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 2012',
 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/channels/keypeele/75629013',
 126             'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
 127             'note': 'Video is freely available via original URL '
 128                     'and protected with password when accessed via http://vimeo.com/75629013',
 129             'info_dict': {
 130                 'id': '75629013',
 131                 'ext': 'mp4',
 132                 'title': 'Key & Peele: Terrorist Interrogation',
 133                 'description': 'md5:8678b246399b070816b12313e8b4eb5c',
 134                 'uploader_id': 'atencio',
 135                 'uploader': 'Peter Atencio',
 136                 'duration': 187,
 137             },
 138         },
 139         {
 140             'url': 'http://vimeo.com/76979871',
 141             'md5': '3363dd6ffebe3784d56f4132317fd446',
 142             'note': 'Video with subtitles',
 143             'info_dict': {
 144                 'id': '76979871',
 145                 'ext': 'mp4',
 146                 'title': 'The New Vimeo Player (You Know, For Videos)',
 147                 'description': 'md5:2ec900bf97c3f389378a96aee11260ea',
 148                 'upload_date': '20131015',
 149                 'uploader_id': 'staff',
 150                 'uploader': 'Vimeo Staff',
 151                 'duration': 62,
 152             }
 153         },
 154         {
 155             'note': 'video player needs Referer',
 156             'url': 'http://vimeo.com/user22258446/review/91613211/13f927e053',
 157             'md5': '6295fdab8f4bf6a002d058b2c6dce276',
 158             'info_dict': {
 159                 'id': '91613211',
 160                 'ext': 'mp4',
 161                 'title': 'Death by dogma versus assembling agile - Sander Hoogendoorn',
 162                 'uploader': 'DevWeek Events',
 163                 'duration': 2773,
 164                 'thumbnail': 're:^https?://.*\.jpg$',
 165             }
 166         }
 167     ]
 168
 169     @classmethod
 170     def suitable(cls, url):
 171         if VimeoChannelIE.suitable(url):
 172             # Otherwise channel urls like http://vimeo.com/channels/31259 would
 173             # match
 174             return False
 175         else:
 176             return super(VimeoIE, cls).suitable(url)
 177
 178     def _verify_video_password(self, url, video_id, webpage):
 179         password = self._downloader.params.get('videopassword', None)
 180         if password is None:
 181             raise ExtractorError('This video is protected by a password, use the --video-password option')
 182         token = self._search_regex(r'xsrft: \'(.*?)\'', webpage, 'login token')
 183         data = compat_urllib_parse.urlencode({
 184             'password': password,
 185             'token': token,
 186         })
 187         # I didn't manage to use the password with https
 188         if url.startswith('https'):
 189             pass_url = url.replace('https', 'http')
 190         else:
 191             pass_url = url
 192         password_request = compat_urllib_request.Request(pass_url + '/password', data)
 193         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 194         password_request.add_header('Cookie', 'xsrft=%s' % token)
 195         self._download_webpage(password_request, video_id,
 196                                'Verifying the password',
 197                                'Wrong password')
 198
 199     def _verify_player_video_password(self, url, video_id):
 200         password = self._downloader.params.get('videopassword', None)
 201         if password is None:
 202             raise ExtractorError('This video is protected by a password, use the --video-password option')
 203         data = compat_urllib_parse.urlencode({'password': password})
 204         pass_url = url + '/check-password'
 205         password_request = compat_urllib_request.Request(pass_url, data)
 206         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 207         return self._download_json(
 208             password_request, video_id,
 209             'Verifying the password',
 210             'Wrong password')
 211
 212     def _real_initialize(self):
 213         self._login()
 214
 215     def _real_extract(self, url):
 216         url, data = unsmuggle_url(url)
 217         headers = std_headers
 218         if data is not None:
 219             headers = headers.copy()
 220             headers.update(data)
 221         if 'Referer' not in headers:
 222             headers['Referer'] = url
 223
 224         # Extract ID from URL
 225         mobj = re.match(self._VALID_URL, url)
 226         video_id = mobj.group('id')
 227         if mobj.group('pro') or mobj.group('player'):
 228             url = 'http://player.vimeo.com/video/' + video_id
 229
 230         # Retrieve video webpage to extract further information
 231         request = compat_urllib_request.Request(url, None, headers)
 232         try:
 233             webpage = self._download_webpage(request, video_id)
 234         except ExtractorError as ee:
 235             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 236                 errmsg = ee.cause.read()
 237                 if b'Because of its privacy settings, this video cannot be played here' in errmsg:
 238                     raise ExtractorError(
 239                         'Cannot download embed-only video without embedding '
 240                         'URL. Please call youtube-dl with the URL of the page '
 241                         'that embeds this video.',
 242                         expected=True)
 243             raise
 244
 245         # Now we begin extracting as much information as we can from what we
 246         # retrieved. First we extract the information common to all extractors,
 247         # and latter we extract those that are Vimeo specific.
 248         self.report_extraction(video_id)
 249
 250         # Extract the config JSON
 251         try:
 252             try:
 253                 config_url = self._html_search_regex(
 254                     r' data-config-url="(.+?)"', webpage, 'config URL')
 255                 config_json = self._download_webpage(config_url, video_id)
 256                 config = json.loads(config_json)
 257             except RegexNotFoundError:
 258                 # For pro videos or player.vimeo.com urls
 259                 # We try to find out to which variable is assigned the config dic
 260                 m_variable_name = re.search('(\w)\.video\.id', webpage)
 261                 if m_variable_name is not None:
 262                     config_re = r'%s=({.+?});' % re.escape(m_variable_name.group(1))
 263                 else:
 264                     config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
 265                 config = self._search_regex(config_re, webpage, 'info section',
 266                     flags=re.DOTALL)
 267                 config = json.loads(config)
 268         except Exception as e:
 269             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
 270                 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
 271
 272             if re.search('<form[^>]+?id="pw_form"', webpage) is not None:
 273                 self._verify_video_password(url, video_id, webpage)
 274                 return self._real_extract(url)
 275             else:
 276                 raise ExtractorError('Unable to extract info section',
 277                                      cause=e)
 278         else:
 279             if config.get('view') == 4:
 280                 config = self._verify_player_video_password(url, video_id)
 281
 282         # Extract title
 283         video_title = config["video"]["title"]
 284
 285         # Extract uploader and uploader_id
 286         video_uploader = config["video"]["owner"]["name"]
 287         video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] if config["video"]["owner"]["url"] else None
 288
 289         # Extract video thumbnail
 290         video_thumbnail = config["video"].get("thumbnail")
 291         if video_thumbnail is None:
 292             video_thumbs = config["video"].get("thumbs")
 293             if video_thumbs and isinstance(video_thumbs, dict):
 294                 _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1]
 295
 296         # Extract video description
 297         video_description = None
 298         try:
 299             video_description = get_element_by_attribute("class", "description_wrapper", webpage)
 300             if video_description:
 301                 video_description = clean_html(video_description)
 302         except AssertionError as err:
 303             # On some pages like (http://player.vimeo.com/video/54469442) the
 304             # html tags are not closed, python 2.6 cannot handle it
 305             if err.args[0] == 'we should not get here!':
 306                 pass
 307             else:
 308                 raise
 309
 310         # Extract video duration
 311         video_duration = int_or_none(config["video"].get("duration"))
 312
 313         # Extract upload date
 314         video_upload_date = None
 315         mobj = re.search(r'<meta itemprop="dateCreated" content="(\d{4})-(\d{2})-(\d{2})T', webpage)
 316         if mobj is not None:
 317             video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3)
 318
 319         try:
 320             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
 321             like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
 322             comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
 323         except RegexNotFoundError:
 324             # This info is only available in vimeo.com/{id} urls
 325             view_count = None
 326             like_count = None
 327             comment_count = None
 328
 329         # Vimeo specific: extract request signature and timestamp
 330         sig = config['request']['signature']
 331         timestamp = config['request']['timestamp']
 332
 333         # Vimeo specific: extract video codec and quality information
 334         # First consider quality, then codecs, then take everything
 335         codecs = [('vp6', 'flv'), ('vp8', 'flv'), ('h264', 'mp4')]
 336         files = {'hd': [], 'sd': [], 'other': []}
 337         config_files = config["video"].get("files") or config["request"].get("files")
 338         for codec_name, codec_extension in codecs:
 339             for quality in config_files.get(codec_name, []):
 340                 format_id = '-'.join((codec_name, quality)).lower()
 341                 key = quality if quality in files else 'other'
 342                 video_url = None
 343                 if isinstance(config_files[codec_name], dict):
 344                     file_info = config_files[codec_name][quality]
 345                     video_url = file_info.get('url')
 346                 else:
 347                     file_info = {}
 348                 if video_url is None:
 349                     video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
 350                         % (video_id, sig, timestamp, quality, codec_name.upper())
 351
 352                 files[key].append({
 353                     'ext': codec_extension,
 354                     'url': video_url,
 355                     'format_id': format_id,
 356                     'width': file_info.get('width'),
 357                     'height': file_info.get('height'),
 358                 })
 359         formats = []
 360         for key in ('other', 'sd', 'hd'):
 361             formats += files[key]
 362         if len(formats) == 0:
 363             raise ExtractorError('No known codec found')
 364
 365         subtitles = {}
 366         text_tracks = config['request'].get('text_tracks')
 367         if text_tracks:
 368             for tt in text_tracks:
 369                 subtitles[tt['lang']] = 'http://vimeo.com' + tt['url']
 370
 371         video_subtitles = self.extract_subtitles(video_id, subtitles)
 372         if self._downloader.params.get('listsubtitles', False):
 373             self._list_available_subtitles(video_id, subtitles)
 374             return
 375
 376         return {
 377             'id': video_id,
 378             'uploader': video_uploader,
 379             'uploader_id': video_uploader_id,
 380             'upload_date': video_upload_date,
 381             'title': video_title,
 382             'thumbnail': video_thumbnail,
 383             'description': video_description,
 384             'duration': video_duration,
 385             'formats': formats,
 386             'webpage_url': url,
 387             'view_count': view_count,
 388             'like_count': like_count,
 389             'comment_count': comment_count,
 390             'subtitles': video_subtitles,
 391         }
 392
 393
 394 class VimeoChannelIE(InfoExtractor):
 395     IE_NAME = 'vimeo:channel'
 396     _VALID_URL = r'(?:https?://)?vimeo\.com/channels/(?P<id>[^/]+)/?(\?.*)?$'
 397     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 398     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
 399
 400     def _page_url(self, base_url, pagenum):
 401         return '%s/videos/page:%d/' % (base_url, pagenum)
 402
 403     def _extract_list_title(self, webpage):
 404         return self._html_search_regex(self._TITLE_RE, webpage, 'list title')
 405
 406     def _extract_videos(self, list_id, base_url):
 407         video_ids = []
 408         for pagenum in itertools.count(1):
 409             webpage = self._download_webpage(
 410                 self._page_url(base_url, pagenum), list_id,
 411                 'Downloading page %s' % pagenum)
 412             video_ids.extend(re.findall(r'id="clip_(\d+?)"', webpage))
 413             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 414                 break
 415
 416         entries = [self.url_result('http://vimeo.com/%s' % video_id, 'Vimeo')
 417                    for video_id in video_ids]
 418         return {'_type': 'playlist',
 419                 'id': list_id,
 420                 'title': self._extract_list_title(webpage),
 421                 'entries': entries,
 422                 }
 423
 424     def _real_extract(self, url):
 425         mobj = re.match(self._VALID_URL, url)
 426         channel_id = mobj.group('id')
 427         return self._extract_videos(channel_id, 'http://vimeo.com/channels/%s' % channel_id)
 428
 429
 430 class VimeoUserIE(VimeoChannelIE):
 431     IE_NAME = 'vimeo:user'
 432     _VALID_URL = r'(?:https?://)?vimeo\.com/(?P<name>[^/]+)(?:/videos|[#?]|$)'
 433     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
 434
 435     @classmethod
 436     def suitable(cls, url):
 437         if VimeoChannelIE.suitable(url) or VimeoIE.suitable(url) or VimeoAlbumIE.suitable(url) or VimeoGroupsIE.suitable(url):
 438             return False
 439         return super(VimeoUserIE, cls).suitable(url)
 440
 441     def _real_extract(self, url):
 442         mobj = re.match(self._VALID_URL, url)
 443         name = mobj.group('name')
 444         return self._extract_videos(name, 'http://vimeo.com/%s' % name)
 445
 446
 447 class VimeoAlbumIE(VimeoChannelIE):
 448     IE_NAME = 'vimeo:album'
 449     _VALID_URL = r'(?:https?://)?vimeo\.com/album/(?P<id>\d+)'
 450     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
 451
 452     def _page_url(self, base_url, pagenum):
 453         return '%s/page:%d/' % (base_url, pagenum)
 454
 455     def _real_extract(self, url):
 456         mobj = re.match(self._VALID_URL, url)
 457         album_id = mobj.group('id')
 458         return self._extract_videos(album_id, 'http://vimeo.com/album/%s' % album_id)
 459
 460
 461 class VimeoGroupsIE(VimeoAlbumIE):
 462     IE_NAME = 'vimeo:group'
 463     _VALID_URL = r'(?:https?://)?vimeo\.com/groups/(?P<name>[^/]+)'
 464
 465     def _extract_list_title(self, webpage):
 466         return self._og_search_title(webpage)
 467
 468     def _real_extract(self, url):
 469         mobj = re.match(self._VALID_URL, url)
 470         name = mobj.group('name')
 471         return self._extract_videos(name, 'http://vimeo.com/groups/%s' % name)
 472
 473
 474 class VimeoReviewIE(InfoExtractor):
 475     IE_NAME = 'vimeo:review'
 476     IE_DESC = 'Review pages on vimeo'
 477     _VALID_URL = r'(?:https?://)?vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
 478     _TEST = {
 479         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
 480         'file': '75524534.mp4',
 481         'md5': 'c507a72f780cacc12b2248bb4006d253',
 482         'info_dict': {
 483             'title': "DICK HARDWICK 'Comedian'",
 484             'uploader': 'Richard Hardwick',
 485         }
 486     }
 487
 488     def _real_extract(self, url):
 489         mobj = re.match(self._VALID_URL, url)
 490         video_id = mobj.group('id')
 491         player_url = 'https://player.vimeo.com/player/' + video_id
 492         return self.url_result(player_url, 'Vimeo', video_id)
 493
 494
 495 class VimeoWatchLaterIE(VimeoBaseInfoExtractor, VimeoChannelIE):
 496     IE_NAME = 'vimeo:watchlater'
 497     IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)'
 498     _VALID_URL = r'https?://vimeo\.com/home/watchlater|:vimeowatchlater'
 499     _LOGIN_REQUIRED = True
 500     _TITLE_RE = r'href="/home/watchlater".*?>(.*?)<'
 501
 502     def _real_initialize(self):
 503         self._login()
 504
 505     def _page_url(self, base_url, pagenum):
 506         url = '%s/page:%d/' % (base_url, pagenum)
 507         request = compat_urllib_request.Request(url)
 508         # Set the header to get a partial html page with the ids,
 509         # the normal page doesn't contain them.
 510         request.add_header('X-Requested-With', 'XMLHttpRequest')
 511         return request
 512
 513     def _real_extract(self, url):
 514         return self._extract_videos('watchlater', 'https://vimeo.com/home/watchlater')