_ 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     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
  32     _FORMATS = [
  33         (u'stream_h264_ld_url', u'ld'),
  34         (u'stream_h264_url', u'standard'),
  35         (u'stream_h264_hq_url', u'hq'),
  36         (u'stream_h264_hd_url', u'hd'),
  37         (u'stream_h264_hd1080_url', u'hd180'),
  38     ]
  39
  40     _TESTS = [
  41         {
  42             u'url': u'http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech',
  43             u'file': u'x33vw9.mp4',
  44             u'md5': u'392c4b85a60a90dc4792da41ce3144eb',
  45             u'info_dict': {
  46                 u"uploader": u"Amphora Alex and Van .",
  47                 u"title": u"Tutoriel de Youtubeur\"DL DES VIDEO DE YOUTUBE\""
  48             }
  49         },
  50         # Vevo video
  51         {
  52             u'url': u'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
  53             u'file': u'USUV71301934.mp4',
  54             u'info_dict': {
  55                 u'title': u'Roar (Official)',
  56                 u'uploader': u'Katy Perry',
  57                 u'upload_date': u'20130905',
  58             },
  59             u'params': {
  60                 u'skip_download': True,
  61             },
  62             u'skip': u'VEVO is only available in some countries',
  63         },
  64     ]
  65
  66     def _real_extract(self, url):
  67         # Extract id and simplified title from URL
  68         mobj = re.match(self._VALID_URL, url)
  69
  70         video_id = mobj.group(1).split('_')[0].split('?')[0]
  71
  72         url = 'http://www.dailymotion.com/video/%s' % video_id
  73
  74         # Retrieve video webpage to extract further information
  75         request = self._build_request(url)
  76         webpage = self._download_webpage(request, video_id)
  77
  78         # Extract URL, uploader and title from webpage
  79         self.report_extraction(video_id)
  80
  81         # It may just embed a vevo video:
  82         m_vevo = re.search(
  83             r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?videoId=(?P<id>[\w]*)',
  84             webpage)
  85         if m_vevo is not None:
  86             vevo_id = m_vevo.group('id')
  87             self.to_screen(u'Vevo video detected: %s' % vevo_id)
  88             return self.url_result(u'vevo:%s' % vevo_id, ie='Vevo')
  89
  90         video_uploader = self._search_regex([r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>',
  91                                              # Looking for official user
  92                                              r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'],
  93                                             webpage, 'video uploader')
  94
  95         video_upload_date = None
  96         mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
  97         if mobj is not None:
  98             video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1)
  99
 100         embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id
 101         embed_page = self._download_webpage(embed_url, video_id,
 102                                             u'Downloading embed page')
 103         info = self._search_regex(r'var info = ({.*?}),$', embed_page,
 104             'video info', flags=re.MULTILINE)
 105         info = json.loads(info)
 106         if info.get('error') is not None:
 107             msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title']
 108             raise ExtractorError(msg, expected=True)
 109
 110         formats = []
 111         for (key, format_id) in self._FORMATS:
 112             video_url = info.get(key)
 113             if video_url is not None:
 114                 m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
 115                 if m_size is not None:
 116                     width, height = m_size.group(1), m_size.group(2)
 117                 else:
 118                     width, height = None, None
 119                 formats.append({
 120                     'url': video_url,
 121                     'ext': 'mp4',
 122                     'format_id': format_id,
 123                     'width': width,
 124                     'height': height,
 125                 })
 126         if not formats:
 127             raise ExtractorError(u'Unable to extract video URL')
 128
 129         # subtitles
 130         video_subtitles = self.extract_subtitles(video_id)
 131         if self._downloader.params.get('listsubtitles', False):
 132             self._list_available_subtitles(video_id)
 133             return
 134
 135         return [{
 136             'id':       video_id,
 137             'formats': formats,
 138             'uploader': video_uploader,
 139             'upload_date':  video_upload_date,
 140             'title':    self._og_search_title(webpage),
 141             'subtitles':    video_subtitles,
 142             'thumbnail': info['thumbnail_url']
 143         }]
 144
 145     def _get_available_subtitles(self, video_id):
 146         try:
 147             sub_list = self._download_webpage(
 148                 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
 149                 video_id, note=False)
 150         except ExtractorError as err:
 151             self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err))
 152             return {}
 153         info = json.loads(sub_list)
 154         if (info['total'] > 0):
 155             sub_lang_list = dict((l['language'], l['url']) for l in info['list'])
 156             return sub_lang_list
 157         self._downloader.report_warning(u'video doesn\'t have subtitles')
 158         return {}
 159
 160
 161 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 162     IE_NAME = u'dailymotion:playlist'
 163     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
 164     _MORE_PAGES_INDICATOR = r'<div class="next">.*?<a.*?href="/playlist/.+?".*?>.*?</a>.*?</div>'
 165     _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
 166
 167     def _extract_entries(self, id):
 168         video_ids = []
 169         for pagenum in itertools.count(1):
 170             request = self._build_request(self._PAGE_TEMPLATE % (id, pagenum))
 171             webpage = self._download_webpage(request,
 172                                              id, u'Downloading page %s' % pagenum)
 173
 174             playlist_el = get_element_by_attribute(u'class', u'video_list', webpage)
 175             video_ids.extend(re.findall(r'data-id="(.+?)"', playlist_el))
 176
 177             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 178                 break
 179         return [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
 180                    for video_id in orderedSet(video_ids)]
 181
 182     def _real_extract(self, url):
 183         mobj = re.match(self._VALID_URL, url)
 184         playlist_id = mobj.group('id')
 185         webpage = self._download_webpage(url, playlist_id)
 186
 187         return {'_type': 'playlist',
 188                 'id': playlist_id,
 189                 'title': get_element_by_id(u'playlist_name', webpage),
 190                 'entries': self._extract_entries(playlist_id),
 191                 }
 192
 193
 194 class DailymotionUserIE(DailymotionPlaylistIE):
 195     IE_NAME = u'dailymotion:user'
 196     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/user/(?P<user>[^/]+)'
 197     _MORE_PAGES_INDICATOR = r'<div class="next">.*?<a.*?href="/user/.+?".*?>.*?</a>.*?</div>'
 198     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
 199
 200     def _real_extract(self, url):
 201         mobj = re.match(self._VALID_URL, url)
 202         user = mobj.group('user')
 203         webpage = self._download_webpage(url, user)
 204         full_user = self._html_search_regex(
 205             r'<a class="label" href="/%s".*?>(.*?)</' % re.escape(user),
 206             webpage, u'user', flags=re.DOTALL)
 207
 208         return {
 209             '_type': 'playlist',
 210             'id': user,
 211             'title': full_user,
 212             'entries': self._extract_entries(user),
 213         }