_ Git - youtube-dl/blob - youtube_dl/extractor/vimeo.py

   1 import json
   2 import re
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     compat_urllib_parse,
   7     compat_urllib_request,
   8
   9     clean_html,
  10     get_element_by_attribute,
  11     ExtractorError,
  12     std_headers,
  13 )
  14
  15 class VimeoIE(InfoExtractor):
  16     """Information extractor for vimeo.com."""
  17
  18     # _VALID_URL matches Vimeo URLs
  19     _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]+)(?:[?].*)?$'
  20     IE_NAME = u'vimeo'
  21     _TEST = {
  22         u'url': u'http://vimeo.com/56015672',
  23         u'file': u'56015672.mp4',
  24         u'md5': u'8879b6cc097e987f02484baf890129e5',
  25         u'info_dict': {
  26             u"upload_date": u"20121220",
  27             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",
  28             u"uploader_id": u"user7108434",
  29             u"uploader": u"Filippo Valsorda",
  30             u"title": u"youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550"
  31         }
  32     }
  33
  34     def _verify_video_password(self, url, video_id, webpage):
  35         password = self._downloader.params.get('videopassword', None)
  36         if password is None:
  37             raise ExtractorError(u'This video is protected by a password, use the --video-password option')
  38         token = re.search(r'xsrft: \'(.*?)\'', webpage).group(1)
  39         data = compat_urllib_parse.urlencode({'password': password,
  40                                               'token': token})
  41         # I didn't manage to use the password with https
  42         if url.startswith('https'):
  43             pass_url = url.replace('https','http')
  44         else:
  45             pass_url = url
  46         password_request = compat_urllib_request.Request(pass_url+'/password', data)
  47         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  48         password_request.add_header('Cookie', 'xsrft=%s' % token)
  49         self._download_webpage(password_request, video_id,
  50                                u'Verifying the password',
  51                                u'Wrong password')
  52
  53     def _real_extract(self, url, new_video=True):
  54         # Extract ID from URL
  55         mobj = re.match(self._VALID_URL, url)
  56         if mobj is None:
  57             raise ExtractorError(u'Invalid URL: %s' % url)
  58
  59         video_id = mobj.group('id')
  60         if not mobj.group('proto'):
  61             url = 'https://' + url
  62         if mobj.group('direct_link') or mobj.group('pro'):
  63             url = 'https://vimeo.com/' + video_id
  64
  65         # Retrieve video webpage to extract further information
  66         request = compat_urllib_request.Request(url, None, std_headers)
  67         webpage = self._download_webpage(request, video_id)
  68
  69         # Now we begin extracting as much information as we can from what we
  70         # retrieved. First we extract the information common to all extractors,
  71         # and latter we extract those that are Vimeo specific.
  72         self.report_extraction(video_id)
  73
  74         # Extract the config JSON
  75         try:
  76             config = webpage.split(' = {config:')[1].split(',assets:')[0]
  77             config = json.loads(config)
  78         except:
  79             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
  80                 raise ExtractorError(u'The author has restricted the access to this video, try with the "--referer" option')
  81
  82             if re.search('If so please provide the correct password.', webpage):
  83                 self._verify_video_password(url, video_id, webpage)
  84                 return self._real_extract(url)
  85             else:
  86                 raise ExtractorError(u'Unable to extract info section')
  87
  88         # Extract title
  89         video_title = config["video"]["title"]
  90
  91         # Extract uploader and uploader_id
  92         video_uploader = config["video"]["owner"]["name"]
  93         video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] if config["video"]["owner"]["url"] else None
  94
  95         # Extract video thumbnail
  96         video_thumbnail = config["video"]["thumbnail"]
  97
  98         # Extract video description
  99         video_description = get_element_by_attribute("itemprop", "description", webpage)
 100         if video_description: video_description = clean_html(video_description)
 101         else: video_description = u''
 102
 103         # Extract upload date
 104         video_upload_date = None
 105         mobj = re.search(r'<meta itemprop="dateCreated" content="(\d{4})-(\d{2})-(\d{2})T', webpage)
 106         if mobj is not None:
 107             video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3)
 108
 109         # Vimeo specific: extract request signature and timestamp
 110         sig = config['request']['signature']
 111         timestamp = config['request']['timestamp']
 112
 113         # Vimeo specific: extract video codec and quality information
 114         # First consider quality, then codecs, then take everything
 115         # TODO bind to format param
 116         codecs = [('h264', 'mp4'), ('vp8', 'flv'), ('vp6', 'flv')]
 117         files = { 'hd': [], 'sd': [], 'other': []}
 118         for codec_name, codec_extension in codecs:
 119             if codec_name in config["video"]["files"]:
 120                 if 'hd' in config["video"]["files"][codec_name]:
 121                     files['hd'].append((codec_name, codec_extension, 'hd'))
 122                 elif 'sd' in config["video"]["files"][codec_name]:
 123                     files['sd'].append((codec_name, codec_extension, 'sd'))
 124                 else:
 125                     files['other'].append((codec_name, codec_extension, config["video"]["files"][codec_name][0]))
 126
 127         for quality in ('hd', 'sd', 'other'):
 128             if len(files[quality]) > 0:
 129                 video_quality = files[quality][0][2]
 130                 video_codec = files[quality][0][0]
 131                 video_extension = files[quality][0][1]
 132                 self.to_screen(u'%s: Downloading %s file at %s quality' % (video_id, video_codec.upper(), video_quality))
 133                 break
 134         else:
 135             raise ExtractorError(u'No known codec found')
 136
 137         video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
 138                     %(video_id, sig, timestamp, video_quality, video_codec.upper())
 139
 140         return [{
 141             'id':       video_id,
 142             'url':      video_url,
 143             'uploader': video_uploader,
 144             'uploader_id': video_uploader_id,
 145             'upload_date':  video_upload_date,
 146             'title':    video_title,
 147             'ext':      video_extension,
 148             'thumbnail':    video_thumbnail,
 149             'description':  video_description,
 150         }]