230                 # For pro videos or player.vimeo.com urls
231                 # We try to find out to which variable is assigned the config dic
232                 m_variable_name = re.search('(\w)\.video\.id', webpage)
233                 if m_variable_name is not None:
234                     config_re = r'%s=({.+?});' % re.escape(m_variable_name.group(1))
235                 else:
236                     config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
237                 config = self._search_regex(config_re, webpage, 'info section',
238                     flags=re.DOTALL)
239                 config = json.loads(config)
240         except Exception as e:
241             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
242                 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
243
244             if re.search('<form[^>]+?id="pw_form"', webpage) is not None:
245                 self._verify_video_password(url, video_id, webpage)
246                 return self._real_extract(url)
247             else:
248                 raise ExtractorError('Unable to extract info section',
249                                      cause=e)
250         else:
251             if config.get('view') == 4:
252                 config = self._verify_player_video_password(url, video_id)
253
254         # Extract title
255         video_title = config["video"]["title"]
256
257         # Extract uploader and uploader_id
258         video_uploader = config["video"]["owner"]["name"]
259         video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] if config["video"]["owner"]["url"] else None
260
261         # Extract video thumbnail
262         video_thumbnail = config["video"].get("thumbnail")
263         if video_thumbnail is None:
264             video_thumbs = config["video"].get("thumbs")
265             if video_thumbs and isinstance(video_thumbs, dict):
266                 _, video_thumbnail = sorted((int(width), t_url) for (width, t_url) in video_thumbs.items())[-1]
267
268         # Extract video description
269         video_description = None
270         try:
271             video_description = get_element_by_attribute("class", "description_wrapper", webpage)
272             if video_description:
273                 video_description = clean_html(video_description)
274         except AssertionError as err:
275             # On some pages like (http://player.vimeo.com/video/54469442) the
276             # html tags are not closed, python 2.6 cannot handle it
277             if err.args[0] == 'we should not get here!':
278                 pass
279             else:
280                 raise
281
282         # Extract video duration
283         video_duration = int_or_none(config["video"].get("duration"))
284
285         # Extract upload date
286         video_upload_date = None
287         mobj = re.search(r'<meta itemprop="dateCreated" content="(\d{4})-(\d{2})-(\d{2})T', webpage)
288         if mobj is not None:
289             video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3)
290
291         try:
292             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
293             like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
294             comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
295         except RegexNotFoundError:
296             # This info is only available in vimeo.com/{id} urls
297             view_count = None
298             like_count = None
299             comment_count = None
300
301         # Vimeo specific: extract request signature and timestamp
302         sig = config['request']['signature']
303         timestamp = config['request']['timestamp']
304
305         # Vimeo specific: extract video codec and quality information
306         # First consider quality, then codecs, then take everything
307         codecs = [('vp6', 'flv'), ('vp8', 'flv'), ('h264', 'mp4')]
308         files = {'hd': [], 'sd': [], 'other': []}
309         config_files = config["video"].get("files") or config["request"].get("files")
310         for codec_name, codec_extension in codecs:
311             for quality in config_files.get(codec_name, []):
312                 format_id = '-'.join((codec_name, quality)).lower()
313                 key = quality if quality in files else 'other'
314                 video_url = None
315                 if isinstance(config_files[codec_name], dict):
316                     file_info = config_files[codec_name][quality]
317                     video_url = file_info.get('url')
318                 else:
319                     file_info = {}
320                 if video_url is None:
321                     video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
322                         % (video_id, sig, timestamp, quality, codec_name.upper())
323
324                 files[key].append({
325                     'ext': codec_extension,
326                     'url': video_url,
327                     'format_id': format_id,
328                     'width': file_info.get('width'),
329                     'height': file_info.get('height'),
330                 })
331         formats = []
332         for key in ('other', 'sd', 'hd'):
333             formats += files[key]
334         if len(formats) == 0:
335             raise ExtractorError('No known codec found')
336
337         subtitles = {}
338         text_tracks = config['request'].get('text_tracks')
339         if text_tracks:
340             for tt in text_tracks:
341                 subtitles[tt['lang']] = 'http://vimeo.com' + tt['url']
342
343         video_subtitles = self.extract_subtitles(video_id, subtitles)
344         if self._downloader.params.get('listsubtitles', False):
345             self._list_available_subtitles(video_id, subtitles)
346             return
347
348         return {
349             'id': video_id,
350             'uploader': video_uploader,
351             'uploader_id': video_uploader_id,
352             'upload_date': video_upload_date,
353             'title': video_title,
354             'thumbnail': video_thumbnail,
355             'description': video_description,
356             'duration': video_duration,
357             'formats': formats,
358             'webpage_url': url,
359             'view_count': view_count,
360             'like_count': like_count,
361             'comment_count': comment_count,
362             'subtitles': video_subtitles,
363         }
364
365
366 class VimeoChannelIE(InfoExtractor):
367     IE_NAME = 'vimeo:channel'
368     _VALID_URL = r'(?:https?://)?vimeo\.com/channels/(?P<id>[^/]+)/?(\?.*)?$'
369     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
370     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
371
372     def _page_url(self, base_url, pagenum):
373         return '%s/videos/page:%d/' % (base_url, pagenum)
374
375     def _extract_list_title(self, webpage):
376         return self._html_search_regex(self._TITLE_RE, webpage, 'list title')
377
378     def _extract_videos(self, list_id, base_url):
379         video_ids = []
380         for pagenum in itertools.count(1):
381             webpage = self._download_webpage(
382                 self._page_url(base_url, pagenum), list_id,
383                 'Downloading page %s' % pagenum)
384             video_ids.extend(re.findall(r'id="clip_(\d+?)"', webpage))
385             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
386                 break
387
388         entries = [self.url_result('http://vimeo.com/%s' % video_id, 'Vimeo')
389                    for video_id in video_ids]
390         return {'_type': 'playlist',
391                 'id': list_id,
392                 'title': self._extract_list_title(webpage),
393                 'entries': entries,
394                 }
395
396     def _real_extract(self, url):
397         mobj = re.match(self._VALID_URL, url)
398         channel_id = mobj.group('id')
399         return self._extract_videos(channel_id, 'http://vimeo.com/channels/%s' % channel_id)
400
401
402 class VimeoUserIE(VimeoChannelIE):
403     IE_NAME = 'vimeo:user'
404     _VALID_URL = r'(?:https?://)?vimeo\.com/(?P<name>[^/]+)(?:/videos|[#?]|$)'
405     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
406
407     @classmethod
408     def suitable(cls, url):
409         if VimeoChannelIE.suitable(url) or VimeoIE.suitable(url) or VimeoAlbumIE.suitable(url) or VimeoGroupsIE.suitable(url):
410             return False
411         return super(VimeoUserIE, cls).suitable(url)
412
413     def _real_extract(self, url):
414         mobj = re.match(self._VALID_URL, url)
415         name = mobj.group('name')
416         return self._extract_videos(name, 'http://vimeo.com/%s' % name)
417
418
419 class VimeoAlbumIE(VimeoChannelIE):
420     IE_NAME = 'vimeo:album'
421     _VALID_URL = r'(?:https?://)?vimeo\.com/album/(?P<id>\d+)'
422     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
423
424     def _page_url(self, base_url, pagenum):
425         return '%s/page:%d/' % (base_url, pagenum)
426
427     def _real_extract(self, url):
428         mobj = re.match(self._VALID_URL, url)
429         album_id = mobj.group('id')
430         return self._extract_videos(album_id, 'http://vimeo.com/album/%s' % album_id)
431
432
433 class VimeoGroupsIE(VimeoAlbumIE):
434     IE_NAME = 'vimeo:group'
435     _VALID_URL = r'(?:https?://)?vimeo\.com/groups/(?P<name>[^/]+)'
436
437     def _extract_list_title(self, webpage):
438         return self._og_search_title(webpage)
439
440     def _real_extract(self, url):
441         mobj = re.match(self._VALID_URL, url)
442         name = mobj.group('name')
443         return self._extract_videos(name, 'http://vimeo.com/groups/%s' % name)
444
445
446 class VimeoReviewIE(InfoExtractor):
447     IE_NAME = 'vimeo:review'
448     IE_DESC = 'Review pages on vimeo'
449     _VALID_URL = r'(?:https?://)?vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
450     _TEST = {
451         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
452         'file': '75524534.mp4',
453         'md5': 'c507a72f780cacc12b2248bb4006d253',
454         'info_dict': {
455             'title': "DICK HARDWICK 'Comedian'",
456             'uploader': 'Richard Hardwick',
457         }
458     }
459
460     def _real_extract(self, url):
461         mobj = re.match(self._VALID_URL, url)
462         video_id = mobj.group('id')
463         player_url = 'https://player.vimeo.com/player/' + video_id
464         return self.url_result(player_url, 'Vimeo', video_id)
465
466
467 class VimeoWatchLaterIE(VimeoBaseInfoExtractor, VimeoChannelIE):
468     IE_NAME = 'vimeo:watchlater'
469     IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)'
470     _VALID_URL = r'https?://vimeo\.com/home/watchlater|:vimeowatchlater'
471     _LOGIN_REQUIRED = True
472     _TITLE_RE = r'href="/home/watchlater".*?>(.*?)<'
473
474     def _real_initialize(self):
475         self._login()
476
477     def _page_url(self, base_url, pagenum):
478         url = '%s/page:%d/' % (base_url, pagenum)
479         request = compat_urllib_request.Request(url)
480         # Set the header to get a partial html page with the ids,
481         # the normal page doesn't contain them.
482         request.add_header('X-Requested-With', 'XMLHttpRequest')
483         return request
484
485     def _real_extract(self, url):
486         return self._extract_videos('watchlater', 'https://vimeo.com/home/watchlater')