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     _GEO_COUNTRIES = ['US']
  10     _VALID_URL = r'(?:crackle:|https?://(?:(?:www|m)\.)?crackle\.com/(?:playlist/\d+/|(?:[^/]+/)+))(?P<id>\d+)'
  11     _TEST = {
  12         'url': 'http://www.crackle.com/comedians-in-cars-getting-coffee/2498934',
  13         'info_dict': {
  14             'id': '2498934',
  15             'ext': 'mp4',
  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',
  19             'duration': 906,
  20             'series': 'Comedians In Cars Getting Coffee',
  21             'season_number': 8,
  22             'episode_number': 4,
  23             'subtitles': {
  24                 'en-US': [{
  25                     'ext': 'ttml',
  26                 }]
  27             },
  28         },
  29         'params': {
  30             # m3u8 download
  31             'skip_download': True,
  32         }
  33     }
  34
  35     _THUMBNAIL_RES = [
  36         (120, 90),
  37         (208, 156),
  38         (220, 124),
  39         (220, 220),
  40         (240, 180),
  41         (250, 141),
  42         (315, 236),
  43         (320, 180),
  44         (360, 203),
  45         (400, 300),
  46         (421, 316),
  47         (460, 330),
  48         (460, 460),
  49         (462, 260),
  50         (480, 270),
  51         (587, 330),
  52         (640, 480),
  53         (700, 330),
  54         (700, 394),
  55         (854, 480),
  56         (1024, 1024),
  57         (1920, 1080),
  58     ]
  59
  60     # extracted from http://legacyweb-us.crackle.com/flash/ReferrerRedirect.ashx
  61     _MEDIA_FILE_SLOTS = {
  62         'c544.flv': {
  63             'width': 544,
  64             'height': 306,
  65         },
  66         '360p.mp4': {
  67             'width': 640,
  68             'height': 360,
  69         },
  70         '480p.mp4': {
  71             'width': 852,
  72             'height': 478,
  73         },
  74         '480p_1mbps.mp4': {
  75             'width': 852,
  76             'height': 478,
  77         },
  78     }
  79
  80     def _real_extract(self, url):
  81         video_id = self._match_id(url)
  82
  83         config_doc = self._download_xml(
  84             'http://legacyweb-us.crackle.com/flash/QueryReferrer.ashx?site=16',
  85             video_id, 'Downloading config')
  86
  87         item = self._download_xml(
  88             'http://legacyweb-us.crackle.com/app/revamp/vidwallcache.aspx?flags=-1&fm=%s' % video_id,
  89             video_id, headers=self.geo_verification_headers()).find('i')
  90         title = item.attrib['t']
  91
  92         subtitles = {}
  93         formats = self._extract_m3u8_formats(
  94             'http://content.uplynk.com/ext/%s/%s.m3u8' % (config_doc.attrib['strUplynkOwnerId'], video_id),
  95             video_id, 'mp4', m3u8_id='hls', fatal=None)
  96         thumbnails = []
  97         path = item.attrib.get('p')
  98         if path:
  99             for width, height in self._THUMBNAIL_RES:
 100                 res = '%dx%d' % (width, height)
 101                 thumbnails.append({
 102                     'id': res,
 103                     'url': 'http://images-us-am.crackle.com/%stnl_%s.jpg' % (path, res),
 104                     'width': width,
 105                     'height': height,
 106                     'resolution': res,
 107                 })
 108             http_base_url = 'http://ahttp.crackle.com/' + path
 109             for mfs_path, mfs_info in self._MEDIA_FILE_SLOTS.items():
 110                 formats.append({
 111                     'url': http_base_url + mfs_path,
 112                     'format_id': 'http-' + mfs_path.split('.')[0],
 113                     'width': mfs_info['width'],
 114                     'height': mfs_info['height'],
 115                 })
 116             for cc in item.findall('cc'):
 117                 locale = cc.attrib.get('l')
 118                 v = cc.attrib.get('v')
 119                 if locale and v:
 120                     if locale not in subtitles:
 121                         subtitles[locale] = []
 122                     for url_ext, ext in (('vtt', 'vtt'), ('xml', 'tt')):
 123                         subtitles.setdefault(locale, []).append({
 124                             'url': '%s/%s%s_%s.%s' % (config_doc.attrib['strSubtitleServer'], path, locale, v, url_ext),
 125                             'ext': ext,
 126                         })
 127         self._sort_formats(formats, ('width', 'height', 'tbr', 'format_id'))
 128
 129         return {
 130             'id': video_id,
 131             'title': title,
 132             'description': item.attrib.get('d'),
 133             'duration': int(item.attrib.get('r'), 16) / 1000 if item.attrib.get('r') else None,
 134             'series': item.attrib.get('sn'),
 135             'season_number': int_or_none(item.attrib.get('se')),
 136             'episode_number': int_or_none(item.attrib.get('ep')),
 137             'thumbnails': thumbnails,
 138             'subtitles': subtitles,
 139             'formats': formats,
 140         }