_ 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
  18 class DailymotionIE(SubtitlesInfoExtractor):
  19     """Information Extractor for Dailymotion"""
  20
  21     _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/(?:embed/)?video/([^/]+)'
  22     IE_NAME = u'dailymotion'
  23     _TEST = {
  24         u'url': u'http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech',
  25         u'file': u'x33vw9.mp4',
  26         u'md5': u'392c4b85a60a90dc4792da41ce3144eb',
  27         u'info_dict': {
  28             u"uploader": u"Amphora Alex and Van .",
  29             u"title": u"Tutoriel de Youtubeur\"DL DES VIDEO DE YOUTUBE\""
  30         }
  31     }
  32
  33     def _real_extract(self, url):
  34         # Extract id and simplified title from URL
  35         mobj = re.match(self._VALID_URL, url)
  36
  37         video_id = mobj.group(1).split('_')[0].split('?')[0]
  38
  39         video_extension = 'mp4'
  40         url = 'http://www.dailymotion.com/video/%s' % video_id
  41
  42         # Retrieve video webpage to extract further information
  43         request = compat_urllib_request.Request(url)
  44         request.add_header('Cookie', 'family_filter=off')
  45         webpage = self._download_webpage(request, video_id)
  46
  47         # Extract URL, uploader and title from webpage
  48         self.report_extraction(video_id)
  49
  50         video_uploader = self._search_regex([r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>',
  51                                              # Looking for official user
  52                                              r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'],
  53                                             webpage, 'video uploader')
  54
  55         video_upload_date = None
  56         mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
  57         if mobj is not None:
  58             video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1)
  59
  60         embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id
  61         embed_page = self._download_webpage(embed_url, video_id,
  62                                             u'Downloading embed page')
  63         info = self._search_regex(r'var info = ({.*?}),$', embed_page,
  64             'video info', flags=re.MULTILINE)
  65         info = json.loads(info)
  66
  67         # TODO: support choosing qualities
  68
  69         for key in ['stream_h264_hd1080_url','stream_h264_hd_url',
  70                     'stream_h264_hq_url','stream_h264_url',
  71                     'stream_h264_ld_url']:
  72             if info.get(key):#key in info and info[key]:
  73                 max_quality = key
  74                 self.to_screen(u'Using %s' % key)
  75                 break
  76         else:
  77             raise ExtractorError(u'Unable to extract video URL')
  78         video_url = info[max_quality]
  79
  80         # subtitles
  81         video_subtitles = self.extract_subtitles(video_id)
  82         if self._downloader.params.get('listsubtitles', False):
  83             self._list_available_subtitles(video_id)
  84             return
  85
  86         return [{
  87             'id':       video_id,
  88             'url':      video_url,
  89             'uploader': video_uploader,
  90             'upload_date':  video_upload_date,
  91             'title':    self._og_search_title(webpage),
  92             'ext':      video_extension,
  93             'subtitles':    video_subtitles,
  94             'thumbnail': info['thumbnail_url']
  95         }]
  96
  97     def _get_available_subtitles(self, video_id):
  98         try:
  99             sub_list = self._download_webpage(
 100                 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
 101                 video_id, note=False)
 102         except ExtractorError as err:
 103             self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err))
 104             return {}
 105         info = json.loads(sub_list)
 106         if (info['total'] > 0):
 107             sub_lang_list = dict((l['language'], l['url']) for l in info['list'])
 108             return sub_lang_list
 109         self._downloader.report_warning(u'video doesn\'t have subtitles')
 110         return {}
 111
 112
 113 class DailymotionPlaylistIE(InfoExtractor):
 114     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
 115     _MORE_PAGES_INDICATOR = r'<div class="next">.*?<a.*?href="/playlist/.+?".*?>.*?</a>.*?</div>'
 116
 117     def _real_extract(self, url):
 118         mobj = re.match(self._VALID_URL, url)
 119         playlist_id =  mobj.group('id')
 120         video_ids = []
 121
 122         for pagenum in itertools.count(1):
 123             webpage = self._download_webpage('https://www.dailymotion.com/playlist/%s/%s' % (playlist_id, pagenum),
 124                                              playlist_id, u'Downloading page %s' % pagenum)
 125
 126             playlist_el = get_element_by_attribute(u'class', u'video_list', webpage)
 127             video_ids.extend(re.findall(r'data-id="(.+?)" data-ext-id', playlist_el))
 128
 129             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 130                 break
 131
 132         entries = [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
 133                    for video_id in video_ids]
 134         return {'_type': 'playlist',
 135                 'id': playlist_id,
 136                 'title': get_element_by_id(u'playlist_name', webpage),
 137                 'entries': entries,
 138                 }