2 from __future__ import unicode_literals, division
6 from .common import InfoExtractor
7 from ..utils import int_or_none
10 class CrackleIE(InfoExtractor):
11 _VALID_URL = r'(?:crackle:|https?://(?:www\.)?crackle\.com/(?:playlist/\d+/|(?:[^/]+/)+))(?P<id>\d+)'
13 'url': 'http://www.crackle.com/comedians-in-cars-getting-coffee/2498934',
17 'title': 'Everybody Respects A Bloody Nose',
18 '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.',
19 'thumbnail': 're:^https?://.*\.jpg',
21 'series': 'Comedians In Cars Getting Coffee',
32 'skip_download': True,
36 # extracted from http://legacyweb-us.crackle.com/flash/ReferrerRedirect.ashx
56 def _real_extract(self, url):
57 video_id = self._match_id(url)
59 config_doc = self._download_xml(
60 'http://legacyweb-us.crackle.com/flash/QueryReferrer.ashx?site=16',
61 video_id, 'Downloading config')
63 item = self._download_xml(
64 'http://legacyweb-us.crackle.com/app/revamp/vidwallcache.aspx?flags=-1&fm=%s' % video_id,
66 title = item.attrib['t']
69 formats = self._extract_m3u8_formats(
70 'http://content.uplynk.com/ext/%s/%s.m3u8' % (config_doc.attrib['strUplynkOwnerId'], video_id),
71 video_id, 'mp4', m3u8_id='hls', fatal=None)
72 path = item.attrib.get('p')
74 http_base_url = 'http://ahttp.crackle.com/' + path
75 for mfs_path, mfs_info in self._MEDIA_FILE_SLOTS.items():
77 'url': http_base_url + mfs_path,
78 'format_id': 'http-' + mfs_path.split('.')[0],
79 'width': mfs_info['width'],
80 'height': mfs_info['height'],
82 for cc in item.findall('cc'):
83 locale = cc.attrib.get('l')
84 v = cc.attrib.get('v')
86 if locale not in subtitles:
87 subtitles[locale] = []
88 subtitles[locale] = [{
89 'url': '%s/%s%s_%s.xml' % (config_doc.attrib['strSubtitleServer'], path, locale, v),
92 self._sort_formats(formats, ('width', 'height', 'tbr', 'format_id'))
94 media_details = self._download_json(
95 'https://web-api-us.crackle.com/Service.svc/details/media/%s/TW?format=json' % video_id,
96 video_id, fatal=False)
99 for key, value in media_details.items():
100 mobj = re.match('^Thumbnail_(\d+)x(\d+)$', key)
102 width, height = list(map(int, mobj.groups()))
104 'id': '%dp' % height,
113 'description': item.attrib.get('d'),
114 'duration': int(item.attrib.get('r'), 16) / 1000 if item.attrib.get('r') else None,
115 'series': item.attrib.get('sn'),
116 'season_number': int_or_none(item.attrib.get('se')),
117 'episode_number': int_or_none(item.attrib.get('ep')),
118 'thumbnails': thumbnails,
119 'subtitles': subtitles,