Raphaël G. Git Repositories - youtubedl/blob - 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
  10 from ..compat import (
  11     compat_str,
  12     compat_urllib_request,
  13 )
  14 from ..utils import (
  15     ExtractorError,
  16     int_or_none,
  17     orderedSet,
  18     str_to_int,
  19     unescapeHTML,
  20 )
  21
  22
  23 class DailymotionBaseInfoExtractor(InfoExtractor):
  24     @staticmethod
  25     def _build_request(url):
  26         """Build a request with the family filter disabled"""
  27         request = compat_urllib_request.Request(url)
  28         request.add_header('Cookie', 'family_filter=off; ff=off')
  29         return request
  30
  31
  32 class DailymotionIE(DailymotionBaseInfoExtractor):
  33     """Information Extractor for Dailymotion"""
  34
  35     _VALID_URL = r'(?i)(?:https?://)?(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(embed|#)/)?video/(?P<id>[^/?_]+)'
  36     IE_NAME = 'dailymotion'
  37
  38     _FORMATS = [
  39         ('stream_h264_ld_url', 'ld'),
  40         ('stream_h264_url', 'standard'),
  41         ('stream_h264_hq_url', 'hq'),
  42         ('stream_h264_hd_url', 'hd'),
  43         ('stream_h264_hd1080_url', 'hd180'),
  44     ]
  45
  46     _TESTS = [
  47         {
  48             'url': 'https://www.dailymotion.com/video/x2iuewm_steam-machine-models-pricing-listed-on-steam-store-ign-news_videogames',
  49             'md5': '2137c41a8e78554bb09225b8eb322406',
  50             'info_dict': {
  51                 'id': 'x2iuewm',
  52                 'ext': 'mp4',
  53                 'uploader': 'IGN',
  54                 'title': 'Steam Machine Models, Pricing Listed on Steam Store - IGN News',
  55                 'upload_date': '20150306',
  56             }
  57         },
  58         # Vevo video
  59         {
  60             'url': 'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
  61             'info_dict': {
  62                 'title': 'Roar (Official)',
  63                 'id': 'USUV71301934',
  64                 'ext': 'mp4',
  65                 'uploader': 'Katy Perry',
  66                 'upload_date': '20130905',
  67             },
  68             'params': {
  69                 'skip_download': True,
  70             },
  71             'skip': 'VEVO is only available in some countries',
  72         },
  73         # age-restricted video
  74         {
  75             'url': 'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband',
  76             'md5': '0d667a7b9cebecc3c89ee93099c4159d',
  77             'info_dict': {
  78                 'id': 'xyh2zz',
  79                 'ext': 'mp4',
  80                 'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]',
  81                 'uploader': 'HotWaves1012',
  82                 'age_limit': 18,
  83             }
  84         }
  85     ]
  86
  87     def _real_extract(self, url):
  88         video_id = self._match_id(url)
  89         url = 'https://www.dailymotion.com/video/%s' % video_id
  90
  91         # Retrieve video webpage to extract further information
  92         request = self._build_request(url)
  93         webpage = self._download_webpage(request, video_id)
  94
  95         # Extract URL, uploader and title from webpage
  96         self.report_extraction(video_id)
  97
  98         # It may just embed a vevo video:
  99         m_vevo = re.search(
 100             r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?video=(?P<id>[\w]*)',
 101             webpage)
 102         if m_vevo is not None:
 103             vevo_id = m_vevo.group('id')
 104             self.to_screen('Vevo video detected: %s' % vevo_id)
 105             return self.url_result('vevo:%s' % vevo_id, ie='Vevo')
 106
 107         age_limit = self._rta_search(webpage)
 108
 109         video_upload_date = None
 110         mobj = re.search(r'<meta property="video:release_date" content="([0-9]{4})-([0-9]{2})-([0-9]{2}).+?"/>', webpage)
 111         if mobj is not None:
 112             video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3)
 113
 114         embed_url = 'https://www.dailymotion.com/embed/video/%s' % video_id
 115         embed_request = self._build_request(embed_url)
 116         embed_page = self._download_webpage(
 117             embed_request, video_id, '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
 147         view_count = str_to_int(self._search_regex(
 148             r'video_views_count[^>]+>\s+([\d\.,]+)',
 149             webpage, 'view count', fatal=False))
 150
 151         title = self._og_search_title(webpage, default=None)
 152         if title is None:
 153             title = self._html_search_regex(
 154                 r'(?s)<span\s+id="video_title"[^>]*>(.*?)</span>', webpage,
 155                 'title')
 156
 157         return {
 158             'id': video_id,
 159             'formats': formats,
 160             'uploader': info['owner.screenname'],
 161             'upload_date': video_upload_date,
 162             'title': title,
 163             'subtitles': video_subtitles,
 164             'thumbnail': info['thumbnail_url'],
 165             'age_limit': age_limit,
 166             'view_count': view_count,
 167         }
 168
 169     def _get_subtitles(self, video_id, webpage):
 170         try:
 171             sub_list = self._download_webpage(
 172                 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
 173                 video_id, note=False)
 174         except ExtractorError as err:
 175             self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
 176             return {}
 177         info = json.loads(sub_list)
 178         if (info['total'] > 0):
 179             sub_lang_list = dict((l['language'], [{'url': l['url'], 'ext': 'srt'}]) for l in info['list'])
 180             return sub_lang_list
 181         self._downloader.report_warning('video doesn\'t have subtitles')
 182         return {}
 183
 184
 185 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 186     IE_NAME = 'dailymotion:playlist'
 187     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
 188     _MORE_PAGES_INDICATOR = r'(?s)<div class="pages[^"]*">.*?<a\s+class="[^"]*?icon-arrow_right[^"]*?"'
 189     _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
 190     _TESTS = [{
 191         'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q',
 192         'info_dict': {
 193             'title': 'SPORT',
 194             'id': 'xv4bw_nqtv_sport',
 195         },
 196         'playlist_mincount': 20,
 197     }]
 198
 199     def _extract_entries(self, id):
 200         video_ids = []
 201         for pagenum in itertools.count(1):
 202             request = self._build_request(self._PAGE_TEMPLATE % (id, pagenum))
 203             webpage = self._download_webpage(request,
 204                                              id, 'Downloading page %s' % pagenum)
 205
 206             video_ids.extend(re.findall(r'data-xid="(.+?)"', webpage))
 207
 208             if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:
 209                 break
 210         return [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
 211                 for video_id in orderedSet(video_ids)]
 212
 213     def _real_extract(self, url):
 214         mobj = re.match(self._VALID_URL, url)
 215         playlist_id = mobj.group('id')
 216         webpage = self._download_webpage(url, playlist_id)
 217
 218         return {
 219             '_type': 'playlist',
 220             'id': playlist_id,
 221             'title': self._og_search_title(webpage),
 222             'entries': self._extract_entries(playlist_id),
 223         }
 224
 225
 226 class DailymotionUserIE(DailymotionPlaylistIE):
 227     IE_NAME = 'dailymotion:user'
 228     _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?:(?:old/)?user/)?(?P<user>[^/]+)$'
 229     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
 230     _TESTS = [{
 231         'url': 'https://www.dailymotion.com/user/nqtv',
 232         'info_dict': {
 233             'id': 'nqtv',
 234             'title': 'Rémi Gaillard',
 235         },
 236         'playlist_mincount': 100,
 237     }]
 238
 239     def _real_extract(self, url):
 240         mobj = re.match(self._VALID_URL, url)
 241         user = mobj.group('user')
 242         webpage = self._download_webpage(
 243             'https://www.dailymotion.com/user/%s' % user, user)
 244         full_user = unescapeHTML(self._html_search_regex(
 245             r'<a class="nav-image" title="([^"]+)" href="/%s">' % re.escape(user),
 246             webpage, 'user'))
 247
 248         return {
 249             '_type': 'playlist',
 250             'id': user,
 251             'title': full_user,
 252             'entries': self._extract_entries(user),
 253         }