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 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
  22 class DailymotionBaseInfoExtractor(InfoExtractor):
  23     @staticmethod
  24     def _build_request(url):
  25         """Build a request with the family filter disabled"""
  26         request = compat_urllib_request.Request(url)
  27         request.add_header('Cookie', 'family_filter=off')
  28         request.add_header('Cookie', 'ff=off')
  29         return request
  30
  31
  32 class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor):
  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': 'http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech',
  49             'md5': '392c4b85a60a90dc4792da41ce3144eb',
  50             'info_dict': {
  51                 'id': 'x33vw9',
  52                 'ext': 'mp4',
  53                 'uploader': 'Amphora Alex and Van .',
  54                 'title': 'Tutoriel de Youtubeur"DL DES VIDEO DE YOUTUBE"',
  55             }
  56         },
  57         # Vevo video
  58         {
  59             'url': 'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
  60             'info_dict': {
  61                 'title': 'Roar (Official)',
  62                 'id': 'USUV71301934',
  63                 'ext': 'mp4',
  64                 'uploader': 'Katy Perry',
  65                 'upload_date': '20130905',
  66             },
  67             'params': {
  68                 'skip_download': True,
  69             },
  70             'skip': 'VEVO is only available in some countries',
  71         },
  72         # age-restricted video
  73         {
  74             'url': 'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband',
  75             'md5': '0d667a7b9cebecc3c89ee93099c4159d',
  76             'info_dict': {
  77                 'id': 'xyh2zz',
  78                 'ext': 'mp4',
  79                 'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]',
  80                 'uploader': 'HotWaves1012',
  81                 'age_limit': 18,
  82             }
  83         }
  84     ]
  85
  86     def _real_extract(self, url):
  87         video_id = self._match_id(url)
  88         url = 'http://www.dailymotion.com/video/%s' % video_id
  89
  90         # Retrieve video webpage to extract further information
  91         request = self._build_request(url)
  92         webpage = self._download_webpage(request, video_id)
  93
  94         # Extract URL, uploader and title from webpage
  95         self.report_extraction(video_id)
  96
  97         # It may just embed a vevo video:
  98         m_vevo = re.search(
  99             r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?video=(?P<id>[\w]*)',
 100             webpage)
 101         if m_vevo is not None:
 102             vevo_id = m_vevo.group('id')
 103             self.to_screen('Vevo video detected: %s' % vevo_id)
 104             return self.url_result('vevo:%s' % vevo_id, ie='Vevo')
 105
 106         age_limit = self._rta_search(webpage)
 107
 108         video_upload_date = None
 109         mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
 110         if mobj is not None:
 111             video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1)
 112
 113         embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id
 114         embed_page = self._download_webpage(embed_url, video_id,
 115                                             'Downloading embed page')
 116         info = self._search_regex(r'var info = ({.*?}),$', embed_page,
 117                                   'video info', flags=re.MULTILINE)
 118         info = json.loads(info)
 119         if info.get('error') is not None:
 120             msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title']
 121             raise ExtractorError(msg, expected=True)
 122
 123         formats = []
 124         for (key, format_id) in self._FORMATS:
 125             video_url = info.get(key)
 126             if video_url is not None:
 127                 m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
 128                 if m_size is not None:
 129                     width, height = map(int_or_none, (m_size.group(1), m_size.group(2)))
 130                 else:
 131                     width, height = None, None
 132                 formats.append({
 133                     'url': video_url,
 134                     'ext': 'mp4',
 135                     'format_id': format_id,
 136                     'width': width,
 137                     'height': height,
 138                 })
 139         if not formats:
 140             raise ExtractorError('Unable to extract video URL')
 141
 142         # subtitles
 143         video_subtitles = self.extract_subtitles(video_id, webpage)
 144         if self._downloader.params.get('listsubtitles', False):
 145             self._list_available_subtitles(video_id, webpage)
 146             return
 147
 148         view_count = str_to_int(self._search_regex(
 149             r'video_views_count[^>]+>\s+([\d\.,]+)',
 150             webpage, 'view count', fatal=False))
 151
 152         title = self._og_search_title(webpage, default=None)
 153         if title is None:
 154             title = self._html_search_regex(
 155                 r'(?s)<span\s+id="video_title"[^>]*>(.*?)</span>', webpage,
 156                 'title')
 157
 158         return {
 159             'id': video_id,
 160             'formats': formats,
 161             'uploader': info['owner.screenname'],
 162             'upload_date': video_upload_date,
 163             'title': title,
 164             'subtitles': video_subtitles,
 165             'thumbnail': info['thumbnail_url'],
 166             'age_limit': age_limit,
 167             'view_count': view_count,
 168         }
 169
 170     def _get_available_subtitles(self, video_id, webpage):
 171         try:
 172             sub_list = self._download_webpage(
 173                 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
 174                 video_id, note=False)
 175         except ExtractorError as err:
 176             self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
 177             return {}
 178         info = json.loads(sub_list)
 179         if (info['total'] > 0):
 180             sub_lang_list = dict((l['language'], l['url']) for l in info['list'])
 181             return sub_lang_list
 182         self._downloader.report_warning('video doesn\'t have subtitles')
 183         return {}
 184
 185
 186 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 187     IE_NAME = 'dailymotion:playlist'
 188     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
 189     _MORE_PAGES_INDICATOR = r'(?s)<div class="pages[^"]*">.*?<a\s+class="[^"]*?icon-arrow_right[^"]*?"'
 190     _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
 191     _TESTS = [{
 192         'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q',
 193         'info_dict': {
 194             'title': '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}/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(url, user)
 243         full_user = unescapeHTML(self._html_search_regex(
 244             r'<a class="nav-image" title="([^"]+)" href="/%s">' % re.escape(user),
 245             webpage, 'user'))
 246
 247         return {
 248             '_type': 'playlist',
 249             'id': user,
 250             'title': full_user,
 251             'entries': self._extract_entries(user),
 252         }