_ Git - youtube-dl/blob - youtube_dl/extractor/go.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .adobepass import AdobePassIE
   7 from ..utils import (
   8     int_or_none,
   9     determine_ext,
  10     parse_age_limit,
  11     urlencode_postdata,
  12     ExtractorError,
  13 )
  14
  15
  16 class GoIE(AdobePassIE):
  17     _SITE_INFO = {
  18         'abc': {
  19             'brand': '001',
  20             'requestor_id': 'ABC',
  21         },
  22         'freeform': {
  23             'brand': '002',
  24             'requestor_id': 'ABCFamily',
  25         },
  26         'watchdisneychannel': {
  27             'brand': '004',
  28             'resource_id': 'Disney',
  29         },
  30         'watchdisneyjunior': {
  31             'brand': '008',
  32             'resource_id': 'DisneyJunior',
  33         },
  34         'watchdisneyxd': {
  35             'brand': '009',
  36             'resource_id': 'DisneyXD',
  37         },
  38         'disneynow': {
  39             'brand': '011',
  40             'resource_id': 'Disney',
  41         }
  42     }
  43     _VALID_URL = r'https?://(?:(?:(?P<sub_domain>%s)\.)?go|(?P<sub_domain_2>abc|freeform|disneynow))\.com/(?:(?:[^/]+/)*(?P<id>vdka\w+)|(?:[^/]+/)*(?P<display_id>[^/?#]+))'\
  44                  % '|'.join(list(_SITE_INFO.keys()))
  45     _TESTS = [{
  46         'url': 'http://abc.go.com/shows/designated-survivor/video/most-recent/VDKA3807643',
  47         'info_dict': {
  48             'id': 'VDKA3807643',
  49             'ext': 'mp4',
  50             'title': 'The Traitor in the White House',
  51             'description': 'md5:05b009d2d145a1e85d25111bd37222e8',
  52         },
  53         'params': {
  54             # m3u8 download
  55             'skip_download': True,
  56         },
  57         'skip': 'This content is no longer available.',
  58     }, {
  59         'url': 'http://watchdisneyxd.go.com/doraemon',
  60         'info_dict': {
  61             'title': 'Doraemon',
  62             'id': 'SH55574025',
  63         },
  64         'playlist_mincount': 51,
  65     }, {
  66         'url': 'http://freeform.go.com/shows/shadowhunters/episodes/season-2/1-this-guilty-blood',
  67         'info_dict': {
  68             'id': 'VDKA3609139',
  69             'ext': 'mp4',
  70             'title': 'This Guilty Blood',
  71             'description': 'md5:f18e79ad1c613798d95fdabfe96cd292',
  72             'age_limit': 14,
  73         },
  74         'params': {
  75             'geo_bypass_ip_block': '3.244.239.0/24',
  76             # m3u8 download
  77             'skip_download': True,
  78         },
  79     }, {
  80         'url': 'https://abc.com/shows/the-rookie/episode-guide/season-02/03-the-bet',
  81         'info_dict': {
  82             'id': 'VDKA13435179',
  83             'ext': 'mp4',
  84             'title': 'The Bet',
  85             'description': 'md5:c66de8ba2e92c6c5c113c3ade84ab404',
  86             'age_limit': 14,
  87         },
  88         'params': {
  89             'geo_bypass_ip_block': '3.244.239.0/24',
  90             # m3u8 download
  91             'skip_download': True,
  92         },
  93     }, {
  94         'url': 'http://abc.go.com/shows/the-catch/episode-guide/season-01/10-the-wedding',
  95         'only_matching': True,
  96     }, {
  97         'url': 'http://abc.go.com/shows/world-news-tonight/episode-guide/2017-02/17-021717-intense-stand-off-between-man-with-rifle-and-police-in-oakland',
  98         'only_matching': True,
  99     }, {
 100         # brand 004
 101         'url': 'http://disneynow.go.com/shows/big-hero-6-the-series/season-01/episode-10-mr-sparkles-loses-his-sparkle/vdka4637915',
 102         'only_matching': True,
 103     }, {
 104         # brand 008
 105         'url': 'http://disneynow.go.com/shows/minnies-bow-toons/video/happy-campers/vdka4872013',
 106         'only_matching': True,
 107     }, {
 108         'url': 'https://disneynow.com/shows/minnies-bow-toons/video/happy-campers/vdka4872013',
 109         'only_matching': True,
 110     }]
 111
 112     def _extract_videos(self, brand, video_id='-1', show_id='-1'):
 113         display_id = video_id if video_id != '-1' else show_id
 114         return self._download_json(
 115             'http://api.contents.watchabc.go.com/vp2/ws/contents/3000/videos/%s/001/-1/%s/-1/%s/-1/-1.json' % (brand, show_id, video_id),
 116             display_id)['video']
 117
 118     def _real_extract(self, url):
 119         mobj = re.match(self._VALID_URL, url)
 120         sub_domain = mobj.group('sub_domain') or mobj.group('sub_domain_2')
 121         video_id, display_id = mobj.group('id', 'display_id')
 122         site_info = self._SITE_INFO.get(sub_domain, {})
 123         brand = site_info.get('brand')
 124         if not video_id or not site_info:
 125             webpage = self._download_webpage(url, display_id or video_id)
 126             video_id = self._search_regex(
 127                 (
 128                     # There may be inner quotes, e.g. data-video-id="'VDKA3609139'"
 129                     # from http://freeform.go.com/shows/shadowhunters/episodes/season-2/1-this-guilty-blood
 130                     r'data-video-id=["\']*(VDKA\w+)',
 131                     # https://abc.com/shows/the-rookie/episode-guide/season-02/03-the-bet
 132                     r'\b(?:video)?id["\']\s*:\s*["\'](VDKA\w+)'
 133                 ), webpage, 'video id', default=video_id)
 134             if not site_info:
 135                 brand = self._search_regex(
 136                     (r'data-brand=\s*["\']\s*(\d+)',
 137                      r'data-page-brand=\s*["\']\s*(\d+)'), webpage, 'brand',
 138                     default='004')
 139                 site_info = next(
 140                     si for _, si in self._SITE_INFO.items()
 141                     if si.get('brand') == brand)
 142             if not video_id:
 143                 # show extraction works for Disney, DisneyJunior and DisneyXD
 144                 # ABC and Freeform has different layout
 145                 show_id = self._search_regex(r'data-show-id=["\']*(SH\d+)', webpage, 'show id')
 146                 videos = self._extract_videos(brand, show_id=show_id)
 147                 show_title = self._search_regex(r'data-show-title="([^"]+)"', webpage, 'show title', fatal=False)
 148                 entries = []
 149                 for video in videos:
 150                     entries.append(self.url_result(
 151                         video['url'], 'Go', video.get('id'), video.get('title')))
 152                 entries.reverse()
 153                 return self.playlist_result(entries, show_id, show_title)
 154         video_data = self._extract_videos(brand, video_id)[0]
 155         video_id = video_data['id']
 156         title = video_data['title']
 157
 158         formats = []
 159         for asset in video_data.get('assets', {}).get('asset', []):
 160             asset_url = asset.get('value')
 161             if not asset_url:
 162                 continue
 163             format_id = asset.get('format')
 164             ext = determine_ext(asset_url)
 165             if ext == 'm3u8':
 166                 video_type = video_data.get('type')
 167                 data = {
 168                     'video_id': video_data['id'],
 169                     'video_type': video_type,
 170                     'brand': brand,
 171                     'device': '001',
 172                 }
 173                 if video_data.get('accesslevel') == '1':
 174                     requestor_id = site_info.get('requestor_id', 'DisneyChannels')
 175                     resource = site_info.get('resource_id') or self._get_mvpd_resource(
 176                         requestor_id, title, video_id, None)
 177                     auth = self._extract_mvpd_auth(
 178                         url, video_id, requestor_id, resource)
 179                     data.update({
 180                         'token': auth,
 181                         'token_type': 'ap',
 182                         'adobe_requestor_id': requestor_id,
 183                     })
 184                 else:
 185                     self._initialize_geo_bypass({'countries': ['US']})
 186                 entitlement = self._download_json(
 187                     'https://api.entitlement.watchabc.go.com/vp2/ws-secure/entitlement/2020/authorize.json',
 188                     video_id, data=urlencode_postdata(data))
 189                 errors = entitlement.get('errors', {}).get('errors', [])
 190                 if errors:
 191                     for error in errors:
 192                         if error.get('code') == 1002:
 193                             self.raise_geo_restricted(
 194                                 error['message'], countries=['US'])
 195                     error_message = ', '.join([error['message'] for error in errors])
 196                     raise ExtractorError('%s said: %s' % (self.IE_NAME, error_message), expected=True)
 197                 asset_url += '?' + entitlement['uplynkData']['sessionKey']
 198                 formats.extend(self._extract_m3u8_formats(
 199                     asset_url, video_id, 'mp4', m3u8_id=format_id or 'hls', fatal=False))
 200             else:
 201                 f = {
 202                     'format_id': format_id,
 203                     'url': asset_url,
 204                     'ext': ext,
 205                 }
 206                 if re.search(r'(?:/mp4/source/|_source\.mp4)', asset_url):
 207                     f.update({
 208                         'format_id': ('%s-' % format_id if format_id else '') + 'SOURCE',
 209                         'preference': 1,
 210                     })
 211                 else:
 212                     mobj = re.search(r'/(\d+)x(\d+)/', asset_url)
 213                     if mobj:
 214                         height = int(mobj.group(2))
 215                         f.update({
 216                             'format_id': ('%s-' % format_id if format_id else '') + '%dP' % height,
 217                             'width': int(mobj.group(1)),
 218                             'height': height,
 219                         })
 220                 formats.append(f)
 221         self._sort_formats(formats)
 222
 223         subtitles = {}
 224         for cc in video_data.get('closedcaption', {}).get('src', []):
 225             cc_url = cc.get('value')
 226             if not cc_url:
 227                 continue
 228             ext = determine_ext(cc_url)
 229             if ext == 'xml':
 230                 ext = 'ttml'
 231             subtitles.setdefault(cc.get('lang'), []).append({
 232                 'url': cc_url,
 233                 'ext': ext,
 234             })
 235
 236         thumbnails = []
 237         for thumbnail in video_data.get('thumbnails', {}).get('thumbnail', []):
 238             thumbnail_url = thumbnail.get('value')
 239             if not thumbnail_url:
 240                 continue
 241             thumbnails.append({
 242                 'url': thumbnail_url,
 243                 'width': int_or_none(thumbnail.get('width')),
 244                 'height': int_or_none(thumbnail.get('height')),
 245             })
 246
 247         return {
 248             'id': video_id,
 249             'title': title,
 250             'description': video_data.get('longdescription') or video_data.get('description'),
 251             'duration': int_or_none(video_data.get('duration', {}).get('value'), 1000),
 252             'age_limit': parse_age_limit(video_data.get('tvrating', {}).get('rating')),
 253             'episode_number': int_or_none(video_data.get('episodenumber')),
 254             'series': video_data.get('show', {}).get('title'),
 255             'season_number': int_or_none(video_data.get('season', {}).get('num')),
 256             'thumbnails': thumbnails,
 257             'formats': formats,
 258             'subtitles': subtitles,
 259         }