2 from __future__
import unicode_literals
, division
4 from .common
import InfoExtractor
5 from ..utils
import int_or_none
8 class CrackleIE(InfoExtractor
):
9 _GEO_COUNTRIES
= ['US']
10 _VALID_URL
= r
'(?:crackle:|https?://(?:(?:www|m)\.)?crackle\.com/(?:playlist/\d+/|(?:[^/]+/)+))(?P<id>\d+)'
12 'url': 'http://www.crackle.com/comedians-in-cars-getting-coffee/2498934',
16 'title': 'Everybody Respects A Bloody Nose',
17 '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.',
18 'thumbnail': r
're:^https?://.*\.jpg',
20 'series': 'Comedians In Cars Getting Coffee',
32 'skip_download': True,
61 # extracted from http://legacyweb-us.crackle.com/flash/ReferrerRedirect.ashx
81 def _real_extract(self
, url
):
82 video_id
= self
._match
_id
(url
)
84 config_doc
= self
._download
_xml
(
85 'http://legacyweb-us.crackle.com/flash/QueryReferrer.ashx?site=16',
86 video_id
, 'Downloading config')
88 item
= self
._download
_xml
(
89 'http://legacyweb-us.crackle.com/app/revamp/vidwallcache.aspx?flags=-1&fm=%s' % video_id
,
90 video_id
, headers
=self
.geo_verification_headers()).find('i')
91 title
= item
.attrib
['t']
94 formats
= self
._extract
_m
3u8_formats
(
95 'http://content.uplynk.com/ext/%s/%s.m3u8' % (config_doc
.attrib
['strUplynkOwnerId'], video_id
),
96 video_id
, 'mp4', m3u8_id
='hls', fatal
=None)
98 path
= item
.attrib
.get('p')
100 for width
, height
in self
._THUMBNAIL
_RES
:
101 res
= '%dx%d' % (width
, height
)
104 'url': 'http://images-us-am.crackle.com/%stnl_%s.jpg' % (path
, res
),
109 http_base_url
= 'http://ahttp.crackle.com/' + path
110 for mfs_path
, mfs_info
in self
._MEDIA
_FILE
_SLOTS
.items():
112 'url': http_base_url
+ mfs_path
,
113 'format_id': 'http-' + mfs_path
.split('.')[0],
114 'width': mfs_info
['width'],
115 'height': mfs_info
['height'],
117 for cc
in item
.findall('cc'):
118 locale
= cc
.attrib
.get('l')
119 v
= cc
.attrib
.get('v')
121 if locale
not in subtitles
:
122 subtitles
[locale
] = []
123 for url_ext
, ext
in (('vtt', 'vtt'), ('xml', 'tt')):
124 subtitles
.setdefault(locale
, []).append({
125 'url': '%s/%s%s_%s.%s' % (config_doc
.attrib
['strSubtitleServer'], path
, locale
, v
, url_ext
),
128 self
._sort
_formats
(formats
, ('width', 'height', 'tbr', 'format_id'))
133 'description': item
.attrib
.get('d'),
134 'duration': int(item
.attrib
.get('r'), 16) / 1000 if item
.attrib
.get('r') else None,
135 'series': item
.attrib
.get('sn'),
136 'season_number': int_or_none(item
.attrib
.get('se')),
137 'episode_number': int_or_none(item
.attrib
.get('ep')),
138 'thumbnails': thumbnails
,
139 'subtitles': subtitles
,