youtube_dl/extractor/dailymotion.py

   1 import re
   2 import json
   3 import itertools
   4
   5 from .common import InfoExtractor
   6 from .subtitles import SubtitlesInfoExtractor
   7
   8 from ..utils import (
   9     compat_urllib_request,
  10     compat_str,
  11     orderedSet,
  12     str_to_int,
  13     int_or_none,
  14     ExtractorError,
  15     unescapeHTML,
  16 )
  17
  18 class DailymotionBaseInfoExtractor(InfoExtractor):
  19     @staticmethod
  20     def _build_request(url):
  21         """Build a request with the family filter disabled"""
  22         request = compat_urllib_request.Request(url)
  23         request.add_header('Cookie', 'family_filter=off')
  24         request.add_header('Cookie', 'ff=off')
  25         return request
  26
  27 class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor):
  28     """Information Extractor for Dailymotion"""
  29
  30     _VALID_URL = r'(?i)(?:https?://)?(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(embed|#)/)?video/(?P<id>[^/?_]+)'
  31     IE_NAME = u'dailymotion'
  32
  33     _FORMATS = [
  34         (u'stream_h264_ld_url', u'ld'),
  35         (u'stream_h264_url', u'standard'),
  36         (u'stream_h264_hq_url', u'hq'),
  37         (u'stream_h264_hd_url', u'hd'),
  38         (u'stream_h264_hd1080_url', u'hd180'),
  39     ]
  40
  41     _TESTS = [
  42         {
  43             u'url': u'http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech',
  44             u'file': u'x33vw9.mp4',
  45             u'md5': u'392c4b85a60a90dc4792da41ce3144eb',
  46             u'info_dict': {
  47                 u"uploader": u"Amphora Alex and Van .",
  48                 u"title": u"Tutoriel de Youtubeur\"DL DES VIDEO DE YOUTUBE\""
  49             }
  50         },
  51         # Vevo video
  52         {
  53             u'url': u'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
  54             u'file': u'USUV71301934.mp4',
  55             u'info_dict': {
  56                 u'title': u'Roar (Official)',
  57                 u'uploader': u'Katy Perry',
  58                 u'upload_date': u'20130905',
  59             },
  60             u'params': {
  61                 u'skip_download': True,
  62             },
  63             u'skip': u'VEVO is only available in some countries',
  64         },
  65         # age-restricted video
  66         {
  67             u'url': u'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband',
  68             u'file': u'xyh2zz.mp4',
  69             u'md5': u'0d667a7b9cebecc3c89ee93099c4159d',
  70             u'info_dict': {
  71                 u'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]',
  72                 u'uploader': 'HotWaves1012',
  73                 u'age_limit': 18,
  74             }
  75
  76         }
  77     ]
  78
  79     def _real_extract(self, url):
  80         # Extract id and simplified title from URL
  81         mobj = re.match(self._VALID_URL, url)
  82
  83         video_id = mobj.group('id')
  84
  85         url = 'http://www.dailymotion.com/video/%s' % video_id
  86
  87         # Retrieve video webpage to extract further information
  88         request = self._build_request(url)
  89         webpage = self._download_webpage(request, video_id)
  90
  91         # Extract URL, uploader and title from webpage
  92         self.report_extraction(video_id)
  93
  94         # It may just embed a vevo video:
  95         m_vevo = re.search(
  96             r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?videoId=(?P<id>[\w]*)',
  97             webpage)
  98         if m_vevo is not None:
  99             vevo_id = m_vevo.group('id')
 100             self.to_screen(u'Vevo video detected: %s' % vevo_id)
 101             return self.url_result(u'vevo:%s' % vevo_id, ie='Vevo')
 102
 103         age_limit = self._rta_search(webpage)
 104
 105         video_upload_date = None
 106         mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
 107         if mobj is not None:
 108             video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1)
 109
 110         embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id
 111         embed_page = self._download_webpage(embed_url, video_id,
 112                                             u'Downloading embed page')
 113         info = self._search_regex(r'var info = ({.*?}),$', embed_page,
 114             'video info', flags=re.MULTILINE)
 115         info = json.loads(info)
 116         if info.get('error') is not None:
 117             msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title']
 118             raise ExtractorError(msg, expected=True)
 119
 120         formats = []
 121         for (key, format_id) in self._FORMATS:
 122             video_url = info.get(key)
 123             if video_url is not None:
 124                 m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
 125                 if m_size is not None:
 126                     width, height = map(int_or_none, (m_size.group(1), m_size.group(2)))
 127                 else:
 128                     width, height = None, None
 129                 formats.append({
 130                     'url': video_url,
 131                     'ext': 'mp4',
 132                     'format_id': format_id,
 133                     'width': width,
 134                     'height': height,
 135                 })
 136         if not formats:
 137             raise ExtractorError(u'Unable to extract video URL')
 138
 139         # subtitles
 140         video_subtitles = self.extract_subtitles(video_id, webpage)
 141         if self._downloader.params.get('listsubtitles', False):
 142             self._list_available_subtitles(video_id, webpage)
 143             return
 144
 145         view_count = self._search_regex(
 146             r'video_views_count[^>]+>\s+([\d\.,]+)', webpage, u'view count', fatal=False)
 147         if view_count is not None:
 148             view_count = str_to_int(view_count)
 149
 150         return {
 151             'id':       video_id,
 152             'formats': formats,
 153             'uploader': info['owner.screenname'],
 154             'upload_date':  video_upload_date,
 155             'title':    self._og_search_title(webpage),
 156             'subtitles':    video_subtitles,
 157             'thumbnail': info['thumbnail_url'],
 158             'age_limit': age_limit,
 159             'view_count': view_count,
 160         }
 161
 162     def _get_available_subtitles(self, video_id, webpage):
 163         try:
 164             sub_list = self._download_webpage(
 165                 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
 166                 video_id, note=False)
 167         except ExtractorError as err:
 168             self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err))
 169             return {}
 170         info = json.loads(sub_list)
 171         if (info['total'] > 0):
 172             sub_lang_list = dict((l['language'], l['url']) for l in info['list'])
 173             return sub_lang_list
 174         self._downloader.report_warning(u'video doesn\'t have subtitles')
 175         return {}
 176
 177
 178 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 179     IE_NAME = u'dailymotion:playlist'
 180     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
 181     _MORE_PAGES_INDICATOR = r'(?s)<div class="pages[^"]*">.*?<a\s+class="[^"]*?icon-arrow_right[^"]*?"'
 182     _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
 183
 184     def _extract_entries(self, id):
 185         video_ids = []
 186         for pagenum in itertools.count(1):
 187             request = self._build_request(self._PAGE_TEMPLATE % (id, pagenum))
 188             webpage = self._download_webpage(request,
 189                                              id, u'Downloading page %s' % pagenum)
 190
 191             video_ids.extend(re.findall(r'data-xid="(.+?)"', webpage))
 192
 193             if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:
 194                 break
 195         return [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
 196                    for video_id in orderedSet(video_ids)]
 197
 198     def _real_extract(self, url):
 199         mobj = re.match(self._VALID_URL, url)
 200         playlist_id = mobj.group('id')
 201         webpage = self._download_webpage(url, playlist_id)
 202
 203         return {
 204             '_type': 'playlist',
 205             'id': playlist_id,
 206             'title': self._og_search_title(webpage),
 207             'entries': self._extract_entries(playlist_id),
 208         }
 209
 210
 211 class DailymotionUserIE(DailymotionPlaylistIE):
 212     IE_NAME = u'dailymotion:user'
 213     _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/user/(?P<user>[^/]+)'
 214     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
 215
 216     def _real_extract(self, url):
 217         mobj = re.match(self._VALID_URL, url)
 218         user = mobj.group('user')
 219         webpage = self._download_webpage(url, user)
 220         full_user = unescapeHTML(self._html_search_regex(
 221             r'<a class="nav-image" title="([^"]+)" href="/%s">' % re.escape(user),
 222             webpage, u'user', flags=re.DOTALL))
 223
 224         return {
 225             '_type': 'playlist',
 226             'id': user,
 227             'title': full_user,
 228             'entries': self._extract_entries(user),
 229         }