_ Git - youtube-dl/blob - 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     get_element_by_attribute,
  12     get_element_by_id,
  13     orderedSet,
  14     str_to_int,
  15     int_or_none,
  16
  17     ExtractorError,
  18 )
  19
  20 class DailymotionBaseInfoExtractor(InfoExtractor):
  21     @staticmethod
  22     def _build_request(url):
  23         """Build a request with the family filter disabled"""
  24         request = compat_urllib_request.Request(url)
  25         request.add_header('Cookie', 'family_filter=off')
  26         request.add_header('Cookie', 'ff=off')
  27         return request
  28
  29 class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor):
  30     """Information Extractor for Dailymotion"""
  31
  32     _VALID_URL = r'(?i)(?:https?://)?(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(embed|#)/)?video/(?P<id>[^/?_]+)'
  33     IE_NAME = u'dailymotion'
  34
  35     _FORMATS = [
  36         (u'stream_h264_ld_url', u'ld'),
  37         (u'stream_h264_url', u'standard'),
  38         (u'stream_h264_hq_url', u'hq'),
  39         (u'stream_h264_hd_url', u'hd'),
  40         (u'stream_h264_hd1080_url', u'hd180'),
  41     ]
  42
  43     _TESTS = [
  44         {
  45             u'url': u'http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech',
  46             u'file': u'x33vw9.mp4',
  47             u'md5': u'392c4b85a60a90dc4792da41ce3144eb',
  48             u'info_dict': {
  49                 u"uploader": u"Amphora Alex and Van .",
  50                 u"title": u"Tutoriel de Youtubeur\"DL DES VIDEO DE YOUTUBE\""
  51             }
  52         },
  53         # Vevo video
  54         {
  55             u'url': u'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
  56             u'file': u'USUV71301934.mp4',
  57             u'info_dict': {
  58                 u'title': u'Roar (Official)',
  59                 u'uploader': u'Katy Perry',
  60                 u'upload_date': u'20130905',
  61             },
  62             u'params': {
  63                 u'skip_download': True,
  64             },
  65             u'skip': u'VEVO is only available in some countries',
  66         },
  67         # age-restricted video
  68         {
  69             u'url': u'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband',
  70             u'file': u'xyh2zz.mp4',
  71             u'md5': u'0d667a7b9cebecc3c89ee93099c4159d',
  72             u'info_dict': {
  73                 u'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]',
  74                 u'uploader': 'HotWaves1012',
  75                 u'age_limit': 18,
  76             }
  77
  78         }
  79     ]
  80
  81     def _real_extract(self, url):
  82         # Extract id and simplified title from URL
  83         mobj = re.match(self._VALID_URL, url)
  84
  85         video_id = mobj.group('id')
  86
  87         url = 'http://www.dailymotion.com/video/%s' % video_id
  88
  89         # Retrieve video webpage to extract further information
  90         request = self._build_request(url)
  91         webpage = self._download_webpage(request, video_id)
  92
  93         # Extract URL, uploader and title from webpage
  94         self.report_extraction(video_id)
  95
  96         # It may just embed a vevo video:
  97         m_vevo = re.search(
  98             r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?videoId=(?P<id>[\w]*)',
  99             webpage)
 100         if m_vevo is not None:
 101             vevo_id = m_vevo.group('id')
 102             self.to_screen(u'Vevo video detected: %s' % vevo_id)
 103             return self.url_result(u'vevo:%s' % vevo_id, ie='Vevo')
 104
 105         age_limit = self._rta_search(webpage)
 106
 107         video_upload_date = None
 108         mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
 109         if mobj is not None:
 110             video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1)
 111
 112         embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id
 113         embed_page = self._download_webpage(embed_url, video_id,
 114                                             u'Downloading embed page')
 115         info = self._search_regex(r'var info = ({.*?}),$', embed_page,
 116             'video info', flags=re.MULTILINE)
 117         info = json.loads(info)
 118         if info.get('error') is not None:
 119             msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title']
 120             raise ExtractorError(msg, expected=True)
 121
 122         formats = []
 123         for (key, format_id) in self._FORMATS:
 124             video_url = info.get(key)
 125             if video_url is not None:
 126                 m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
 127                 if m_size is not None:
 128                     width, height = map(int_or_none, (m_size.group(1), m_size.group(2)))
 129                 else:
 130                     width, height = None, None
 131                 formats.append({
 132                     'url': video_url,
 133                     'ext': 'mp4',
 134                     'format_id': format_id,
 135                     'width': width,
 136                     'height': height,
 137                 })
 138         if not formats:
 139             raise ExtractorError(u'Unable to extract video URL')
 140
 141         # subtitles
 142         video_subtitles = self.extract_subtitles(video_id, webpage)
 143         if self._downloader.params.get('listsubtitles', False):
 144             self._list_available_subtitles(video_id, webpage)
 145             return
 146
 147         view_count = self._search_regex(
 148             r'video_views_count[^>]+>\s+([\d\.,]+)', webpage, u'view count', fatal=False)
 149         if view_count is not None:
 150             view_count = str_to_int(view_count)
 151
 152         return {
 153             'id':       video_id,
 154             'formats': formats,
 155             'uploader': info['owner_screenname'],
 156             'upload_date':  video_upload_date,
 157             'title':    self._og_search_title(webpage),
 158             'subtitles':    video_subtitles,
 159             'thumbnail': info['thumbnail_url'],
 160             'age_limit': age_limit,
 161             'view_count': view_count,
 162         }
 163
 164     def _get_available_subtitles(self, video_id, webpage):
 165         try:
 166             sub_list = self._download_webpage(
 167                 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
 168                 video_id, note=False)
 169         except ExtractorError as err:
 170             self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err))
 171             return {}
 172         info = json.loads(sub_list)
 173         if (info['total'] > 0):
 174             sub_lang_list = dict((l['language'], l['url']) for l in info['list'])
 175             return sub_lang_list
 176         self._downloader.report_warning(u'video doesn\'t have subtitles')
 177         return {}
 178
 179
 180 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 181     IE_NAME = u'dailymotion:playlist'
 182     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
 183     _MORE_PAGES_INDICATOR = r'<div class="next">.*?<a.*?href="/playlist/.+?".*?>.*?</a>.*?</div>'
 184     _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
 185
 186     def _extract_entries(self, id):
 187         video_ids = []
 188         for pagenum in itertools.count(1):
 189             request = self._build_request(self._PAGE_TEMPLATE % (id, pagenum))
 190             webpage = self._download_webpage(request,
 191                                              id, u'Downloading page %s' % pagenum)
 192
 193             playlist_el = get_element_by_attribute(u'class', u'row video_list', webpage)
 194             video_ids.extend(re.findall(r'data-id="(.+?)"', playlist_el))
 195
 196             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 197                 break
 198         return [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
 199                    for video_id in orderedSet(video_ids)]
 200
 201     def _real_extract(self, url):
 202         mobj = re.match(self._VALID_URL, url)
 203         playlist_id = mobj.group('id')
 204         webpage = self._download_webpage(url, playlist_id)
 205
 206         return {'_type': 'playlist',
 207                 'id': playlist_id,
 208                 'title': get_element_by_id(u'playlist_name', webpage),
 209                 'entries': self._extract_entries(playlist_id),
 210                 }
 211
 212
 213 class DailymotionUserIE(DailymotionPlaylistIE):
 214     IE_NAME = u'dailymotion:user'
 215     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/user/(?P<user>[^/]+)'
 216     _MORE_PAGES_INDICATOR = r'<div class="next">.*?<a.*?href="/user/.+?".*?>.*?</a>.*?</div>'
 217     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
 218
 219     def _real_extract(self, url):
 220         mobj = re.match(self._VALID_URL, url)
 221         user = mobj.group('user')
 222         webpage = self._download_webpage(url, user)
 223         full_user = self._html_search_regex(
 224             r'<a class="label" href="/%s".*?>(.*?)</' % re.escape(user),
 225             webpage, u'user', flags=re.DOTALL)
 226
 227         return {
 228             '_type': 'playlist',
 229             'id': user,
 230             'title': full_user,
 231             'entries': self._extract_entries(user),
 232         }