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')
  29         request.add_header('Cookie', 'ff=off')
  30         return request
  31
  32
  33 class DailymotionIE(DailymotionBaseInfoExtractor):
  34     """Information Extractor for Dailymotion"""
  35
  36     _VALID_URL = r'(?i)(?:https?://)?(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(embed|#)/)?video/(?P<id>[^/?_]+)'
  37     IE_NAME = 'dailymotion'
  38
  39     _FORMATS = [
  40         ('stream_h264_ld_url', 'ld'),
  41         ('stream_h264_url', 'standard'),
  42         ('stream_h264_hq_url', 'hq'),
  43         ('stream_h264_hd_url', 'hd'),
  44         ('stream_h264_hd1080_url', 'hd180'),
  45     ]
  46
  47     _TESTS = [
  48         {
  49             'url': 'http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech',
  50             'md5': '392c4b85a60a90dc4792da41ce3144eb',
  51             'info_dict': {
  52                 'id': 'x33vw9',
  53                 'ext': 'mp4',
  54                 'uploader': 'Amphora Alex and Van .',
  55                 'title': 'Tutoriel de Youtubeur"DL DES VIDEO DE YOUTUBE"',
  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 = 'http://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'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
 111         if mobj is not None:
 112             video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1)
 113
 114         embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id
 115         embed_page = self._download_webpage(embed_url, video_id,
 116                                             'Downloading embed page')
 117         info = self._search_regex(r'var info = ({.*?}),$', embed_page,
 118                                   'video info', flags=re.MULTILINE)
 119         info = json.loads(info)
 120         if info.get('error') is not None:
 121             msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title']
 122             raise ExtractorError(msg, expected=True)
 123
 124         formats = []
 125         for (key, format_id) in self._FORMATS:
 126             video_url = info.get(key)
 127             if video_url is not None:
 128                 m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
 129                 if m_size is not None:
 130                     width, height = map(int_or_none, (m_size.group(1), m_size.group(2)))
 131                 else:
 132                     width, height = None, None
 133                 formats.append({
 134                     'url': video_url,
 135                     'ext': 'mp4',
 136                     'format_id': format_id,
 137                     'width': width,
 138                     'height': height,
 139                 })
 140         if not formats:
 141             raise ExtractorError('Unable to extract video URL')
 142
 143         # subtitles
 144         video_subtitles = self.extract_subtitles(video_id, webpage)
 145
 146         view_count = str_to_int(self._search_regex(
 147             r'video_views_count[^>]+>\s+([\d\.,]+)',
 148             webpage, 'view count', fatal=False))
 149
 150         title = self._og_search_title(webpage, default=None)
 151         if title is None:
 152             title = self._html_search_regex(
 153                 r'(?s)<span\s+id="video_title"[^>]*>(.*?)</span>', webpage,
 154                 'title')
 155
 156         return {
 157             'id': video_id,
 158             'formats': formats,
 159             'uploader': info['owner.screenname'],
 160             'upload_date': video_upload_date,
 161             'title': title,
 162             'subtitles': video_subtitles,
 163             'thumbnail': info['thumbnail_url'],
 164             'age_limit': age_limit,
 165             'view_count': view_count,
 166         }
 167
 168     def _get_subtitles(self, video_id, webpage):
 169         try:
 170             sub_list = self._download_webpage(
 171                 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
 172                 video_id, note=False)
 173         except ExtractorError as err:
 174             self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
 175             return {}
 176         info = json.loads(sub_list)
 177         if (info['total'] > 0):
 178             sub_lang_list = dict((l['language'], [{'url': l['url'], 'ext': 'srt'}]) for l in info['list'])
 179             return sub_lang_list
 180         self._downloader.report_warning('video doesn\'t have subtitles')
 181         return {}
 182
 183
 184 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 185     IE_NAME = 'dailymotion:playlist'
 186     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
 187     _MORE_PAGES_INDICATOR = r'(?s)<div class="pages[^"]*">.*?<a\s+class="[^"]*?icon-arrow_right[^"]*?"'
 188     _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
 189     _TESTS = [{
 190         'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q',
 191         'info_dict': {
 192             'title': 'SPORT',
 193             'id': 'xv4bw_nqtv_sport',
 194         },
 195         'playlist_mincount': 20,
 196     }]
 197
 198     def _extract_entries(self, id):
 199         video_ids = []
 200         for pagenum in itertools.count(1):
 201             request = self._build_request(self._PAGE_TEMPLATE % (id, pagenum))
 202             webpage = self._download_webpage(request,
 203                                              id, 'Downloading page %s' % pagenum)
 204
 205             video_ids.extend(re.findall(r'data-xid="(.+?)"', webpage))
 206
 207             if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:
 208                 break
 209         return [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
 210                 for video_id in orderedSet(video_ids)]
 211
 212     def _real_extract(self, url):
 213         mobj = re.match(self._VALID_URL, url)
 214         playlist_id = mobj.group('id')
 215         webpage = self._download_webpage(url, playlist_id)
 216
 217         return {
 218             '_type': 'playlist',
 219             'id': playlist_id,
 220             'title': self._og_search_title(webpage),
 221             'entries': self._extract_entries(playlist_id),
 222         }
 223
 224
 225 class DailymotionUserIE(DailymotionPlaylistIE):
 226     IE_NAME = 'dailymotion:user'
 227     _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/user/(?P<user>[^/]+)'
 228     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
 229     _TESTS = [{
 230         'url': 'https://www.dailymotion.com/user/nqtv',
 231         'info_dict': {
 232             'id': 'nqtv',
 233             'title': 'Rémi Gaillard',
 234         },
 235         'playlist_mincount': 100,
 236     }]
 237
 238     def _real_extract(self, url):
 239         mobj = re.match(self._VALID_URL, url)
 240         user = mobj.group('user')
 241         webpage = self._download_webpage(url, user)
 242         full_user = unescapeHTML(self._html_search_regex(
 243             r'<a class="nav-image" title="([^"]+)" href="/%s">' % re.escape(user),
 244             webpage, 'user'))
 245
 246         return {
 247             '_type': 'playlist',
 248             'id': user,
 249             'title': full_user,
 250             'entries': self._extract_entries(user),
 251         }