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