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

   1 # encoding: utf-8
   2
   3 import re
   4 import json
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     compat_urllib_request,
   9     ExtractorError,
  10 )
  11
  12
  13 class IviIE(InfoExtractor):
  14     IE_DESC = u'ivi.ru'
  15     IE_NAME = u'ivi'
  16     _VALID_URL = r'^https?://(?:www\.)?ivi\.ru/watch(?:/(?P<compilationid>[^/]+))?/(?P<videoid>\d+)'
  17
  18     _TESTS = [
  19         # Single movie
  20         {
  21             u'url': u'http://www.ivi.ru/watch/53141',
  22             u'file': u'53141.mp4',
  23             u'md5': u'6ff5be2254e796ed346251d117196cf4',
  24             u'info_dict': {
  25                 u'title': u'Иван Васильевич меняет профессию',
  26                 u'description': u'md5:14d8eda24e9d93d29b5857012c6d6346',
  27                 u'duration': 5498,
  28                 u'thumbnail': u'http://thumbs.ivi.ru/f20.vcp.digitalaccess.ru/contents/d/1/c3c885163a082c29bceeb7b5a267a6.jpg',
  29             },
  30             u'skip': u'Only works from Russia',
  31         },
  32         # Serial's serie
  33         {
  34             u'url': u'http://www.ivi.ru/watch/dezhurnyi_angel/74791',
  35             u'file': u'74791.mp4',
  36             u'md5': u'3e6cc9a848c1d2ebcc6476444967baa9',
  37             u'info_dict': {
  38                 u'title': u'Дежурный ангел - 1 серия',
  39                 u'duration': 2490,
  40                 u'thumbnail': u'http://thumbs.ivi.ru/f7.vcp.digitalaccess.ru/contents/8/e/bc2f6c2b6e5d291152fdd32c059141.jpg',
  41             },
  42             u'skip': u'Only works from Russia',
  43          }
  44     ]
  45
  46     # Sorted by quality
  47     _known_formats = ['MP4-low-mobile', 'MP4-mobile', 'FLV-lo', 'MP4-lo', 'FLV-hi', 'MP4-hi', 'MP4-SHQ']
  48
  49     # Sorted by size
  50     _known_thumbnails = ['Thumb-120x90', 'Thumb-160', 'Thumb-640x480']
  51
  52     def _extract_description(self, html):
  53         m = re.search(r'<meta name="description" content="(?P<description>[^"]+)"/>', html)
  54         return m.group('description') if m is not None else None
  55
  56     def _extract_comment_count(self, html):
  57         m = re.search(u'(?s)<a href="#" id="view-comments" class="action-button dim gradient">\s*Комментарии:\s*(?P<commentcount>\d+)\s*</a>', html)
  58         return int(m.group('commentcount')) if m is not None else 0
  59
  60     def _real_extract(self, url):
  61         mobj = re.match(self._VALID_URL, url)
  62         video_id = mobj.group('videoid')
  63
  64         api_url = 'http://api.digitalaccess.ru/api/json/'
  65
  66         data = {u'method': u'da.content.get',
  67                 u'params': [video_id, {u'site': u's183',
  68                                        u'referrer': u'http://www.ivi.ru/watch/%s' % video_id,
  69                                        u'contentid': video_id
  70                                     }
  71                             ]
  72                 }
  73
  74         request = compat_urllib_request.Request(api_url, json.dumps(data))
  75
  76         video_json_page = self._download_webpage(request, video_id, u'Downloading video JSON')
  77         video_json = json.loads(video_json_page)
  78
  79         if u'error' in video_json:
  80             error = video_json[u'error']
  81             if error[u'origin'] == u'NoRedisValidData':
  82                 raise ExtractorError(u'Video %s does not exist' % video_id, expected=True)
  83             raise ExtractorError(u'Unable to download video %s: %s' % (video_id, error[u'message']), expected=True)
  84
  85         result = video_json[u'result']
  86
  87         formats = [{
  88             'url': x[u'url'],
  89             'format_id': x[u'content_format'],
  90             'preference': self._known_formats.index(x[u'content_format']),
  91         } for x in result[u'files'] if x[u'content_format'] in self._known_formats]
  92
  93         self._sort_formats(formats)
  94
  95         if not formats:
  96             raise ExtractorError(u'No media links available for %s' % video_id)
  97
  98         duration = result[u'duration']
  99         compilation = result[u'compilation']
 100         title = result[u'title']
 101
 102         title = '%s - %s' % (compilation, title) if compilation is not None else title
 103
 104         previews = result[u'preview']
 105         previews.sort(key=lambda fmt: self._known_thumbnails.index(fmt['content_format']))
 106         thumbnail = previews[-1][u'url'] if len(previews) > 0 else None
 107
 108         video_page = self._download_webpage(url, video_id, u'Downloading video page')
 109         description = self._extract_description(video_page)
 110         comment_count = self._extract_comment_count(video_page)
 111
 112         return {
 113             'id': video_id,
 114             'title': title,
 115             'thumbnail': thumbnail,
 116             'description': description,
 117             'duration': duration,
 118             'comment_count': comment_count,
 119             'formats': formats,
 120         }
 121
 122
 123 class IviCompilationIE(InfoExtractor):
 124     IE_DESC = u'ivi.ru compilations'
 125     IE_NAME = u'ivi:compilation'
 126     _VALID_URL = r'^https?://(?:www\.)?ivi\.ru/watch/(?!\d+)(?P<compilationid>[a-z\d_-]+)(?:/season(?P<seasonid>\d+))?$'
 127
 128     def _extract_entries(self, html, compilation_id):
 129         return [self.url_result('http://www.ivi.ru/watch/%s/%s' % (compilation_id, serie), 'Ivi')
 130                 for serie in re.findall(r'<strong><a href="/watch/%s/(\d+)">(?:[^<]+)</a></strong>' % compilation_id, html)]
 131
 132     def _real_extract(self, url):
 133         mobj = re.match(self._VALID_URL, url)
 134         compilation_id = mobj.group('compilationid')
 135         season_id = mobj.group('seasonid')
 136
 137         if season_id is not None: # Season link
 138             season_page = self._download_webpage(url, compilation_id, u'Downloading season %s web page' % season_id)
 139             playlist_id = '%s/season%s' % (compilation_id, season_id)
 140             playlist_title = self._html_search_meta(u'title', season_page, u'title')
 141             entries = self._extract_entries(season_page, compilation_id)
 142         else: # Compilation link
 143             compilation_page = self._download_webpage(url, compilation_id, u'Downloading compilation web page')
 144             playlist_id = compilation_id
 145             playlist_title = self._html_search_meta(u'title', compilation_page, u'title')
 146             seasons = re.findall(r'<a href="/watch/%s/season(\d+)">[^<]+</a>' % compilation_id, compilation_page)
 147             if len(seasons) == 0: # No seasons in this compilation
 148                 entries = self._extract_entries(compilation_page, compilation_id)
 149             else:
 150                 entries = []
 151                 for season_id in seasons:
 152                     season_page = self._download_webpage('http://www.ivi.ru/watch/%s/season%s' % (compilation_id, season_id),
 153                                                          compilation_id, u'Downloading season %s web page' % season_id)
 154                     entries.extend(self._extract_entries(season_page, compilation_id))
 155
 156         return self.playlist_result(entries, playlist_id, playlist_title)