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

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     ExtractorError,
   9     int_or_none,
  10 )
  11
  12
  13 class CinemassacreIE(InfoExtractor):
  14     _VALID_URL = r'http://(?:www\.)?cinemassacre\.com/(?P<date_Y>[0-9]{4})/(?P<date_m>[0-9]{2})/(?P<date_d>[0-9]{2})/(?P<display_id>[^?#/]+)'
  15     _TESTS = [
  16         {
  17             'url': 'http://cinemassacre.com/2012/11/10/avgn-the-movie-trailer/',
  18             'md5': 'fde81fbafaee331785f58cd6c0d46190',
  19             'info_dict': {
  20                 'id': '19911',
  21                 'ext': 'mp4',
  22                 'upload_date': '20121110',
  23                 'title': '“Angry Video Game Nerd: The Movie” – Trailer',
  24                 'description': 'md5:fb87405fcb42a331742a0dce2708560b',
  25             },
  26         },
  27         {
  28             'url': 'http://cinemassacre.com/2013/10/02/the-mummys-hand-1940',
  29             'md5': 'd72f10cd39eac4215048f62ab477a511',
  30             'info_dict': {
  31                 'id': '521be8ef82b16',
  32                 'ext': 'mp4',
  33                 'upload_date': '20131002',
  34                 'title': 'The Mummy’s Hand (1940)',
  35             },
  36         }
  37     ]
  38
  39     def _real_extract(self, url):
  40         mobj = re.match(self._VALID_URL, url)
  41         display_id = mobj.group('display_id')
  42
  43         webpage = self._download_webpage(url, display_id)
  44         video_date = mobj.group('date_Y') + mobj.group('date_m') + mobj.group('date_d')
  45         mobj = re.search(r'src="(?P<embed_url>http://player\.screenwavemedia\.com/play/[a-zA-Z]+\.php\?[^"]*\bid=(?:Cinemassacre-)?(?P<video_id>.+?))"', webpage)
  46         if not mobj:
  47             raise ExtractorError('Can\'t extract embed url and video id')
  48         playerdata_url = mobj.group('embed_url')
  49         video_id = mobj.group('video_id')
  50
  51         video_title = self._html_search_regex(
  52             r'<title>(?P<title>.+?)\|', webpage, 'title')
  53         video_description = self._html_search_regex(
  54             r'<div class="entry-content">(?P<description>.+?)</div>',
  55             webpage, 'description', flags=re.DOTALL, fatal=False)
  56         video_thumbnail = self._og_search_thumbnail(webpage)
  57
  58         playerdata = self._download_webpage(playerdata_url, video_id, 'Downloading player webpage')
  59
  60         vidurl = self._search_regex(
  61             r'\'vidurl\'\s*:\s*"([^\']+)"', playerdata, 'vidurl').replace('\\/', '/')
  62         vidid = self._search_regex(
  63             r'\'vidid\'\s*:\s*"([^\']+)"', playerdata, 'vidid')
  64         videoserver = self._html_search_regex(
  65             r"'videoserver'\s*:\s*'([^']+)'", playerdata, 'videoserver')
  66
  67         videolist_url = 'http://%s/vod/smil:%s.smil/jwplayer.smil' % (videoserver, vidid)
  68         videolist = self._download_xml(videolist_url, video_id, 'Downloading videolist XML')
  69
  70         formats = []
  71         baseurl = vidurl[:vidurl.rfind('/')+1]
  72         for video in videolist.findall('.//video'):
  73             src = video.get('src')
  74             if not src:
  75                 continue
  76             file_ = src.partition(':')[-1]
  77             width = int_or_none(video.get('width'))
  78             height = int_or_none(video.get('height'))
  79             bitrate = int_or_none(video.get('system-bitrate'))
  80             format = {
  81                 'url': baseurl + file_,
  82                 'format_id': src.rpartition('.')[0].rpartition('_')[-1],
  83             }
  84             if width or height:
  85                 format.update({
  86                     'tbr': bitrate // 1000 if bitrate else None,
  87                     'width': width,
  88                     'height': height,
  89                 })
  90             else:
  91                 format.update({
  92                     'abr': bitrate // 1000 if bitrate else None,
  93                     'vcodec': 'none',
  94                 })
  95             formats.append(format)
  96         self._sort_formats(formats)
  97
  98         return {
  99             'id': video_id,
 100             'title': video_title,
 101             'formats': formats,
 102             'description': video_description,
 103             'upload_date': video_date,
 104             'thumbnail': video_thumbnail,
 105         }