_ 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 NoAutoSubtitlesIE
   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 DailyMotionSubtitlesIE(NoAutoSubtitlesIE):
  22
  23     def _get_available_subtitles(self, video_id):
  24         request = compat_urllib_request.Request('https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id)
  25         try:
  26             sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8')
  27         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  28             self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err))
  29             return {}
  30         info = json.loads(sub_list)
  31         if (info['total'] > 0):
  32             sub_lang_list = dict((l['language'], l['url']) for l in info['list'])
  33             return sub_lang_list
  34         self._downloader.report_warning(u'video doesn\'t have subtitles')
  35         return {}
  36
  37 class DailymotionIE(DailyMotionSubtitlesIE, InfoExtractor):
  38     """Information Extractor for Dailymotion"""
  39
  40     _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/(?:embed/)?video/([^/]+)'
  41     IE_NAME = u'dailymotion'
  42     _TEST = {
  43         u'url': u'http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech',
  44         u'file': u'x33vw9.mp4',
  45         u'md5': u'392c4b85a60a90dc4792da41ce3144eb',
  46         u'info_dict': {
  47             u"uploader": u"Amphora Alex and Van .",
  48             u"title": u"Tutoriel de Youtubeur\"DL DES VIDEO DE YOUTUBE\""
  49         }
  50     }
  51
  52     def _real_extract(self, url):
  53         # Extract id and simplified title from URL
  54         mobj = re.match(self._VALID_URL, url)
  55
  56         video_id = mobj.group(1).split('_')[0].split('?')[0]
  57
  58         video_extension = 'mp4'
  59         url = 'http://www.dailymotion.com/video/%s' % video_id
  60
  61         # Retrieve video webpage to extract further information
  62         request = compat_urllib_request.Request(url)
  63         request.add_header('Cookie', 'family_filter=off')
  64         webpage = self._download_webpage(request, video_id)
  65
  66         # Extract URL, uploader and title from webpage
  67         self.report_extraction(video_id)
  68
  69         video_uploader = self._search_regex([r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>',
  70                                              # Looking for official user
  71                                              r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'],
  72                                             webpage, 'video uploader')
  73
  74         video_upload_date = None
  75         mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
  76         if mobj is not None:
  77             video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1)
  78
  79         embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id
  80         embed_page = self._download_webpage(embed_url, video_id,
  81                                             u'Downloading embed page')
  82         info = self._search_regex(r'var info = ({.*?}),$', embed_page,
  83             'video info', flags=re.MULTILINE)
  84         info = json.loads(info)
  85
  86         # TODO: support choosing qualities
  87
  88         for key in ['stream_h264_hd1080_url','stream_h264_hd_url',
  89                     'stream_h264_hq_url','stream_h264_url',
  90                     'stream_h264_ld_url']:
  91             if info.get(key):#key in info and info[key]:
  92                 max_quality = key
  93                 self.to_screen(u'Using %s' % key)
  94                 break
  95         else:
  96             raise ExtractorError(u'Unable to extract video URL')
  97         video_url = info[max_quality]
  98
  99         # subtitles
 100         video_subtitles = None
 101         video_webpage = None
 102
 103         if self._downloader.params.get('writesubtitles', False) or self._downloader.params.get('allsubtitles', False):
 104             video_subtitles = self._extract_subtitles(video_id)
 105         elif self._downloader.params.get('writeautomaticsub', False):
 106             video_subtitles = self._request_automatic_caption(video_id, video_webpage)
 107
 108         if self._downloader.params.get('listsubtitles', False):
 109             self._list_available_subtitles(video_id)
 110             return
 111
 112         return [{
 113             'id':       video_id,
 114             'url':      video_url,
 115             'uploader': video_uploader,
 116             'upload_date':  video_upload_date,
 117             'title':    self._og_search_title(webpage),
 118             'ext':      video_extension,
 119             'subtitles':    video_subtitles,
 120             'thumbnail': info['thumbnail_url']
 121         }]
 122
 123
 124 class DailymotionPlaylistIE(InfoExtractor):
 125     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
 126     _MORE_PAGES_INDICATOR = r'<div class="next">.*?<a.*?href="/playlist/.+?".*?>.*?</a>.*?</div>'
 127
 128     def _real_extract(self, url):
 129         mobj = re.match(self._VALID_URL, url)
 130         playlist_id =  mobj.group('id')
 131         video_ids = []
 132
 133         for pagenum in itertools.count(1):
 134             webpage = self._download_webpage('https://www.dailymotion.com/playlist/%s/%s' % (playlist_id, pagenum),
 135                                              playlist_id, u'Downloading page %s' % pagenum)
 136
 137             playlist_el = get_element_by_attribute(u'class', u'video_list', webpage)
 138             video_ids.extend(re.findall(r'data-id="(.+?)" data-ext-id', playlist_el))
 139
 140             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 141                 break
 142
 143         entries = [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
 144                    for video_id in video_ids]
 145         return {'_type': 'playlist',
 146                 'id': playlist_id,
 147                 'title': get_element_by_id(u'playlist_name', webpage),
 148                 'entries': entries,
 149                 }