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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import int_or_none
   6
   7
   8 class CrackleIE(InfoExtractor):
   9     _VALID_URL = r'(?:crackle:|https?://(?:www\.)?crackle\.com/(?:playlist/\d+/|(?:[^/]+/)+))(?P<id>\d+)'
  10     _TEST = {
  11         'url': 'http://www.crackle.com/the-art-of-more/2496419',
  12         'info_dict': {
  13             'id': '2496419',
  14             'ext': 'mp4',
  15             'title': 'Heavy Lies the Head',
  16             'description': 'md5:bb56aa0708fe7b9a4861535f15c3abca',
  17         },
  18         'params': {
  19             # m3u8 download
  20             'skip_download': True,
  21         }
  22     }
  23
  24     # extracted from http://legacyweb-us.crackle.com/flash/QueryReferrer.ashx
  25     _SUBTITLE_SERVER = 'http://web-us-az.crackle.com'
  26     _UPLYNK_OWNER_ID = 'e8773f7770a44dbd886eee4fca16a66b'
  27     _THUMBNAIL_TEMPLATE = 'http://images-us-am.crackle.com/%stnl_1920x1080.jpg?ts=20140107233116?c=635333335057637614'
  28
  29     # extracted from http://legacyweb-us.crackle.com/flash/ReferrerRedirect.ashx
  30     _MEDIA_FILE_SLOTS = {
  31         'c544.flv': {
  32             'width': 544,
  33             'height': 306,
  34         },
  35         '360p.mp4': {
  36             'width': 640,
  37             'height': 360,
  38         },
  39         '480p.mp4': {
  40             'width': 852,
  41             'height': 478,
  42         },
  43         '480p_1mbps.mp4': {
  44             'width': 852,
  45             'height': 478,
  46         },
  47     }
  48
  49     def _real_extract(self, url):
  50         video_id = self._match_id(url)
  51         item = self._download_xml(
  52             'http://legacyweb-us.crackle.com/app/revamp/vidwallcache.aspx?flags=-1&fm=%s' % video_id,
  53             video_id).find('i')
  54         title = item.attrib['t']
  55
  56         thumbnail = None
  57         subtitles = {}
  58         formats = self._extract_m3u8_formats(
  59             'http://content.uplynk.com/ext/%s/%s.m3u8' % (self._UPLYNK_OWNER_ID, video_id),
  60             video_id, 'mp4', m3u8_id='hls', fatal=None)
  61         path = item.attrib.get('p')
  62         if path:
  63             thumbnail = self._THUMBNAIL_TEMPLATE % path
  64             http_base_url = 'http://ahttp.crackle.com/' + path
  65             for mfs_path, mfs_info in self._MEDIA_FILE_SLOTS.items():
  66                 formats.append({
  67                     'url': http_base_url + mfs_path,
  68                     'format_id': 'http-' + mfs_path.split('.')[0],
  69                     'width': mfs_info['width'],
  70                     'height': mfs_info['height'],
  71                 })
  72             for cc in item.findall('cc'):
  73                 locale = cc.attrib.get('l')
  74                 v = cc.attrib.get('v')
  75                 if locale and v:
  76                     if locale not in subtitles:
  77                         subtitles[locale] = []
  78                     subtitles[locale] = [{
  79                         'url': '%s/%s%s_%s.xml' % (self._SUBTITLE_SERVER, path, locale, v),
  80                         'ext': 'ttml',
  81                     }]
  82         self._sort_formats(formats, ('width', 'height', 'tbr', 'format_id'))
  83
  84         return {
  85             'id': video_id,
  86             'title': title,
  87             'description': item.attrib.get('d'),
  88             'duration': int(item.attrib.get('r'), 16) if item.attrib.get('r') else None,
  89             'series': item.attrib.get('sn'),
  90             'season_number': int_or_none(item.attrib.get('se')),
  91             'episode_number': int_or_none(item.attrib.get('ep')),
  92             'thumbnail': thumbnail,
  93             'subtitles': subtitles,
  94             'formats': formats,
  95         }