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

   1 # coding: utf-8
   2 from __future__ import unicode_literals, division
   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/comedians-in-cars-getting-coffee/2498934',
  12         'info_dict': {
  13             'id': '2498934',
  14             'ext': 'mp4',
  15             'title': 'Everybody Respects A Bloody Nose',
  16             'description': 'Jerry is kaffeeklatsching in L.A. with funnyman J.B. Smoove (Saturday Night Live, Real Husbands of Hollywood). They’re headed for brew at 10 Speed Coffee in a 1964 Studebaker Avanti.',
  17             'thumbnail': 're:^https?://.*\.jpg',
  18             'duration': 906,
  19             'series': 'Comedians In Cars Getting Coffee',
  20             'season_number': 8,
  21             'episode_number': 4,
  22             'subtitles': {
  23                 'en-US': [{
  24                     'ext': 'ttml',
  25                 }]
  26             },
  27         },
  28         'params': {
  29             # m3u8 download
  30             'skip_download': True,
  31         }
  32     }
  33
  34     # extracted from http://legacyweb-us.crackle.com/flash/ReferrerRedirect.ashx
  35     _THUMBNAIL_TEMPLATE = 'http://images-us-am.crackle.com/%stnl_1920x1080.jpg?ts=20140107233116?c=635333335057637614'
  36     _MEDIA_FILE_SLOTS = {
  37         'c544.flv': {
  38             'width': 544,
  39             'height': 306,
  40         },
  41         '360p.mp4': {
  42             'width': 640,
  43             'height': 360,
  44         },
  45         '480p.mp4': {
  46             'width': 852,
  47             'height': 478,
  48         },
  49         '480p_1mbps.mp4': {
  50             'width': 852,
  51             'height': 478,
  52         },
  53     }
  54
  55     def _real_extract(self, url):
  56         video_id = self._match_id(url)
  57
  58         config_doc = self._download_xml(
  59             'http://legacyweb-us.crackle.com/flash/QueryReferrer.ashx?site=16',
  60             video_id, 'Downloading config')
  61
  62         item = self._download_xml(
  63             'http://legacyweb-us.crackle.com/app/revamp/vidwallcache.aspx?flags=-1&fm=%s' % video_id,
  64             video_id).find('i')
  65         title = item.attrib['t']
  66
  67         subtitles = {}
  68         formats = self._extract_m3u8_formats(
  69             'http://content.uplynk.com/ext/%s/%s.m3u8' % (config_doc.attrib['strUplynkOwnerId'], video_id),
  70             video_id, 'mp4', m3u8_id='hls', fatal=None)
  71         thumbnail = None
  72         path = item.attrib.get('p')
  73         if path:
  74             thumbnail = self._THUMBNAIL_TEMPLATE % path
  75             http_base_url = 'http://ahttp.crackle.com/' + path
  76             for mfs_path, mfs_info in self._MEDIA_FILE_SLOTS.items():
  77                 formats.append({
  78                     'url': http_base_url + mfs_path,
  79                     'format_id': 'http-' + mfs_path.split('.')[0],
  80                     'width': mfs_info['width'],
  81                     'height': mfs_info['height'],
  82                 })
  83             for cc in item.findall('cc'):
  84                 locale = cc.attrib.get('l')
  85                 v = cc.attrib.get('v')
  86                 if locale and v:
  87                     if locale not in subtitles:
  88                         subtitles[locale] = []
  89                     subtitles[locale] = [{
  90                         'url': '%s/%s%s_%s.xml' % (config_doc.attrib['strSubtitleServer'], path, locale, v),
  91                         'ext': 'ttml',
  92                     }]
  93         self._sort_formats(formats, ('width', 'height', 'tbr', 'format_id'))
  94
  95         return {
  96             'id': video_id,
  97             'title': title,
  98             'description': item.attrib.get('d'),
  99             'duration': int(item.attrib.get('r'), 16) / 1000 if item.attrib.get('r') else None,
 100             'series': item.attrib.get('sn'),
 101             'season_number': int_or_none(item.attrib.get('se')),
 102             'episode_number': int_or_none(item.attrib.get('ep')),
 103             'thumbnail': thumbnail,
 104             'subtitles': subtitles,
 105             'formats': formats,
 106         }