_ 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     month_by_name,
  10     int_or_none,
  11 )
  12
  13 class ScreenwaveMediaIE(InfoExtractor):
  14     _VALID_URL = r'(?:http://)?(?' \
  15         r':(?P<generic>player\.screenwavemedia\.com/play/[a-zA-Z]+\.php\?[^"]*\bid=(?P<video_id>.+))' \
  16         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>[^?#/]+))' \
  17         r'|(?P<teamfourstar>(?:www\.)?teamfourstar\.com/video/(?P<tfs_display_id>[a-z0-9\-]+)/?)' \
  18         r')'
  19
  20     _TESTS = [
  21         {
  22             'url': 'http://cinemassacre.com/2012/11/10/avgn-the-movie-trailer/',
  23             'md5': 'fde81fbafaee331785f58cd6c0d46190',
  24             'info_dict': {
  25                 'id': 'Cinemasssacre-19911',
  26                 'ext': 'mp4',
  27                 'upload_date': '20121110',
  28                 'title': '“Angry Video Game Nerd: The Movie” – Trailer',
  29                 'description': 'md5:fb87405fcb42a331742a0dce2708560b',
  30             },
  31         },
  32         {
  33             'url': 'http://cinemassacre.com/2013/10/02/the-mummys-hand-1940',
  34             'md5': 'd72f10cd39eac4215048f62ab477a511',
  35             'info_dict': {
  36                 'id': 'Cinemasssacre-521be8ef82b16',
  37                 'ext': 'mp4',
  38                 'upload_date': '20131002',
  39                 'title': 'The Mummy’s Hand (1940)',
  40             },
  41         }
  42     ]
  43
  44     def _cinemassacre_get_info(self, url):
  45         mobj = re.match(self._VALID_URL, url)
  46         display_id = mobj.group('cm_display_id')
  47
  48         webpage = self._download_webpage(url, display_id)
  49         video_date = mobj.group('cm_date_Y') + mobj.group('cm_date_m') + mobj.group('cm_date_d')
  50         mobj = re.search(r'src="(?P<embed_url>http://player\.screenwavemedia\.com/play/[a-zA-Z]+\.php\?[^"]*\bid=.+?)"', webpage)
  51         if not mobj:
  52             raise ExtractorError('Can\'t extract embed url and video id')
  53         playerdata_url = mobj.group('embed_url')
  54
  55         video_title = self._html_search_regex(
  56             r'<title>(?P<title>.+?)\|', webpage, 'title')
  57         video_description = self._html_search_regex(
  58             r'<div class="entry-content">(?P<description>.+?)</div>',
  59             webpage, 'description', flags=re.DOTALL, fatal=False)
  60         video_thumbnail = self._og_search_thumbnail(webpage)
  61
  62         return {
  63             'title': video_title,
  64             'description': video_description,
  65             'upload_date': video_date,
  66             'thumbnail': video_thumbnail,
  67             '_embed_url': playerdata_url,
  68         }
  69
  70     def _teamfourstar_get_info(self, url):
  71         mobj = re.match(self._VALID_URL, url)
  72         display_id = mobj.group('tfs_display_id')
  73         webpage = self._download_webpage(url, display_id)
  74
  75         mobj = re.search(r'src="(?P<embed_url>http://player\.screenwavemedia\.com/play/[a-zA-Z]+\.php\?[^"]*\bid=.+?)"', webpage)
  76         if not mobj:
  77             raise ExtractorError('Can\'t extract embed url and video id')
  78         playerdata_url = mobj.group('embed_url')
  79
  80         video_title = self._html_search_regex(
  81             r'<div class="heroheadingtitle">(?P<title>.+?)</div>', webpage, 'title')
  82         video_date = self._html_search_regex(
  83             r'<div class="heroheadingdate">(?P<date>.+?)</div>', webpage, 'date')
  84         mobj = re.match('(?P<month>[A-Z][a-z]+) (?P<day>\d+), (?P<year>\d+)', video_date)
  85         video_date = '%04u%02u%02u' % (int(mobj.group('year')), month_by_name(mobj.group('month')), int(mobj.group('day')))
  86         video_description = self._html_search_regex(
  87             r'<div class="postcontent">(?P<description>.+?)</div>', webpage, 'description', flags=re.DOTALL)
  88         video_thumbnail = self._og_search_thumbnail(webpage)
  89
  90         return {
  91             'title': video_title,
  92             'description': video_description,
  93             'upload_date': video_date,
  94             'thumbnail': video_thumbnail,
  95             '_embed_url': playerdata_url,
  96         }
  97
  98     def _screenwavemedia_get_info(self, url):
  99         mobj = re.match(self._VALID_URL, url)
 100         if not mobj:
 101             raise ExtractorError('Can\'t extract embed url and video id')
 102         video_id = mobj.group('video_id')
 103
 104         playerdata = self._download_webpage(url, video_id, 'Downloading player webpage')
 105
 106         vidtitle = self._search_regex(
 107             r'\'vidtitle\'\s*:\s*"([^\']+)"', playerdata, 'vidtitle').replace('\\/', '/')
 108         vidurl = self._search_regex(
 109             r'\'vidurl\'\s*:\s*"([^\']+)"', playerdata, 'vidurl').replace('\\/', '/')
 110         pageurl = self._search_regex(
 111             r'\'pageurl\'\s*:\s*"([^\']+)"', playerdata, 'pageurl', fatal=False).replace('\\/', '/')
 112
 113         videolist_url = None
 114
 115         mobj = re.search(r"'videoserver'\s*:\s*'(?P<videoserver>[^']+)'", playerdata)
 116         if mobj:
 117             videoserver = mobj.group('videoserver')
 118             mobj = re.search(r'\'vidid\'\s*:\s*"(?P<vidid>[^\']+)"', playerdata)
 119             vidid = mobj.group('vidid') if mobj else video_id
 120             videolist_url = 'http://%s/vod/smil:%s.smil/jwplayer.smil' % (videoserver, vidid)
 121         else:
 122             mobj = re.search(r"file\s*:\s*'(?P<smil>http.+?/jwplayer\.smil)'", playerdata)
 123             if mobj:
 124                 videolist_url = mobj.group('smil')
 125
 126         if videolist_url:
 127             videolist = self._download_xml(videolist_url, video_id, 'Downloading videolist XML')
 128             formats = []
 129             baseurl = vidurl[:vidurl.rfind('/') + 1]
 130             for video in videolist.findall('.//video'):
 131                 src = video.get('src')
 132                 if not src:
 133                     continue
 134                 file_ = src.partition(':')[-1]
 135                 width = int_or_none(video.get('width'))
 136                 height = int_or_none(video.get('height'))
 137                 bitrate = int_or_none(video.get('system-bitrate'))
 138                 format = {
 139                     'url': baseurl + file_,
 140                     'format_id': src.rpartition('.')[0].rpartition('_')[-1],
 141                 }
 142                 if width or height:
 143                     format.update({
 144                         'tbr': bitrate // 1000 if bitrate else None,
 145                         'width': width,
 146                         'height': height,
 147                     })
 148                 else:
 149                     format.update({
 150                         'abr': bitrate // 1000 if bitrate else None,
 151                         'vcodec': 'none',
 152                     })
 153                 formats.append(format)
 154             self._sort_formats(formats)
 155         else:
 156             formats = [{
 157                 'url': vidurl,
 158             }]
 159
 160         return {
 161             'id': video_id,
 162             'title': vidtitle,
 163             'formats': formats,
 164             '_episode_page': pageurl,
 165         }
 166
 167     def _real_extract(self, url):
 168         mobj = re.match(self._VALID_URL, url)
 169
 170         swm_info = None
 171         site_info = None
 172
 173         if mobj.group('generic'):
 174             swm_info = self._screenwavemedia_get_info(url)
 175             url = swm_info['_episode_page']
 176             mobj = re.match(self._VALID_URL, url)
 177
 178         if mobj:
 179             if mobj.group('cinemassacre'):
 180                 site_info = self._cinemassacre_get_info(url)
 181             elif mobj.group('teamfourstar'):
 182                 site_info = self._teamfourstar_get_info(url)
 183
 184         if not swm_info:
 185             if site_info:
 186                 swm_info = self._screenwavemedia_get_info(site_info['_embed_url'])
 187
 188         if not swm_info:
 189             raise ExtractorError("Failed to extract metadata for this URL")
 190
 191         if site_info:
 192             swm_info.update(site_info)
 193
 194         return swm_info