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