youtube_dlc/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'''(?x)
  44                     https?://
  45                         (?:
  46                             (?:(?P<sub_domain>%s)\.)?go|
  47                             (?P<sub_domain_2>abc|freeform|disneynow)
  48                         )\.com/
  49                         (?:
  50                             (?:[^/]+/)*(?P<id>[Vv][Dd][Kk][Aa]\w+)|
  51                             (?:[^/]+/)*(?P<display_id>[^/?\#]+)
  52                         )
  53                     ''' % '|'.join(list(_SITE_INFO.keys()))
  54     _TESTS = [{
  55         'url': 'http://abc.go.com/shows/designated-survivor/video/most-recent/VDKA3807643',
  56         'info_dict': {
  57             'id': 'VDKA3807643',
  58             'ext': 'mp4',
  59             'title': 'The Traitor in the White House',
  60             'description': 'md5:05b009d2d145a1e85d25111bd37222e8',
  61         },
  62         'params': {
  63             # m3u8 download
  64             'skip_download': True,
  65         },
  66         'skip': 'This content is no longer available.',
  67     }, {
  68         'url': 'http://watchdisneyxd.go.com/doraemon',
  69         'info_dict': {
  70             'title': 'Doraemon',
  71             'id': 'SH55574025',
  72         },
  73         'playlist_mincount': 51,
  74     }, {
  75         'url': 'http://freeform.go.com/shows/shadowhunters/episodes/season-2/1-this-guilty-blood',
  76         'info_dict': {
  77             'id': 'VDKA3609139',
  78             'ext': 'mp4',
  79             'title': 'This Guilty Blood',
  80             'description': 'md5:f18e79ad1c613798d95fdabfe96cd292',
  81             'age_limit': 14,
  82         },
  83         'params': {
  84             'geo_bypass_ip_block': '3.244.239.0/24',
  85             # m3u8 download
  86             'skip_download': True,
  87         },
  88     }, {
  89         'url': 'https://abc.com/shows/the-rookie/episode-guide/season-02/03-the-bet',
  90         'info_dict': {
  91             'id': 'VDKA13435179',
  92             'ext': 'mp4',
  93             'title': 'The Bet',
  94             'description': 'md5:c66de8ba2e92c6c5c113c3ade84ab404',
  95             'age_limit': 14,
  96         },
  97         'params': {
  98             'geo_bypass_ip_block': '3.244.239.0/24',
  99             # m3u8 download
 100             'skip_download': True,
 101         },
 102     }, {
 103         'url': 'http://abc.go.com/shows/the-catch/episode-guide/season-01/10-the-wedding',
 104         'only_matching': True,
 105     }, {
 106         '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',
 107         'only_matching': True,
 108     }, {
 109         # brand 004
 110         'url': 'http://disneynow.go.com/shows/big-hero-6-the-series/season-01/episode-10-mr-sparkles-loses-his-sparkle/vdka4637915',
 111         'only_matching': True,
 112     }, {
 113         # brand 008
 114         'url': 'http://disneynow.go.com/shows/minnies-bow-toons/video/happy-campers/vdka4872013',
 115         'only_matching': True,
 116     }, {
 117         'url': 'https://disneynow.com/shows/minnies-bow-toons/video/happy-campers/vdka4872013',
 118         'only_matching': True,
 119     }]
 120
 121     def _extract_videos(self, brand, video_id='-1', show_id='-1'):
 122         display_id = video_id if video_id != '-1' else show_id
 123         return self._download_json(
 124             '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),
 125             display_id)['video']
 126
 127     def _real_extract(self, url):
 128         mobj = re.match(self._VALID_URL, url)
 129         sub_domain = mobj.group('sub_domain') or mobj.group('sub_domain_2')
 130         video_id, display_id = mobj.group('id', 'display_id')
 131         site_info = self._SITE_INFO.get(sub_domain, {})
 132         brand = site_info.get('brand')
 133         if not video_id or not site_info:
 134             webpage = self._download_webpage(url, display_id or video_id)
 135             video_id = self._search_regex(
 136                 (
 137                     # There may be inner quotes, e.g. data-video-id="'VDKA3609139'"
 138                     # from http://freeform.go.com/shows/shadowhunters/episodes/season-2/1-this-guilty-blood
 139                     r'data-video-id=["\']*(VDKA\w+)',
 140                     # https://github.com/ytdl-org/youtube-dl/pull/25216/files
 141                     #  The following is based on the pull request on the line above. Changed the ABC.com URL to a show available now.
 142                     # https://abc.com/shows/the-rookie/episode-guide/season-02/19-the-q-word
 143                     r'\bvideoIdCode["\']\s*:\s*["\'](vdka\w+)',
 144                     # Deprecated  fallback pattern
 145                     r'\b(?:video)?id["\']\s*:\s*["\'](VDKA\w+)'
 146                 ), webpage, 'video id', default=video_id)
 147             if not site_info:
 148                 brand = self._search_regex(
 149                     (r'data-brand=\s*["\']\s*(\d+)',
 150                      r'data-page-brand=\s*["\']\s*(\d+)'), webpage, 'brand',
 151                     default='004')
 152                 site_info = next(
 153                     si for _, si in self._SITE_INFO.items()
 154                     if si.get('brand') == brand)
 155             if not video_id:
 156                 # show extraction works for Disney, DisneyJunior and DisneyXD
 157                 # ABC and Freeform has different layout
 158                 show_id = self._search_regex(r'data-show-id=["\']*(SH\d+)', webpage, 'show id')
 159                 videos = self._extract_videos(brand, show_id=show_id)
 160                 show_title = self._search_regex(r'data-show-title="([^"]+)"', webpage, 'show title', fatal=False)
 161                 entries = []
 162                 for video in videos:
 163                     entries.append(self.url_result(
 164                         video['url'], 'Go', video.get('id'), video.get('title')))
 165                 entries.reverse()
 166                 return self.playlist_result(entries, show_id, show_title)
 167         video_data = self._extract_videos(brand, video_id)[0]
 168         video_id = video_data['id']
 169         title = video_data['title']
 170
 171         formats = []
 172         for asset in video_data.get('assets', {}).get('asset', []):
 173             asset_url = asset.get('value')
 174             if not asset_url:
 175                 continue
 176             format_id = asset.get('format')
 177             ext = determine_ext(asset_url)
 178             if ext == 'm3u8':
 179                 video_type = video_data.get('type')
 180                 data = {
 181                     'video_id': video_data['id'],
 182                     'video_type': video_type,
 183                     'brand': brand,
 184                     'device': '001',
 185                 }
 186                 if video_data.get('accesslevel') == '1':
 187                     requestor_id = site_info.get('requestor_id', 'DisneyChannels')
 188                     resource = site_info.get('resource_id') or self._get_mvpd_resource(
 189                         requestor_id, title, video_id, None)
 190                     auth = self._extract_mvpd_auth(
 191                         url, video_id, requestor_id, resource)
 192                     data.update({
 193                         'token': auth,
 194                         'token_type': 'ap',
 195                         'adobe_requestor_id': requestor_id,
 196                     })
 197                 else:
 198                     self._initialize_geo_bypass({'countries': ['US']})
 199                 entitlement = self._download_json(
 200                     'https://api.entitlement.watchabc.go.com/vp2/ws-secure/entitlement/2020/authorize.json',
 201                     video_id, data=urlencode_postdata(data))
 202                 errors = entitlement.get('errors', {}).get('errors', [])
 203                 if errors:
 204                     for error in errors:
 205                         if error.get('code') == 1002:
 206                             self.raise_geo_restricted(
 207                                 error['message'], countries=['US'])
 208                     error_message = ', '.join([error['message'] for error in errors])
 209                     raise ExtractorError('%s said: %s' % (self.IE_NAME, error_message), expected=True)
 210                 asset_url += '?' + entitlement['uplynkData']['sessionKey']
 211                 formats.extend(self._extract_m3u8_formats(
 212                     asset_url, video_id, 'mp4', m3u8_id=format_id or 'hls', fatal=False))
 213             else:
 214                 f = {
 215                     'format_id': format_id,
 216                     'url': asset_url,
 217                     'ext': ext,
 218                 }
 219                 if re.search(r'(?:/mp4/source/|_source\.mp4)', asset_url):
 220                     f.update({
 221                         'format_id': ('%s-' % format_id if format_id else '') + 'SOURCE',
 222                         'preference': 1,
 223                     })
 224                 else:
 225                     mobj = re.search(r'/(\d+)x(\d+)/', asset_url)
 226                     if mobj:
 227                         height = int(mobj.group(2))
 228                         f.update({
 229                             'format_id': ('%s-' % format_id if format_id else '') + '%dP' % height,
 230                             'width': int(mobj.group(1)),
 231                             'height': height,
 232                         })
 233                 formats.append(f)
 234         self._sort_formats(formats)
 235
 236         subtitles = {}
 237         for cc in video_data.get('closedcaption', {}).get('src', []):
 238             cc_url = cc.get('value')
 239             if not cc_url:
 240                 continue
 241             ext = determine_ext(cc_url)
 242             if ext == 'xml':
 243                 ext = 'ttml'
 244             subtitles.setdefault(cc.get('lang'), []).append({
 245                 'url': cc_url,
 246                 'ext': ext,
 247             })
 248
 249         thumbnails = []
 250         for thumbnail in video_data.get('thumbnails', {}).get('thumbnail', []):
 251             thumbnail_url = thumbnail.get('value')
 252             if not thumbnail_url:
 253                 continue
 254             thumbnails.append({
 255                 'url': thumbnail_url,
 256                 'width': int_or_none(thumbnail.get('width')),
 257                 'height': int_or_none(thumbnail.get('height')),
 258             })
 259
 260         return {
 261             'id': video_id,
 262             'title': title,
 263             'description': video_data.get('longdescription') or video_data.get('description'),
 264             'duration': int_or_none(video_data.get('duration', {}).get('value'), 1000),
 265             'age_limit': parse_age_limit(video_data.get('tvrating', {}).get('rating')),
 266             'episode_number': int_or_none(video_data.get('episodenumber')),
 267             'series': video_data.get('show', {}).get('title'),
 268             'season_number': int_or_none(video_data.get('season', {}).get('num')),
 269             'thumbnails': thumbnails,
 270             'formats': formats,
 271             'subtitles': subtitles,
 272         }