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|m)\.)?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': r'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     _THUMBNAIL_RES = [
  35         (120, 90),
  36         (208, 156),
  37         (220, 124),
  38         (220, 220),
  39         (240, 180),
  40         (250, 141),
  41         (315, 236),
  42         (320, 180),
  43         (360, 203),
  44         (400, 300),
  45         (421, 316),
  46         (460, 330),
  47         (460, 460),
  48         (462, 260),
  49         (480, 270),
  50         (587, 330),
  51         (640, 480),
  52         (700, 330),
  53         (700, 394),
  54         (854, 480),
  55         (1024, 1024),
  56         (1920, 1080),
  57     ]
  58
  59     # extracted from http://legacyweb-us.crackle.com/flash/ReferrerRedirect.ashx
  60     _MEDIA_FILE_SLOTS = {
  61         'c544.flv': {
  62             'width': 544,
  63             'height': 306,
  64         },
  65         '360p.mp4': {
  66             'width': 640,
  67             'height': 360,
  68         },
  69         '480p.mp4': {
  70             'width': 852,
  71             'height': 478,
  72         },
  73         '480p_1mbps.mp4': {
  74             'width': 852,
  75             'height': 478,
  76         },
  77     }
  78
  79     def _real_extract(self, url):
  80         video_id = self._match_id(url)
  81
  82         config_doc = self._download_xml(
  83             'http://legacyweb-us.crackle.com/flash/QueryReferrer.ashx?site=16',
  84             video_id, 'Downloading config')
  85
  86         item = self._download_xml(
  87             'http://legacyweb-us.crackle.com/app/revamp/vidwallcache.aspx?flags=-1&fm=%s' % video_id,
  88             video_id, headers=self.geo_verification_headers()).find('i')
  89         title = item.attrib['t']
  90
  91         subtitles = {}
  92         formats = self._extract_m3u8_formats(
  93             'http://content.uplynk.com/ext/%s/%s.m3u8' % (config_doc.attrib['strUplynkOwnerId'], video_id),
  94             video_id, 'mp4', m3u8_id='hls', fatal=None)
  95         thumbnails = []
  96         path = item.attrib.get('p')
  97         if path:
  98             for width, height in self._THUMBNAIL_RES:
  99                 res = '%dx%d' % (width, height)
 100                 thumbnails.append({
 101                     'id': res,
 102                     'url': 'http://images-us-am.crackle.com/%stnl_%s.jpg' % (path, res),
 103                     'width': width,
 104                     'height': height,
 105                     'resolution': res,
 106                 })
 107             http_base_url = 'http://ahttp.crackle.com/' + path
 108             for mfs_path, mfs_info in self._MEDIA_FILE_SLOTS.items():
 109                 formats.append({
 110                     'url': http_base_url + mfs_path,
 111                     'format_id': 'http-' + mfs_path.split('.')[0],
 112                     'width': mfs_info['width'],
 113                     'height': mfs_info['height'],
 114                 })
 115             for cc in item.findall('cc'):
 116                 locale = cc.attrib.get('l')
 117                 v = cc.attrib.get('v')
 118                 if locale and v:
 119                     if locale not in subtitles:
 120                         subtitles[locale] = []
 121                     for url_ext, ext in (('vtt', 'vtt'), ('xml', 'tt')):
 122                         subtitles.setdefault(locale, []).append({
 123                             'url': '%s/%s%s_%s.%s' % (config_doc.attrib['strSubtitleServer'], path, locale, v, url_ext),
 124                             'ext': ext,
 125                         })
 126         self._sort_formats(formats, ('width', 'height', 'tbr', 'format_id'))
 127
 128         return {
 129             'id': video_id,
 130             'title': title,
 131             'description': item.attrib.get('d'),
 132             'duration': int(item.attrib.get('r'), 16) / 1000 if item.attrib.get('r') else None,
 133             'series': item.attrib.get('sn'),
 134             'season_number': int_or_none(item.attrib.get('se')),
 135             'episode_number': int_or_none(item.attrib.get('ep')),
 136             'thumbnails': thumbnails,
 137             'subtitles': subtitles,
 138             'formats': formats,
 139         }