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                 'duration': 74,
  57             }
  58         },
  59         # Vevo video
  60         {
  61             'url': 'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
  62             'info_dict': {
  63                 'title': 'Roar (Official)',
  64                 'id': 'USUV71301934',
  65                 'ext': 'mp4',
  66                 'uploader': 'Katy Perry',
  67                 'upload_date': '20130905',
  68             },
  69             'params': {
  70                 'skip_download': True,
  71             },
  72             'skip': 'VEVO is only available in some countries',
  73         },
  74         # age-restricted video
  75         {
  76             'url': 'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband',
  77             'md5': '0d667a7b9cebecc3c89ee93099c4159d',
  78             'info_dict': {
  79                 'id': 'xyh2zz',
  80                 'ext': 'mp4',
  81                 'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]',
  82                 'uploader': 'HotWaves1012',
  83                 'age_limit': 18,
  84             }
  85         }
  86     ]
  87
  88     def _real_extract(self, url):
  89         video_id = self._match_id(url)
  90         url = 'https://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[^"]*?video=(?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'<meta property="video:release_date" content="([0-9]{4})-([0-9]{2})-([0-9]{2}).+?"/>', webpage)
 112         if mobj is not None:
 113             video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3)
 114
 115         embed_url = 'https://www.dailymotion.com/embed/video/%s' % video_id
 116         embed_request = self._build_request(embed_url)
 117         embed_page = self._download_webpage(
 118             embed_request, video_id, 'Downloading embed page')
 119         info = self._search_regex(r'var info = ({.*?}),$', embed_page,
 120                                   'video info', flags=re.MULTILINE)
 121         info = json.loads(info)
 122         if info.get('error') is not None:
 123             msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title']
 124             raise ExtractorError(msg, expected=True)
 125
 126         formats = []
 127         for (key, format_id) in self._FORMATS:
 128             video_url = info.get(key)
 129             if video_url is not None:
 130                 m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
 131                 if m_size is not None:
 132                     width, height = map(int_or_none, (m_size.group(1), m_size.group(2)))
 133                 else:
 134                     width, height = None, None
 135                 formats.append({
 136                     'url': video_url,
 137                     'ext': 'mp4',
 138                     'format_id': format_id,
 139                     'width': width,
 140                     'height': height,
 141                 })
 142         if not formats:
 143             raise ExtractorError('Unable to extract video URL')
 144
 145         # subtitles
 146         video_subtitles = self.extract_subtitles(video_id, webpage)
 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             'duration': info['duration']
 169         }
 170
 171     def _get_subtitles(self, video_id, webpage):
 172         try:
 173             sub_list = self._download_webpage(
 174                 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
 175                 video_id, note=False)
 176         except ExtractorError as err:
 177             self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
 178             return {}
 179         info = json.loads(sub_list)
 180         if (info['total'] > 0):
 181             sub_lang_list = dict((l['language'], [{'url': l['url'], 'ext': 'srt'}]) for l in info['list'])
 182             return sub_lang_list
 183         self._downloader.report_warning('video doesn\'t have subtitles')
 184         return {}
 185
 186
 187 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 188     IE_NAME = 'dailymotion:playlist'
 189     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
 190     _MORE_PAGES_INDICATOR = r'(?s)<div class="pages[^"]*">.*?<a\s+class="[^"]*?icon-arrow_right[^"]*?"'
 191     _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
 192     _TESTS = [{
 193         'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q',
 194         'info_dict': {
 195             'title': 'SPORT',
 196             'id': 'xv4bw_nqtv_sport',
 197         },
 198         'playlist_mincount': 20,
 199     }]
 200
 201     def _extract_entries(self, id):
 202         video_ids = []
 203         for pagenum in itertools.count(1):
 204             request = self._build_request(self._PAGE_TEMPLATE % (id, pagenum))
 205             webpage = self._download_webpage(request,
 206                                              id, 'Downloading page %s' % pagenum)
 207
 208             video_ids.extend(re.findall(r'data-xid="(.+?)"', webpage))
 209
 210             if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:
 211                 break
 212         return [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
 213                 for video_id in orderedSet(video_ids)]
 214
 215     def _real_extract(self, url):
 216         mobj = re.match(self._VALID_URL, url)
 217         playlist_id = mobj.group('id')
 218         webpage = self._download_webpage(url, playlist_id)
 219
 220         return {
 221             '_type': 'playlist',
 222             'id': playlist_id,
 223             'title': self._og_search_title(webpage),
 224             'entries': self._extract_entries(playlist_id),
 225         }
 226
 227
 228 class DailymotionUserIE(DailymotionPlaylistIE):
 229     IE_NAME = 'dailymotion:user'
 230     _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?:(?:old/)?user/)?(?P<user>[^/]+)$'
 231     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
 232     _TESTS = [{
 233         'url': 'https://www.dailymotion.com/user/nqtv',
 234         'info_dict': {
 235             'id': 'nqtv',
 236             'title': 'Rémi Gaillard',
 237         },
 238         'playlist_mincount': 100,
 239     }]
 240
 241     def _real_extract(self, url):
 242         mobj = re.match(self._VALID_URL, url)
 243         user = mobj.group('user')
 244         webpage = self._download_webpage(
 245             'https://www.dailymotion.com/user/%s' % user, user)
 246         full_user = unescapeHTML(self._html_search_regex(
 247             r'<a class="nav-image" title="([^"]+)" href="/%s">' % re.escape(user),
 248             webpage, 'user'))
 249
 250         return {
 251             '_type': 'playlist',
 252             'id': user,
 253             'title': full_user,
 254             'entries': self._extract_entries(user),
 255         }
 256
 257
 258 class DailymotionCloudIE(DailymotionBaseInfoExtractor):
 259     _VALID_URL_PREFIX = r'http://api\.dmcloud\.net/(?:player/)?embed/'
 260     _VALID_URL = r'%s[^/]+/(?P<id>[^/?]+)' % _VALID_URL_PREFIX
 261     _VALID_EMBED_URL = r'%s[^/]+/[^\'"]+' % _VALID_URL_PREFIX
 262
 263     _TESTS = [{
 264         # From http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html
 265         # Tested at FranceTvInfo_2
 266         'url': 'http://api.dmcloud.net/embed/4e7343f894a6f677b10006b4/556e03339473995ee145930c?auth=1464865870-0-jyhsm84b-ead4c701fb750cf9367bf4447167a3db&autoplay=1',
 267         'only_matching': True,
 268     }, {
 269         # http://www.francetvinfo.fr/societe/larguez-les-amarres-le-cobaturage-se-developpe_980101.html
 270         'url': 'http://api.dmcloud.net/player/embed/4e7343f894a6f677b10006b4/559545469473996d31429f06?auth=1467430263-0-90tglw2l-a3a4b64ed41efe48d7fccad85b8b8fda&autoplay=1',
 271         'only_matching': True,
 272     }]
 273
 274     @classmethod
 275     def _extract_dmcloud_url(self, webpage):
 276         mobj = re.search(r'<iframe[^>]+src=[\'"](%s)[\'"]' % self._VALID_EMBED_URL, webpage)
 277         if mobj:
 278             return mobj.group(1)
 279
 280         mobj = re.search(
 281             r'<input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=[\'"](%s)[\'"]' % self._VALID_EMBED_URL,
 282             webpage)
 283         if mobj:
 284             return mobj.group(1)
 285
 286     def _real_extract(self, url):
 287         video_id = self._match_id(url)
 288
 289         request = self._build_request(url)
 290         webpage = self._download_webpage(request, video_id)
 291
 292         title = self._html_search_regex(r'<title>([^>]+)</title>', webpage, 'title')
 293
 294         video_info = self._parse_json(self._search_regex(
 295             r'var\s+info\s*=\s*([^;]+);', webpage, 'video info'), video_id)
 296
 297         # TODO: parse ios_url, which is in fact a manifest
 298         video_url = video_info['mp4_url']
 299
 300         return {
 301             'id': video_id,
 302             'url': video_url,
 303             'title': title,
 304             'thumbnail': video_info.get('thumbnail_url'),
 305         }