_ Git - youtube-dl/blob - youtube_dl/extractor/screenwavemedia.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 class ScreenwaveMediaIE(InfoExtractor):
  13     _VALID_URL = r'(?:http://)?(?' \
  14         r':(?P<generic>player\.screenwavemedia\.com/play/[a-zA-Z]+\.php\?[^"]*\bid=(?P<video_id>.+))' \
  15         r'|(?P<cinemassacre>(?:www\.)?cinemassacre\.com/(?P<cm_date_Y>[0-9]{4})/(?P<cm_date_m>[0-9]{2})/(?P<cm_date_d>[0-9]{2})/(?P<cm_display_id>[^?#/]+))' \
  16         r')'
  17
  18     _TESTS = [
  19         {
  20             'url': 'http://cinemassacre.com/2012/11/10/avgn-the-movie-trailer/',
  21             'md5': 'fde81fbafaee331785f58cd6c0d46190',
  22             'info_dict': {
  23                 'id': 'Cinemasssacre-19911',
  24                 'ext': 'mp4',
  25                 'upload_date': '20121110',
  26                 'title': '“Angry Video Game Nerd: The Movie” – Trailer',
  27                 'description': 'md5:fb87405fcb42a331742a0dce2708560b',
  28             },
  29         },
  30         {
  31             'url': 'http://cinemassacre.com/2013/10/02/the-mummys-hand-1940',
  32             'md5': 'd72f10cd39eac4215048f62ab477a511',
  33             'info_dict': {
  34                 'id': 'Cinemasssacre-521be8ef82b16',
  35                 'ext': 'mp4',
  36                 'upload_date': '20131002',
  37                 'title': 'The Mummy’s Hand (1940)',
  38             },
  39         }
  40     ]
  41
  42     def _cinemassacre_get_info(self, url):
  43         mobj = re.match(self._VALID_URL, url)
  44         display_id = mobj.group('cm_display_id')
  45
  46         webpage = self._download_webpage(url, display_id)
  47         video_date = mobj.group('cm_date_Y') + mobj.group('cm_date_m') + mobj.group('cm_date_d')
  48         mobj = re.search(r'src="(?P<embed_url>http://player\.screenwavemedia\.com/play/[a-zA-Z]+\.php\?[^"]*\bid=.+?)"', webpage)
  49         if not mobj:
  50             raise ExtractorError('Can\'t extract embed url and video id')
  51         playerdata_url = mobj.group('embed_url')
  52
  53         video_title = self._html_search_regex(
  54             r'<title>(?P<title>.+?)\|', webpage, 'title')
  55         video_description = self._html_search_regex(
  56             r'<div class="entry-content">(?P<description>.+?)</div>',
  57             webpage, 'description', flags=re.DOTALL, fatal=False)
  58         video_thumbnail = self._og_search_thumbnail(webpage)
  59
  60         return {
  61             'title': video_title,
  62             'description': video_description,
  63             'upload_date': video_date,
  64             'thumbnail': video_thumbnail,
  65             '_embed_url': playerdata_url,
  66         }
  67
  68     def _screenwavemedia_get_info(self, url):
  69         mobj = re.match(self._VALID_URL, url)
  70         if not mobj:
  71             raise ExtractorError('Can\'t extract embed url and video id')
  72         video_id = mobj.group('video_id')
  73
  74         playerdata = self._download_webpage(url, video_id, 'Downloading player webpage')
  75
  76         vidtitle = self._search_regex(
  77             r'\'vidtitle\'\s*:\s*"([^\']+)"', playerdata, 'vidtitle').replace('\\/', '/')
  78         vidurl = self._search_regex(
  79             r'\'vidurl\'\s*:\s*"([^\']+)"', playerdata, 'vidurl').replace('\\/', '/')
  80         pageurl = self._search_regex(
  81             r'\'pageurl\'\s*:\s*"([^\']+)"', playerdata, 'pageurl', fatal=False).replace('\\/', '/')
  82
  83         videolist_url = None
  84
  85         mobj = re.search(r"'videoserver'\s*:\s*'(?P<videoserver>[^']+)'", playerdata)
  86         if mobj:
  87             videoserver = mobj.group('videoserver')
  88             mobj = re.search(r'\'vidid\'\s*:\s*"(?P<vidid>[^\']+)"', playerdata)
  89             vidid = mobj.group('vidid') if mobj else video_id
  90             videolist_url = 'http://%s/vod/smil:%s.smil/jwplayer.smil' % (videoserver, vidid)
  91         else:
  92             mobj = re.search(r"file\s*:\s*'(?P<smil>http.+?/jwplayer\.smil)'", playerdata)
  93             if mobj:
  94                 videolist_url = mobj.group('smil')
  95
  96         if videolist_url:
  97             videolist = self._download_xml(videolist_url, video_id, 'Downloading videolist XML')
  98             formats = []
  99             baseurl = vidurl[:vidurl.rfind('/') + 1]
 100             for video in videolist.findall('.//video'):
 101                 src = video.get('src')
 102                 if not src:
 103                     continue
 104                 file_ = src.partition(':')[-1]
 105                 width = int_or_none(video.get('width'))
 106                 height = int_or_none(video.get('height'))
 107                 bitrate = int_or_none(video.get('system-bitrate'))
 108                 format = {
 109                     'url': baseurl + file_,
 110                     'format_id': src.rpartition('.')[0].rpartition('_')[-1],
 111                 }
 112                 if width or height:
 113                     format.update({
 114                         'tbr': bitrate // 1000 if bitrate else None,
 115                         'width': width,
 116                         'height': height,
 117                     })
 118                 else:
 119                     format.update({
 120                         'abr': bitrate // 1000 if bitrate else None,
 121                         'vcodec': 'none',
 122                     })
 123                 formats.append(format)
 124             self._sort_formats(formats)
 125         else:
 126             formats = [{
 127                 'url': vidurl,
 128             }]
 129
 130         return {
 131             'id': video_id,
 132             'title': vidtitle,
 133             'formats': formats,
 134             '_episode_page': pageurl,
 135         }
 136
 137     def _real_extract(self, url):
 138         mobj = re.match(self._VALID_URL, url)
 139
 140         swm_info = None
 141         site_info = None
 142
 143         if mobj.group('generic'):
 144             swm_info = self._screenwavemedia_get_info(url)
 145             url = swm_info['_episode_page']
 146             mobj = re.match(self._VALID_URL, url)
 147
 148         if mobj:
 149             if mobj.group('cinemassacre'):
 150                 site_info = self._cinemassacre_get_info(url)
 151
 152         if not swm_info:
 153             if site_info:
 154                 swm_info = self._screenwavemedia_get_info(site_info['_embed_url'])
 155
 156         if not swm_info:
 157             raise ExtractorError("Failed to extract metadata for this URL")
 158
 159         if site_info:
 160             swm_info.update(site_info)
 161
 162         return swm_info