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