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

   1 # coding: utf-8
   2
   3 import json
   4 import re
   5
   6 from .common import InfoExtractor
   7
   8 from ..utils import (
   9     compat_urllib_parse,
  10     unified_strdate,
  11 )
  12
  13
  14 class WatIE(InfoExtractor):
  15     _VALID_URL=r'http://www.wat.tv/.*-(?P<shortID>.*?)_.*?.html'
  16     IE_NAME = 'wat.tv'
  17     _TEST = {
  18         u'url': u'http://www.wat.tv/video/world-war-philadelphia-vost-6bv55_2fjr7_.html',
  19         u'file': u'10631273.mp4',
  20         u'md5': u'0a4fe7870f31eaeabb5e25fd8da8414a',
  21         u'info_dict': {
  22             u'title': u'World War Z - Philadelphia VOST',
  23             u'description': u'La menace est partout. Que se passe-t-il à Philadelphia ?\r\nWORLD WAR Z, avec Brad Pitt, au cinéma le 3 juillet.\r\nhttp://www.worldwarz.fr',
  24         }
  25     }
  26
  27     def download_video_info(self, real_id):
  28         # 'contentv4' is used in the website, but it also returns the related
  29         # videos, we don't need them
  30         info = self._download_webpage('http://www.wat.tv/interface/contentv3/' + real_id, real_id, 'Downloading video info')
  31         info = json.loads(info)
  32         return info['media']
  33
  34
  35     def _real_extract(self, url):
  36         def real_id_for_chapter(chapter):
  37             return chapter['tc_start'].split('-')[0]
  38         mobj = re.match(self._VALID_URL, url)
  39         short_id = mobj.group('shortID')
  40         webpage = self._download_webpage(url, short_id)
  41         real_id = self._search_regex(r'xtpage = ".*-(.*?)";', webpage, 'real id')
  42
  43         video_info = self.download_video_info(real_id)
  44         chapters = video_info['chapters']
  45         first_chapter = chapters[0]
  46
  47         if real_id_for_chapter(first_chapter) != real_id:
  48             self.to_screen('Multipart video detected')
  49             chapter_urls = []
  50             for chapter in chapters:
  51                 chapter_id = real_id_for_chapter(chapter)
  52                 # Yes, when we this chapter is processed by WatIE,
  53                 # it will download the info again
  54                 chapter_info = self.download_video_info(chapter_id)
  55                 chapter_urls.append(chapter_info['url'])
  56             entries = [self.url_result(chapter_url) for chapter_url in chapter_urls]
  57             return self.playlist_result(entries, real_id, video_info['title'])
  58
  59         # Otherwise we can continue and extract just one part, we have to use
  60         # the short id for getting the video url
  61         player_data = compat_urllib_parse.urlencode({'shortVideoId': short_id,
  62                                                      'html5': '1'})
  63         player_info = self._download_webpage('http://www.wat.tv/player?' + player_data,
  64                                              real_id, u'Downloading player info')
  65         player = json.loads(player_info)['player']
  66         html5_player = self._html_search_regex(r'iframe src="(.*?)"', player,
  67                                                'html5 player')
  68         player_webpage = self._download_webpage(html5_player, real_id,
  69                                                 u'Downloading player webpage')
  70
  71         video_url = self._search_regex(r'urlhtml5 : "(.*?)"', player_webpage,
  72                                        'video url')
  73         info = {'id': real_id,
  74                 'url': video_url,
  75                 'ext': 'mp4',
  76                 'title': first_chapter['title'],
  77                 'thumbnail': first_chapter['preview'],
  78                 'description': first_chapter['description'],
  79                 'view_count': video_info['views'],
  80                 }
  81         if 'date_diffusion' in first_chapter:
  82             info['upload_date'] = unified_strdate(first_chapter['date_diffusion'])
  83
  84         return info