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