Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/dailymotion.py

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