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

   1 import re
   2 import json
   3 import socket
   4
   5 from .common import InfoExtractor
   6 from .subtitles import NoAutoSubtitlesIE
   7
   8 from ..utils import (
   9     compat_http_client,
  10     compat_urllib_error,
  11     compat_urllib_request,
  12     compat_str,
  13     get_element_by_attribute,
  14     get_element_by_id,
  15
  16     ExtractorError,
  17 )
  18
  19
  20 class DailyMotionSubtitlesIE(NoAutoSubtitlesIE):
  21
  22     def _get_available_subtitles(self, video_id):
  23         request = compat_urllib_request.Request('https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id)
  24         try:
  25             sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8')
  26         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  27             self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err))
  28             return {}
  29         info = json.loads(sub_list)
  30         if (info['total'] > 0):
  31             sub_lang_list = dict((l['language'], l['url']) for l in info['list'])
  32             return sub_lang_list
  33         self._downloader.report_warning(u'video doesn\'t have subtitles')
  34         return {}
  35
  36 class DailymotionIE(DailyMotionSubtitlesIE):
  37     """Information Extractor for Dailymotion"""
  38
  39     _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/video/([^/]+)'
  40     IE_NAME = u'dailymotion'
  41     _TEST = {
  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"Alex and Van .",
  47             u"title": u"Tutoriel de Youtubeur\"DL DES VIDEO DE YOUTUBE\""
  48         }
  49     }
  50
  51     def _real_extract(self, url):
  52         # Extract id and simplified title from URL
  53         mobj = re.match(self._VALID_URL, url)
  54
  55         video_id = mobj.group(1).split('_')[0].split('?')[0]
  56
  57         video_extension = 'mp4'
  58
  59         # Retrieve video webpage to extract further information
  60         request = compat_urllib_request.Request(url)
  61         request.add_header('Cookie', 'family_filter=off')
  62         webpage = self._download_webpage(request, video_id)
  63
  64         # Extract URL, uploader and title from webpage
  65         self.report_extraction(video_id)
  66
  67         video_uploader = self._search_regex([r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>',
  68                                              # Looking for official user
  69                                              r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'],
  70                                             webpage, 'video uploader')
  71
  72         video_upload_date = None
  73         mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
  74         if mobj is not None:
  75             video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1)
  76
  77         embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id
  78         embed_page = self._download_webpage(embed_url, video_id,
  79                                             u'Downloading embed page')
  80         info = self._search_regex(r'var info = ({.*?}),', embed_page, 'video info')
  81         info = json.loads(info)
  82
  83         # TODO: support choosing qualities
  84
  85         for key in ['stream_h264_hd1080_url','stream_h264_hd_url',
  86                     'stream_h264_hq_url','stream_h264_url',
  87                     'stream_h264_ld_url']:
  88             if info.get(key):  # key in info and info[key]:
  89                 max_quality = key
  90                 self.to_screen(u'%s: Using %s' % (video_id, key))
  91                 break
  92         else:
  93             raise ExtractorError(u'Unable to extract video URL')
  94         video_url = info[max_quality]
  95
  96         # subtitles
  97         video_subtitles = None
  98         video_webpage = None
  99
 100         if self._downloader.params.get('writesubtitles', False) or self._downloader.params.get('allsubtitles', False):
 101             video_subtitles = self._extract_subtitles(video_id)
 102         elif self._downloader.params.get('writeautomaticsub', False):
 103             video_subtitles = self._request_automatic_caption(video_id, video_webpage)
 104
 105         if self._downloader.params.get('listsubtitles', False):
 106             self._list_available_subtitles(video_id)
 107             return
 108
 109         return [{
 110             'id':       video_id,
 111             'url':      video_url,
 112             'uploader': video_uploader,
 113             'upload_date':  video_upload_date,
 114             'title':    self._og_search_title(webpage),
 115             'ext':      video_extension,
 116             'subtitles':    video_subtitles,
 117             'thumbnail': info['thumbnail_url']
 118         }]
 119
 120
 121 class DailymotionPlaylistIE(InfoExtractor):
 122     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
 123     _MORE_PAGES_INDICATOR = r'<div class="next">.*?<a.*?href="/playlist/.+?".*?>.*?</a>.*?</div>'
 124
 125     def _real_extract(self, url):
 126         mobj = re.match(self._VALID_URL, url)
 127         playlist_id =  mobj.group('id')
 128         video_ids = []
 129
 130         for pagenum in itertools.count(1):
 131             webpage = self._download_webpage('https://www.dailymotion.com/playlist/%s/%s' % (playlist_id, pagenum),
 132                                              playlist_id, u'Downloading page %s' % pagenum)
 133
 134             playlist_el = get_element_by_attribute(u'class', u'video_list', webpage)
 135             video_ids.extend(re.findall(r'data-id="(.+?)" data-ext-id', playlist_el))
 136
 137             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 138                 break
 139
 140         entries = [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
 141                    for video_id in video_ids]
 142         return {'_type': 'playlist',
 143                 'id': playlist_id,
 144                 'title': get_element_by_id(u'playlist_name', webpage),
 145                 'entries': entries,
 146                 }