Raphaël G. Git Repositories - youtubedl/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     }
  39     _VALID_URL = r'https?://(?:(?P<sub_domain>%s)\.)?go\.com/(?:(?:[^/]+/)*(?P<id>vdka\w+)|(?:[^/]+/)*(?P<display_id>[^/?#]+))'\
  40                  % '|'.join(list(_SITE_INFO.keys()) + ['disneynow'])
  41     _TESTS = [{
  42         'url': 'http://abc.go.com/shows/designated-survivor/video/most-recent/VDKA3807643',
  43         'info_dict': {
  44             'id': 'VDKA3807643',
  45             'ext': 'mp4',
  46             'title': 'The Traitor in the White House',
  47             'description': 'md5:05b009d2d145a1e85d25111bd37222e8',
  48         },
  49         'params': {
  50             # m3u8 download
  51             'skip_download': True,
  52         },
  53     }, {
  54         'url': 'http://watchdisneyxd.go.com/doraemon',
  55         'info_dict': {
  56             'title': 'Doraemon',
  57             'id': 'SH55574025',
  58         },
  59         'playlist_mincount': 51,
  60     }, {
  61         'url': 'http://abc.go.com/shows/the-catch/episode-guide/season-01/10-the-wedding',
  62         'only_matching': True,
  63     }, {
  64         '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',
  65         'only_matching': True,
  66     }, {
  67         # brand 004
  68         'url': 'http://disneynow.go.com/shows/big-hero-6-the-series/season-01/episode-10-mr-sparkles-loses-his-sparkle/vdka4637915',
  69         'only_matching': True,
  70     }, {
  71         # brand 008
  72         'url': 'http://disneynow.go.com/shows/minnies-bow-toons/video/happy-campers/vdka4872013',
  73         'only_matching': True,
  74     }]
  75
  76     def _extract_videos(self, brand, video_id='-1', show_id='-1'):
  77         display_id = video_id if video_id != '-1' else show_id
  78         return self._download_json(
  79             '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),
  80             display_id)['video']
  81
  82     def _real_extract(self, url):
  83         sub_domain, video_id, display_id = re.match(self._VALID_URL, url).groups()
  84         site_info = self._SITE_INFO.get(sub_domain, {})
  85         brand = site_info.get('brand')
  86         if not video_id or not site_info:
  87             webpage = self._download_webpage(url, display_id or video_id)
  88             video_id = self._search_regex(
  89                 # There may be inner quotes, e.g. data-video-id="'VDKA3609139'"
  90                 # from http://freeform.go.com/shows/shadowhunters/episodes/season-2/1-this-guilty-blood
  91                 r'data-video-id=["\']*(VDKA\w+)', webpage, 'video id',
  92                 default=None)
  93             if not site_info:
  94                 brand = self._search_regex(
  95                     (r'data-brand=\s*["\']\s*(\d+)',
  96                      r'data-page-brand=\s*["\']\s*(\d+)'), webpage, 'brand',
  97                     default='004')
  98                 site_info = next(
  99                     si for _, si in self._SITE_INFO.items()
 100                     if si.get('brand') == brand)
 101             if not video_id:
 102                 # show extraction works for Disney, DisneyJunior and DisneyXD
 103                 # ABC and Freeform has different layout
 104                 show_id = self._search_regex(r'data-show-id=["\']*(SH\d+)', webpage, 'show id')
 105                 videos = self._extract_videos(brand, show_id=show_id)
 106                 show_title = self._search_regex(r'data-show-title="([^"]+)"', webpage, 'show title', fatal=False)
 107                 entries = []
 108                 for video in videos:
 109                     entries.append(self.url_result(
 110                         video['url'], 'Go', video.get('id'), video.get('title')))
 111                 entries.reverse()
 112                 return self.playlist_result(entries, show_id, show_title)
 113         video_data = self._extract_videos(brand, video_id)[0]
 114         video_id = video_data['id']
 115         title = video_data['title']
 116
 117         formats = []
 118         for asset in video_data.get('assets', {}).get('asset', []):
 119             asset_url = asset.get('value')
 120             if not asset_url:
 121                 continue
 122             format_id = asset.get('format')
 123             ext = determine_ext(asset_url)
 124             if ext == 'm3u8':
 125                 video_type = video_data.get('type')
 126                 data = {
 127                     'video_id': video_data['id'],
 128                     'video_type': video_type,
 129                     'brand': brand,
 130                     'device': '001',
 131                 }
 132                 if video_data.get('accesslevel') == '1':
 133                     requestor_id = site_info.get('requestor_id', 'DisneyChannels')
 134                     resource = site_info.get('resource_id') or self._get_mvpd_resource(
 135                         requestor_id, title, video_id, None)
 136                     auth = self._extract_mvpd_auth(
 137                         url, video_id, requestor_id, resource)
 138                     data.update({
 139                         'token': auth,
 140                         'token_type': 'ap',
 141                         'adobe_requestor_id': requestor_id,
 142                     })
 143                 else:
 144                     self._initialize_geo_bypass({'countries': ['US']})
 145                 entitlement = self._download_json(
 146                     'https://api.entitlement.watchabc.go.com/vp2/ws-secure/entitlement/2020/authorize.json',
 147                     video_id, data=urlencode_postdata(data))
 148                 errors = entitlement.get('errors', {}).get('errors', [])
 149                 if errors:
 150                     for error in errors:
 151                         if error.get('code') == 1002:
 152                             self.raise_geo_restricted(
 153                                 error['message'], countries=['US'])
 154                     error_message = ', '.join([error['message'] for error in errors])
 155                     raise ExtractorError('%s said: %s' % (self.IE_NAME, error_message), expected=True)
 156                 asset_url += '?' + entitlement['uplynkData']['sessionKey']
 157                 formats.extend(self._extract_m3u8_formats(
 158                     asset_url, video_id, 'mp4', m3u8_id=format_id or 'hls', fatal=False))
 159             else:
 160                 f = {
 161                     'format_id': format_id,
 162                     'url': asset_url,
 163                     'ext': ext,
 164                 }
 165                 if re.search(r'(?:/mp4/source/|_source\.mp4)', asset_url):
 166                     f.update({
 167                         'format_id': ('%s-' % format_id if format_id else '') + 'SOURCE',
 168                         'preference': 1,
 169                     })
 170                 else:
 171                     mobj = re.search(r'/(\d+)x(\d+)/', asset_url)
 172                     if mobj:
 173                         height = int(mobj.group(2))
 174                         f.update({
 175                             'format_id': ('%s-' % format_id if format_id else '') + '%dP' % height,
 176                             'width': int(mobj.group(1)),
 177                             'height': height,
 178                         })
 179                 formats.append(f)
 180         self._sort_formats(formats)
 181
 182         subtitles = {}
 183         for cc in video_data.get('closedcaption', {}).get('src', []):
 184             cc_url = cc.get('value')
 185             if not cc_url:
 186                 continue
 187             ext = determine_ext(cc_url)
 188             if ext == 'xml':
 189                 ext = 'ttml'
 190             subtitles.setdefault(cc.get('lang'), []).append({
 191                 'url': cc_url,
 192                 'ext': ext,
 193             })
 194
 195         thumbnails = []
 196         for thumbnail in video_data.get('thumbnails', {}).get('thumbnail', []):
 197             thumbnail_url = thumbnail.get('value')
 198             if not thumbnail_url:
 199                 continue
 200             thumbnails.append({
 201                 'url': thumbnail_url,
 202                 'width': int_or_none(thumbnail.get('width')),
 203                 'height': int_or_none(thumbnail.get('height')),
 204             })
 205
 206         return {
 207             'id': video_id,
 208             'title': title,
 209             'description': video_data.get('longdescription') or video_data.get('description'),
 210             'duration': int_or_none(video_data.get('duration', {}).get('value'), 1000),
 211             'age_limit': parse_age_limit(video_data.get('tvrating', {}).get('rating')),
 212             'episode_number': int_or_none(video_data.get('episodenumber')),
 213             'series': video_data.get('show', {}).get('title'),
 214             'season_number': int_or_none(video_data.get('season', {}).get('num')),
 215             'thumbnails': thumbnails,
 216             'formats': formats,
 217             'subtitles': subtitles,
 218         }