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