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\?id=(?: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
  57         playerdata = self._download_webpage(playerdata_url, video_id, 'Downloading player webpage')
  58         video_thumbnail = self._search_regex(
  59             r'image: \'(?P<thumbnail>[^\']+)\'', playerdata, 'thumbnail', fatal=False)
  60         sd_url = self._search_regex(r'file: \'([^\']+)\', label: \'SD\'', playerdata, 'sd_file')
  61         videolist_url = self._search_regex(r'file: \'([^\']+\.smil)\'}', playerdata, 'videolist_url')
  62
  63         videolist = self._download_xml(videolist_url, video_id, 'Downloading videolist XML')
  64
  65         formats = []
  66         baseurl = sd_url[:sd_url.rfind('/')+1]
  67         for video in videolist.findall('.//video'):
  68             src = video.get('src')
  69             if not src:
  70                 continue
  71             file_ = src.partition(':')[-1]
  72             width = int_or_none(video.get('width'))
  73             height = int_or_none(video.get('height'))
  74             bitrate = int_or_none(video.get('system-bitrate'))
  75             format = {
  76                 'url': baseurl + file_,
  77                 'format_id': src.rpartition('.')[0].rpartition('_')[-1],
  78             }
  79             if width or height:
  80                 format.update({
  81                     'tbr': bitrate // 1000 if bitrate else None,
  82                     'width': width,
  83                     'height': height,
  84                 })
  85             else:
  86                 format.update({
  87                     'abr': bitrate // 1000 if bitrate else None,
  88                     'vcodec': 'none',
  89                 })
  90             formats.append(format)
  91         self._sort_formats(formats)
  92
  93         return {
  94             'id': video_id,
  95             'title': video_title,
  96             'formats': formats,
  97             'description': video_description,
  98             'upload_date': video_date,
  99             'thumbnail': video_thumbnail,
 100         }