_ Git - youtube-dl/blob - youtube_dl/extractor/dailymotion.py

   1 import re
   2 import json
   3 import itertools
   4 import socket
   5
   6 from .common import InfoExtractor
   7 from .subtitles import SubtitlesInfoExtractor
   8
   9 from ..utils import (
  10     compat_http_client,
  11     compat_urllib_error,
  12     compat_urllib_request,
  13     compat_str,
  14     get_element_by_attribute,
  15     get_element_by_id,
  16
  17     ExtractorError,
  18 )
  19
  20
  21 class DailymotionIE(SubtitlesInfoExtractor):
  22     """Information Extractor for Dailymotion"""
  23
  24     _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/(?:embed/)?video/([^/]+)'
  25     IE_NAME = u'dailymotion'
  26     _TEST = {
  27         u'url': u'http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech',
  28         u'file': u'x33vw9.mp4',
  29         u'md5': u'392c4b85a60a90dc4792da41ce3144eb',
  30         u'info_dict': {
  31             u"uploader": u"Amphora Alex and Van .",
  32             u"title": u"Tutoriel de Youtubeur\"DL DES VIDEO DE YOUTUBE\""
  33         }
  34     }
  35
  36     def _real_extract(self, url):
  37         # Extract id and simplified title from URL
  38         mobj = re.match(self._VALID_URL, url)
  39
  40         video_id = mobj.group(1).split('_')[0].split('?')[0]
  41
  42         video_extension = 'mp4'
  43         url = 'http://www.dailymotion.com/video/%s' % video_id
  44
  45         # Retrieve video webpage to extract further information
  46         request = compat_urllib_request.Request(url)
  47         request.add_header('Cookie', 'family_filter=off')
  48         webpage = self._download_webpage(request, video_id)
  49
  50         # Extract URL, uploader and title from webpage
  51         self.report_extraction(video_id)
  52
  53         video_uploader = self._search_regex([r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>',
  54                                              # Looking for official user
  55                                              r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'],
  56                                             webpage, 'video uploader')
  57
  58         video_upload_date = None
  59         mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
  60         if mobj is not None:
  61             video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1)
  62
  63         embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id
  64         embed_page = self._download_webpage(embed_url, video_id,
  65                                             u'Downloading embed page')
  66         info = self._search_regex(r'var info = ({.*?}),$', embed_page,
  67             'video info', flags=re.MULTILINE)
  68         info = json.loads(info)
  69
  70         # TODO: support choosing qualities
  71
  72         for key in ['stream_h264_hd1080_url','stream_h264_hd_url',
  73                     'stream_h264_hq_url','stream_h264_url',
  74                     'stream_h264_ld_url']:
  75             if info.get(key):#key in info and info[key]:
  76                 max_quality = key
  77                 self.to_screen(u'Using %s' % key)
  78                 break
  79         else:
  80             raise ExtractorError(u'Unable to extract video URL')
  81         video_url = info[max_quality]
  82
  83         # subtitles
  84         video_subtitles = self.extract_subtitles(video_id)
  85         if self._downloader.params.get('listsubtitles', False):
  86             self._list_available_subtitles(video_id)
  87             return
  88
  89         return [{
  90             'id':       video_id,
  91             'url':      video_url,
  92             'uploader': video_uploader,
  93             'upload_date':  video_upload_date,
  94             'title':    self._og_search_title(webpage),
  95             'ext':      video_extension,
  96             'subtitles':    video_subtitles,
  97             'thumbnail': info['thumbnail_url']
  98         }]
  99
 100     def _get_available_subtitles(self, video_id):
 101         request = compat_urllib_request.Request('https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id)
 102         try:
 103             sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8')
 104         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
 105             self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err))
 106             return {}
 107         info = json.loads(sub_list)
 108         if (info['total'] > 0):
 109             sub_lang_list = dict((l['language'], l['url']) for l in info['list'])
 110             return sub_lang_list
 111         self._downloader.report_warning(u'video doesn\'t have subtitles')
 112         return {}
 113
 114
 115 class DailymotionPlaylistIE(InfoExtractor):
 116     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
 117     _MORE_PAGES_INDICATOR = r'<div class="next">.*?<a.*?href="/playlist/.+?".*?>.*?</a>.*?</div>'
 118
 119     def _real_extract(self, url):
 120         mobj = re.match(self._VALID_URL, url)
 121         playlist_id =  mobj.group('id')
 122         video_ids = []
 123
 124         for pagenum in itertools.count(1):
 125             webpage = self._download_webpage('https://www.dailymotion.com/playlist/%s/%s' % (playlist_id, pagenum),
 126                                              playlist_id, u'Downloading page %s' % pagenum)
 127
 128             playlist_el = get_element_by_attribute(u'class', u'video_list', webpage)
 129             video_ids.extend(re.findall(r'data-id="(.+?)" data-ext-id', playlist_el))
 130
 131             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 132                 break
 133
 134         entries = [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
 135                    for video_id in video_ids]
 136         return {'_type': 'playlist',
 137                 'id': playlist_id,
 138                 'title': get_element_by_id(u'playlist_name', webpage),
 139                 'entries': entries,
 140                 }