youtube_dl/extractor/dailymotion.py

   1 #coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import json
   6 import itertools
   7
   8 from .common import InfoExtractor
   9 from .subtitles import SubtitlesInfoExtractor
  10
  11 from ..utils import (
  12     compat_urllib_request,
  13     compat_str,
  14     orderedSet,
  15     str_to_int,
  16     int_or_none,
  17     ExtractorError,
  18     unescapeHTML,
  19 )
  20
  21 class DailymotionBaseInfoExtractor(InfoExtractor):
  22     @staticmethod
  23     def _build_request(url):
  24         """Build a request with the family filter disabled"""
  25         request = compat_urllib_request.Request(url)
  26         request.add_header('Cookie', 'family_filter=off')
  27         request.add_header('Cookie', 'ff=off')
  28         return request
  29
  30 class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor):
  31     """Information Extractor for Dailymotion"""
  32
  33     _VALID_URL = r'(?i)(?:https?://)?(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(embed|#)/)?video/(?P<id>[^/?_]+)'
  34     IE_NAME = 'dailymotion'
  35
  36     _FORMATS = [
  37         ('stream_h264_ld_url', 'ld'),
  38         ('stream_h264_url', 'standard'),
  39         ('stream_h264_hq_url', 'hq'),
  40         ('stream_h264_hd_url', 'hd'),
  41         ('stream_h264_hd1080_url', 'hd180'),
  42     ]
  43
  44     _TESTS = [
  45         {
  46             'url': 'http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech',
  47             'md5': '392c4b85a60a90dc4792da41ce3144eb',
  48             'info_dict': {
  49                 'id': 'x33vw9',
  50                 'ext': 'mp4',
  51                 'uploader': 'Amphora Alex and Van .',
  52                 'title': 'Tutoriel de Youtubeur"DL DES VIDEO DE YOUTUBE"',
  53             }
  54         },
  55         # Vevo video
  56         {
  57             'url': 'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
  58             'info_dict': {
  59                 'title': 'Roar (Official)',
  60                 'id': 'USUV71301934',
  61                 'ext': 'mp4',
  62                 'uploader': 'Katy Perry',
  63                 'upload_date': '20130905',
  64             },
  65             'params': {
  66                 'skip_download': True,
  67             },
  68             'skip': 'VEVO is only available in some countries',
  69         },
  70         # age-restricted video
  71         {
  72             'url': 'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband',
  73             'md5': '0d667a7b9cebecc3c89ee93099c4159d',
  74             'info_dict': {
  75                 'id': 'xyh2zz',
  76                 'ext': 'mp4',
  77                 'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]',
  78                 'uploader': 'HotWaves1012',
  79                 'age_limit': 18,
  80             }
  81         }
  82     ]
  83
  84     def _real_extract(self, url):
  85         # Extract id and simplified title from URL
  86         mobj = re.match(self._VALID_URL, url)
  87
  88         video_id = mobj.group('id')
  89
  90         url = 'http://www.dailymotion.com/video/%s' % video_id
  91
  92         # Retrieve video webpage to extract further information
  93         request = self._build_request(url)
  94         webpage = self._download_webpage(request, video_id)
  95
  96         # Extract URL, uploader and title from webpage
  97         self.report_extraction(video_id)
  98
  99         # It may just embed a vevo video:
 100         m_vevo = re.search(
 101             r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?videoId=(?P<id>[\w]*)',
 102             webpage)
 103         if m_vevo is not None:
 104             vevo_id = m_vevo.group('id')
 105             self.to_screen('Vevo video detected: %s' % vevo_id)
 106             return self.url_result('vevo:%s' % vevo_id, ie='Vevo')
 107
 108         age_limit = self._rta_search(webpage)
 109
 110         video_upload_date = None
 111         mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
 112         if mobj is not None:
 113             video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1)
 114
 115         embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id
 116         embed_page = self._download_webpage(embed_url, video_id,
 117                                             'Downloading embed page')
 118         info = self._search_regex(r'var info = ({.*?}),$', embed_page,
 119             'video info', flags=re.MULTILINE)
 120         info = json.loads(info)
 121         if info.get('error') is not None:
 122             msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title']
 123             raise ExtractorError(msg, expected=True)
 124
 125         formats = []
 126         for (key, format_id) in self._FORMATS:
 127             video_url = info.get(key)
 128             if video_url is not None:
 129                 m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
 130                 if m_size is not None:
 131                     width, height = map(int_or_none, (m_size.group(1), m_size.group(2)))
 132                 else:
 133                     width, height = None, None
 134                 formats.append({
 135                     'url': video_url,
 136                     'ext': 'mp4',
 137                     'format_id': format_id,
 138                     'width': width,
 139                     'height': height,
 140                 })
 141         if not formats:
 142             raise ExtractorError('Unable to extract video URL')
 143
 144         # subtitles
 145         video_subtitles = self.extract_subtitles(video_id, webpage)
 146         if self._downloader.params.get('listsubtitles', False):
 147             self._list_available_subtitles(video_id, webpage)
 148             return
 149
 150         view_count = self._search_regex(
 151             r'video_views_count[^>]+>\s+([\d\.,]+)', webpage, 'view count', fatal=False)
 152         if view_count is not None:
 153             view_count = str_to_int(view_count)
 154
 155         return {
 156             'id':       video_id,
 157             'formats': formats,
 158             'uploader': info['owner.screenname'],
 159             'upload_date':  video_upload_date,
 160             'title':    self._og_search_title(webpage),
 161             'subtitles':    video_subtitles,
 162             'thumbnail': info['thumbnail_url'],
 163             'age_limit': age_limit,
 164             'view_count': view_count,
 165         }
 166
 167     def _get_available_subtitles(self, video_id, webpage):
 168         try:
 169             sub_list = self._download_webpage(
 170                 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
 171                 video_id, note=False)
 172         except ExtractorError as err:
 173             self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
 174             return {}
 175         info = json.loads(sub_list)
 176         if (info['total'] > 0):
 177             sub_lang_list = dict((l['language'], l['url']) for l in info['list'])
 178             return sub_lang_list
 179         self._downloader.report_warning('video doesn\'t have subtitles')
 180         return {}
 181
 182
 183 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 184     IE_NAME = 'dailymotion:playlist'
 185     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
 186     _MORE_PAGES_INDICATOR = r'(?s)<div class="pages[^"]*">.*?<a\s+class="[^"]*?icon-arrow_right[^"]*?"'
 187     _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
 188     _TESTS = [{
 189         'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q',
 190         'info_dict': {
 191             'title': 'SPORT',
 192         },
 193         'playlist_mincount': 20,
 194     }]
 195
 196     def _extract_entries(self, id):
 197         video_ids = []
 198         for pagenum in itertools.count(1):
 199             request = self._build_request(self._PAGE_TEMPLATE % (id, pagenum))
 200             webpage = self._download_webpage(request,
 201                                              id, 'Downloading page %s' % pagenum)
 202
 203             video_ids.extend(re.findall(r'data-xid="(.+?)"', webpage))
 204
 205             if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:
 206                 break
 207         return [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
 208                    for video_id in orderedSet(video_ids)]
 209
 210     def _real_extract(self, url):
 211         mobj = re.match(self._VALID_URL, url)
 212         playlist_id = mobj.group('id')
 213         webpage = self._download_webpage(url, playlist_id)
 214
 215         return {
 216             '_type': 'playlist',
 217             'id': playlist_id,
 218             'title': self._og_search_title(webpage),
 219             'entries': self._extract_entries(playlist_id),
 220         }
 221
 222
 223 class DailymotionUserIE(DailymotionPlaylistIE):
 224     IE_NAME = 'dailymotion:user'
 225     _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/user/(?P<user>[^/]+)'
 226     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
 227     _TESTS = [{
 228         'url': 'https://www.dailymotion.com/user/nqtv',
 229         'info_dict': {
 230             'id': 'nqtv',
 231             'title': 'Rémi Gaillard',
 232         },
 233         'playlist_mincount': 100,
 234     }]
 235
 236     def _real_extract(self, url):
 237         mobj = re.match(self._VALID_URL, url)
 238         user = mobj.group('user')
 239         webpage = self._download_webpage(url, user)
 240         full_user = unescapeHTML(self._html_search_regex(
 241             r'<a class="nav-image" title="([^"]+)" href="/%s">' % re.escape(user),
 242             webpage, 'user'))
 243
 244         return {
 245             '_type': 'playlist',
 246             'id': user,
 247             'title': full_user,
 248             'entries': self._extract_entries(user),
 249         }