Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/vimeo.py

   1 import json
   2 import re
   3 import itertools
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     compat_urllib_parse,
   8     compat_urllib_request,
   9
  10     clean_html,
  11     get_element_by_attribute,
  12     ExtractorError,
  13     std_headers,
  14 )
  15
  16 class VimeoIE(InfoExtractor):
  17     """Information extractor for vimeo.com."""
  18
  19     # _VALID_URL matches Vimeo URLs
  20     _VALID_URL = r'(?P<proto>https?://)?(?:(?:www|player)\.)?vimeo(?P<pro>pro)?\.com/(?:(?:(?:groups|album)/[^/]+)|(?:.*?)/)?(?P<direct_link>play_redirect_hls\?clip_id=)?(?:videos?/)?(?P<id>[0-9]+)(?:[?].*)?$'
  21     _NETRC_MACHINE = 'vimeo'
  22     IE_NAME = u'vimeo'
  23     _TESTS = [
  24         {
  25             u'url': u'http://vimeo.com/56015672',
  26             u'file': u'56015672.mp4',
  27             u'md5': u'8879b6cc097e987f02484baf890129e5',
  28             u'info_dict': {
  29                 u"upload_date": u"20121220",
  30                 u"description": u"This is a test case for youtube-dl.\nFor more information, see github.com/rg3/youtube-dl\nTest chars: \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
  31                 u"uploader_id": u"user7108434",
  32                 u"uploader": u"Filippo Valsorda",
  33                 u"title": u"youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
  34             },
  35         },
  36         {
  37             u'url': u'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
  38             u'file': u'68093876.mp4',
  39             u'md5': u'3b5ca6aa22b60dfeeadf50b72e44ed82',
  40             u'note': u'Vimeo Pro video (#1197)',
  41             u'info_dict': {
  42                 u'uploader_id': u'openstreetmapus',
  43                 u'uploader': u'OpenStreetMap US',
  44                 u'title': u'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
  45             },
  46         },
  47     ]
  48
  49     def _login(self):
  50         (username, password) = self._get_login_info()
  51         if username is None:
  52             return
  53         self.report_login()
  54         login_url = 'https://vimeo.com/log_in'
  55         webpage = self._download_webpage(login_url, None, False)
  56         token = re.search(r'xsrft: \'(.*?)\'', webpage).group(1)
  57         data = compat_urllib_parse.urlencode({'email': username,
  58                                               'password': password,
  59                                               'action': 'login',
  60                                               'service': 'vimeo',
  61                                               'token': token,
  62                                               })
  63         login_request = compat_urllib_request.Request(login_url, data)
  64         login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  65         login_request.add_header('Cookie', 'xsrft=%s' % token)
  66         self._download_webpage(login_request, None, False, u'Wrong login info')
  67
  68     def _verify_video_password(self, url, video_id, webpage):
  69         password = self._downloader.params.get('videopassword', None)
  70         if password is None:
  71             raise ExtractorError(u'This video is protected by a password, use the --video-password option')
  72         token = re.search(r'xsrft: \'(.*?)\'', webpage).group(1)
  73         data = compat_urllib_parse.urlencode({'password': password,
  74                                               'token': token})
  75         # I didn't manage to use the password with https
  76         if url.startswith('https'):
  77             pass_url = url.replace('https','http')
  78         else:
  79             pass_url = url
  80         password_request = compat_urllib_request.Request(pass_url+'/password', data)
  81         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  82         password_request.add_header('Cookie', 'xsrft=%s' % token)
  83         self._download_webpage(password_request, video_id,
  84                                u'Verifying the password',
  85                                u'Wrong password')
  86
  87     def _real_initialize(self):
  88         self._login()
  89
  90     def _real_extract(self, url, new_video=True):
  91         # Extract ID from URL
  92         mobj = re.match(self._VALID_URL, url)
  93         if mobj is None:
  94             raise ExtractorError(u'Invalid URL: %s' % url)
  95
  96         video_id = mobj.group('id')
  97         if not mobj.group('proto'):
  98             url = 'https://' + url
  99         elif mobj.group('pro'):
 100             url = 'http://player.vimeo.com/video/' + video_id
 101         elif mobj.group('direct_link'):
 102             url = 'https://vimeo.com/' + video_id
 103
 104         # Retrieve video webpage to extract further information
 105         request = compat_urllib_request.Request(url, None, std_headers)
 106         webpage = self._download_webpage(request, video_id)
 107
 108         # Now we begin extracting as much information as we can from what we
 109         # retrieved. First we extract the information common to all extractors,
 110         # and latter we extract those that are Vimeo specific.
 111         self.report_extraction(video_id)
 112
 113         # Extract the config JSON
 114         try:
 115             config = webpage.split(' = {config:')[1].split(',assets:')[0]
 116             config = json.loads(config)
 117         except:
 118             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
 119                 raise ExtractorError(u'The author has restricted the access to this video, try with the "--referer" option')
 120
 121             if re.search('If so please provide the correct password.', webpage):
 122                 self._verify_video_password(url, video_id, webpage)
 123                 return self._real_extract(url)
 124             else:
 125                 raise ExtractorError(u'Unable to extract info section')
 126
 127         # Extract title
 128         video_title = config["video"]["title"]
 129
 130         # Extract uploader and uploader_id
 131         video_uploader = config["video"]["owner"]["name"]
 132         video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] if config["video"]["owner"]["url"] else None
 133
 134         # Extract video thumbnail
 135         video_thumbnail = config["video"]["thumbnail"]
 136
 137         # Extract video description
 138         video_description = get_element_by_attribute("itemprop", "description", webpage)
 139         if video_description: video_description = clean_html(video_description)
 140         else: video_description = u''
 141
 142         # Extract upload date
 143         video_upload_date = None
 144         mobj = re.search(r'<meta itemprop="dateCreated" content="(\d{4})-(\d{2})-(\d{2})T', webpage)
 145         if mobj is not None:
 146             video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3)
 147
 148         # Vimeo specific: extract request signature and timestamp
 149         sig = config['request']['signature']
 150         timestamp = config['request']['timestamp']
 151
 152         # Vimeo specific: extract video codec and quality information
 153         # First consider quality, then codecs, then take everything
 154         # TODO bind to format param
 155         codecs = [('h264', 'mp4'), ('vp8', 'flv'), ('vp6', 'flv')]
 156         files = { 'hd': [], 'sd': [], 'other': []}
 157         for codec_name, codec_extension in codecs:
 158             if codec_name in config["video"]["files"]:
 159                 if 'hd' in config["video"]["files"][codec_name]:
 160                     files['hd'].append((codec_name, codec_extension, 'hd'))
 161                 elif 'sd' in config["video"]["files"][codec_name]:
 162                     files['sd'].append((codec_name, codec_extension, 'sd'))
 163                 else:
 164                     files['other'].append((codec_name, codec_extension, config["video"]["files"][codec_name][0]))
 165
 166         for quality in ('hd', 'sd', 'other'):
 167             if len(files[quality]) > 0:
 168                 video_quality = files[quality][0][2]
 169                 video_codec = files[quality][0][0]
 170                 video_extension = files[quality][0][1]
 171                 self.to_screen(u'%s: Downloading %s file at %s quality' % (video_id, video_codec.upper(), video_quality))
 172                 break
 173         else:
 174             raise ExtractorError(u'No known codec found')
 175
 176         video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
 177                     %(video_id, sig, timestamp, video_quality, video_codec.upper())
 178
 179         return [{
 180             'id':       video_id,
 181             'url':      video_url,
 182             'uploader': video_uploader,
 183             'uploader_id': video_uploader_id,
 184             'upload_date':  video_upload_date,
 185             'title':    video_title,
 186             'ext':      video_extension,
 187             'thumbnail':    video_thumbnail,
 188             'description':  video_description,
 189         }]
 190
 191
 192 class VimeoChannelIE(InfoExtractor):
 193     IE_NAME = u'vimeo:channel'
 194     _VALID_URL = r'(?:https?://)?vimeo.\com/channels/(?P<id>[^/]+)'
 195     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 196
 197     def _real_extract(self, url):
 198         mobj = re.match(self._VALID_URL, url)
 199         channel_id =  mobj.group('id')
 200         video_ids = []
 201
 202         for pagenum in itertools.count(1):
 203             webpage = self._download_webpage('http://vimeo.com/channels/%s/videos/page:%d' % (channel_id, pagenum),
 204                                              channel_id, u'Downloading page %s' % pagenum)
 205             video_ids.extend(re.findall(r'id="clip_(\d+?)"', webpage))
 206             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 207                 break
 208
 209         entries = [self.url_result('http://vimeo.com/%s' % video_id, 'Vimeo')
 210                    for video_id in video_ids]
 211         channel_title = self._html_search_regex(r'<a href="/channels/%s">(.*?)</a>' % channel_id,
 212                                                 webpage, u'channel title')
 213         return {'_type': 'playlist',
 214                 'id': channel_id,
 215                 'title': channel_title,
 216                 'entries': entries,
 217                 }