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

   1 # coding: utf-8
   2 from __future__ import unicode_literals, division
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_HTTPError
   8 from ..utils import (
   9     determine_ext,
  10     float_or_none,
  11     int_or_none,
  12     parse_age_limit,
  13     parse_duration,
  14     url_or_none,
  15     ExtractorError
  16 )
  17
  18
  19 class CrackleIE(InfoExtractor):
  20     _VALID_URL = r'(?:crackle:|https?://(?:(?:www|m)\.)?(?:sony)?crackle\.com/(?:playlist/\d+/|(?:[^/]+/)+))(?P<id>\d+)'
  21     _TESTS = [{
  22         # geo restricted to CA
  23         'url': 'https://www.crackle.com/andromeda/2502343',
  24         'info_dict': {
  25             'id': '2502343',
  26             'ext': 'mp4',
  27             'title': 'Under The Night',
  28             'description': 'md5:d2b8ca816579ae8a7bf28bfff8cefc8a',
  29             'duration': 2583,
  30             'view_count': int,
  31             'average_rating': 0,
  32             'age_limit': 14,
  33             'genre': 'Action, Sci-Fi',
  34             'creator': 'Allan Kroeker',
  35             'artist': 'Keith Hamilton Cobb, Kevin Sorbo, Lisa Ryder, Lexa Doig, Robert Hewitt Wolfe',
  36             'release_year': 2000,
  37             'series': 'Andromeda',
  38             'episode': 'Under The Night',
  39             'season_number': 1,
  40             'episode_number': 1,
  41         },
  42         'params': {
  43             # m3u8 download
  44             'skip_download': True,
  45         }
  46     }, {
  47         'url': 'https://www.sonycrackle.com/andromeda/2502343',
  48         'only_matching': True,
  49     }]
  50
  51     _MEDIA_FILE_SLOTS = {
  52         '360p.mp4': {
  53             'width': 640,
  54             'height': 360,
  55         },
  56         '480p.mp4': {
  57             'width': 768,
  58             'height': 432,
  59         },
  60         '480p_1mbps.mp4': {
  61             'width': 852,
  62             'height': 480,
  63         },
  64     }
  65
  66     def _real_extract(self, url):
  67         video_id = self._match_id(url)
  68
  69         country_code = self._downloader.params.get('geo_bypass_country', None)
  70         countries = [country_code] if country_code else (
  71             'US', 'AU', 'CA', 'AS', 'FM', 'GU', 'MP', 'PR', 'PW', 'MH', 'VI')
  72
  73         last_e = None
  74
  75         for country in countries:
  76             try:
  77                 media = self._download_json(
  78                     'https://web-api-us.crackle.com/Service.svc/details/media/%s/%s'
  79                     % (video_id, country), video_id,
  80                     'Downloading media JSON as %s' % country,
  81                     'Unable to download media JSON', query={
  82                         'disableProtocols': 'true',
  83                         'format': 'json'
  84                     })
  85             except ExtractorError as e:
  86                 # 401 means geo restriction, trying next country
  87                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
  88                     last_e = e
  89                     continue
  90                 raise
  91
  92             media_urls = media.get('MediaURLs')
  93             if not media_urls or not isinstance(media_urls, list):
  94                 continue
  95
  96             title = media['Title']
  97
  98             formats = []
  99             for e in media['MediaURLs']:
 100                 if e.get('UseDRM') is True:
 101                     continue
 102                 format_url = url_or_none(e.get('Path'))
 103                 if not format_url:
 104                     continue
 105                 ext = determine_ext(format_url)
 106                 if ext == 'm3u8':
 107                     formats.extend(self._extract_m3u8_formats(
 108                         format_url, video_id, 'mp4', entry_protocol='m3u8_native',
 109                         m3u8_id='hls', fatal=False))
 110                 elif ext == 'mpd':
 111                     formats.extend(self._extract_mpd_formats(
 112                         format_url, video_id, mpd_id='dash', fatal=False))
 113                 elif format_url.endswith('.ism/Manifest'):
 114                     formats.extend(self._extract_ism_formats(
 115                         format_url, video_id, ism_id='mss', fatal=False))
 116                 else:
 117                     mfs_path = e.get('Type')
 118                     mfs_info = self._MEDIA_FILE_SLOTS.get(mfs_path)
 119                     if not mfs_info:
 120                         continue
 121                     formats.append({
 122                         'url': format_url,
 123                         'format_id': 'http-' + mfs_path.split('.')[0],
 124                         'width': mfs_info['width'],
 125                         'height': mfs_info['height'],
 126                     })
 127             self._sort_formats(formats)
 128
 129             description = media.get('Description')
 130             duration = int_or_none(media.get(
 131                 'DurationInSeconds')) or parse_duration(media.get('Duration'))
 132             view_count = int_or_none(media.get('CountViews'))
 133             average_rating = float_or_none(media.get('UserRating'))
 134             age_limit = parse_age_limit(media.get('Rating'))
 135             genre = media.get('Genre')
 136             release_year = int_or_none(media.get('ReleaseYear'))
 137             creator = media.get('Directors')
 138             artist = media.get('Cast')
 139
 140             if media.get('MediaTypeDisplayValue') == 'Full Episode':
 141                 series = media.get('ShowName')
 142                 episode = title
 143                 season_number = int_or_none(media.get('Season'))
 144                 episode_number = int_or_none(media.get('Episode'))
 145             else:
 146                 series = episode = season_number = episode_number = None
 147
 148             subtitles = {}
 149             cc_files = media.get('ClosedCaptionFiles')
 150             if isinstance(cc_files, list):
 151                 for cc_file in cc_files:
 152                     if not isinstance(cc_file, dict):
 153                         continue
 154                     cc_url = url_or_none(cc_file.get('Path'))
 155                     if not cc_url:
 156                         continue
 157                     lang = cc_file.get('Locale') or 'en'
 158                     subtitles.setdefault(lang, []).append({'url': cc_url})
 159
 160             thumbnails = []
 161             images = media.get('Images')
 162             if isinstance(images, list):
 163                 for image_key, image_url in images.items():
 164                     mobj = re.search(r'Img_(\d+)[xX](\d+)', image_key)
 165                     if not mobj:
 166                         continue
 167                     thumbnails.append({
 168                         'url': image_url,
 169                         'width': int(mobj.group(1)),
 170                         'height': int(mobj.group(2)),
 171                     })
 172
 173             return {
 174                 'id': video_id,
 175                 'title': title,
 176                 'description': description,
 177                 'duration': duration,
 178                 'view_count': view_count,
 179                 'average_rating': average_rating,
 180                 'age_limit': age_limit,
 181                 'genre': genre,
 182                 'creator': creator,
 183                 'artist': artist,
 184                 'release_year': release_year,
 185                 'series': series,
 186                 'episode': episode,
 187                 'season_number': season_number,
 188                 'episode_number': episode_number,
 189                 'thumbnails': thumbnails,
 190                 'subtitles': subtitles,
 191                 'formats': formats,
 192             }
 193
 194         raise last_e