Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/dailymotion.py

   1 import re
   2 import json
   3 import itertools
   4
   5 from .common import InfoExtractor
   6 from .subtitles import SubtitlesInfoExtractor
   7
   8 from ..utils import (
   9     compat_urllib_request,
  10     compat_str,
  11     get_element_by_attribute,
  12     get_element_by_id,
  13     orderedSet,
  14
  15     ExtractorError,
  16 )
  17
  18 class DailymotionBaseInfoExtractor(InfoExtractor):
  19     @staticmethod
  20     def _build_request(url):
  21         """Build a request with the family filter disabled"""
  22         request = compat_urllib_request.Request(url)
  23         request.add_header('Cookie', 'family_filter=off')
  24         return request
  25
  26 class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor):
  27     """Information Extractor for Dailymotion"""
  28
  29     _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/(?:embed/)?video/([^/]+)'
  30     IE_NAME = u'dailymotion'
  31     _TESTS = [
  32         {
  33             u'url': u'http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech',
  34             u'file': u'x33vw9.mp4',
  35             u'md5': u'392c4b85a60a90dc4792da41ce3144eb',
  36             u'info_dict': {
  37                 u"uploader": u"Amphora Alex and Van .",
  38                 u"title": u"Tutoriel de Youtubeur\"DL DES VIDEO DE YOUTUBE\""
  39             }
  40         },
  41         # Vevo video
  42         {
  43             u'url': u'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
  44             u'file': u'USUV71301934.mp4',
  45             u'info_dict': {
  46                 u'title': u'Roar (Official)',
  47                 u'uploader': u'Katy Perry',
  48                 u'upload_date': u'20130905',
  49             },
  50             u'params': {
  51                 u'skip_download': True,
  52             },
  53             u'skip': u'VEVO is only available in some countries',
  54         },
  55     ]
  56
  57     def _real_extract(self, url):
  58         # Extract id and simplified title from URL
  59         mobj = re.match(self._VALID_URL, url)
  60
  61         video_id = mobj.group(1).split('_')[0].split('?')[0]
  62
  63         video_extension = 'mp4'
  64         url = 'http://www.dailymotion.com/video/%s' % video_id
  65
  66         # Retrieve video webpage to extract further information
  67         request = self._build_request(url)
  68         webpage = self._download_webpage(request, video_id)
  69
  70         # Extract URL, uploader and title from webpage
  71         self.report_extraction(video_id)
  72
  73         # It may just embed a vevo video:
  74         m_vevo = re.search(
  75             r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?videoId=(?P<id>[\w]*)',
  76             webpage)
  77         if m_vevo is not None:
  78             vevo_id = m_vevo.group('id')
  79             self.to_screen(u'Vevo video detected: %s' % vevo_id)
  80             return self.url_result(u'vevo:%s' % vevo_id, ie='Vevo')
  81
  82         video_uploader = self._search_regex([r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>',
  83                                              # Looking for official user
  84                                              r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'],
  85                                             webpage, 'video uploader')
  86
  87         video_upload_date = None
  88         mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
  89         if mobj is not None:
  90             video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1)
  91
  92         embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id
  93         embed_page = self._download_webpage(embed_url, video_id,
  94                                             u'Downloading embed page')
  95         info = self._search_regex(r'var info = ({.*?}),$', embed_page,
  96             'video info', flags=re.MULTILINE)
  97         info = json.loads(info)
  98         if info.get('error') is not None:
  99             msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title']
 100             raise ExtractorError(msg, expected=True)
 101
 102         # TODO: support choosing qualities
 103
 104         for key in ['stream_h264_hd1080_url','stream_h264_hd_url',
 105                     'stream_h264_hq_url','stream_h264_url',
 106                     'stream_h264_ld_url']:
 107             if info.get(key):#key in info and info[key]:
 108                 max_quality = key
 109                 self.to_screen(u'Using %s' % key)
 110                 break
 111         else:
 112             raise ExtractorError(u'Unable to extract video URL')
 113         video_url = info[max_quality]
 114
 115         # subtitles
 116         video_subtitles = self.extract_subtitles(video_id)
 117         if self._downloader.params.get('listsubtitles', False):
 118             self._list_available_subtitles(video_id)
 119             return
 120
 121         return [{
 122             'id':       video_id,
 123             'url':      video_url,
 124             'uploader': video_uploader,
 125             'upload_date':  video_upload_date,
 126             'title':    self._og_search_title(webpage),
 127             'ext':      video_extension,
 128             'subtitles':    video_subtitles,
 129             'thumbnail': info['thumbnail_url']
 130         }]
 131
 132     def _get_available_subtitles(self, video_id):
 133         try:
 134             sub_list = self._download_webpage(
 135                 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
 136                 video_id, note=False)
 137         except ExtractorError as err:
 138             self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err))
 139             return {}
 140         info = json.loads(sub_list)
 141         if (info['total'] > 0):
 142             sub_lang_list = dict((l['language'], l['url']) for l in info['list'])
 143             return sub_lang_list
 144         self._downloader.report_warning(u'video doesn\'t have subtitles')
 145         return {}
 146
 147
 148 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 149     IE_NAME = u'dailymotion:playlist'
 150     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
 151     _MORE_PAGES_INDICATOR = r'<div class="next">.*?<a.*?href="/playlist/.+?".*?>.*?</a>.*?</div>'
 152     _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
 153
 154     def _extract_entries(self, id):
 155         video_ids = []
 156         for pagenum in itertools.count(1):
 157             request = self._build_request(self._PAGE_TEMPLATE % (id, pagenum))
 158             webpage = self._download_webpage(request,
 159                                              id, u'Downloading page %s' % pagenum)
 160
 161             playlist_el = get_element_by_attribute(u'class', u'video_list', webpage)
 162             video_ids.extend(re.findall(r'data-id="(.+?)"', playlist_el))
 163
 164             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 165                 break
 166         return [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
 167                    for video_id in orderedSet(video_ids)]
 168
 169     def _real_extract(self, url):
 170         mobj = re.match(self._VALID_URL, url)
 171         playlist_id = mobj.group('id')
 172         webpage = self._download_webpage(url, playlist_id)
 173
 174         return {'_type': 'playlist',
 175                 'id': playlist_id,
 176                 'title': get_element_by_id(u'playlist_name', webpage),
 177                 'entries': self._extract_entries(playlist_id),
 178                 }
 179
 180
 181 class DailymotionUserIE(DailymotionPlaylistIE):
 182     IE_NAME = u'dailymotion:user'
 183     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/user/(?P<user>[^/]+)'
 184     _MORE_PAGES_INDICATOR = r'<div class="next">.*?<a.*?href="/user/.+?".*?>.*?</a>.*?</div>'
 185     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
 186
 187     def _real_extract(self, url):
 188         mobj = re.match(self._VALID_URL, url)
 189         user = mobj.group('user')
 190         webpage = self._download_webpage(url, user)
 191         full_user = self._html_search_regex(
 192             r'<a class="label" href="/%s".*?>(.*?)</' % re.escape(user),
 193             webpage, u'user', flags=re.DOTALL)
 194
 195         return {
 196             '_type': 'playlist',
 197             'id': user,
 198             'title': full_user,
 199             'entries': self._extract_entries(user),
 200         }