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             'requestor_id': 'Disney',
  29         },
  30         'watchdisneyjunior': {
  31             'brand': '008',
  32             'requestor_id': 'DisneyJunior',
  33         },
  34         'watchdisneyxd': {
  35             'brand': '009',
  36             'requestor_id': 'DisneyXD',
  37         }
  38     }
  39     _VALID_URL = r'https?://(?:(?P<sub_domain>%s)\.)?go\.com/(?:(?:[^/]+/)*(?P<id>vdka\w+)|(?:[^/]+/)*(?P<display_id>[^/?#]+))' % '|'.join(_SITE_INFO.keys())
  40     _TESTS = [{
  41         'url': 'http://abc.go.com/shows/designated-survivor/video/most-recent/VDKA3807643',
  42         'info_dict': {
  43             'id': 'VDKA3807643',
  44             'ext': 'mp4',
  45             'title': 'The Traitor in the White House',
  46             'description': 'md5:05b009d2d145a1e85d25111bd37222e8',
  47         },
  48         'params': {
  49             # m3u8 download
  50             'skip_download': True,
  51         },
  52     }, {
  53         'url': 'http://watchdisneyxd.go.com/doraemon',
  54         'info_dict': {
  55             'title': 'Doraemon',
  56             'id': 'SH55574025',
  57         },
  58         'playlist_mincount': 51,
  59     }, {
  60         'url': 'http://abc.go.com/shows/the-catch/episode-guide/season-01/10-the-wedding',
  61         'only_matching': True,
  62     }, {
  63         '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',
  64         'only_matching': True,
  65     }]
  66
  67     def _extract_videos(self, brand, video_id='-1', show_id='-1'):
  68         display_id = video_id if video_id != '-1' else show_id
  69         return self._download_json(
  70             '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),
  71             display_id)['video']
  72
  73     def _real_extract(self, url):
  74         sub_domain, video_id, display_id = re.match(self._VALID_URL, url).groups()
  75         site_info = self._SITE_INFO[sub_domain]
  76         brand = site_info['brand']
  77         if not video_id:
  78             webpage = self._download_webpage(url, display_id)
  79             video_id = self._search_regex(
  80                 # There may be inner quotes, e.g. data-video-id="'VDKA3609139'"
  81                 # from http://freeform.go.com/shows/shadowhunters/episodes/season-2/1-this-guilty-blood
  82                 r'data-video-id=["\']*(VDKA\w+)', webpage, 'video id', default=None)
  83             if not video_id:
  84                 # show extraction works for Disney, DisneyJunior and DisneyXD
  85                 # ABC and Freeform has different layout
  86                 show_id = self._search_regex(r'data-show-id=["\']*(SH\d+)', webpage, 'show id')
  87                 videos = self._extract_videos(brand, show_id=show_id)
  88                 show_title = self._search_regex(r'data-show-title="([^"]+)"', webpage, 'show title', fatal=False)
  89                 entries = []
  90                 for video in videos:
  91                     entries.append(self.url_result(
  92                         video['url'], 'Go', video.get('id'), video.get('title')))
  93                 entries.reverse()
  94                 return self.playlist_result(entries, show_id, show_title)
  95         video_data = self._extract_videos(brand, video_id)[0]
  96         video_id = video_data['id']
  97         title = video_data['title']
  98
  99         formats = []
 100         for asset in video_data.get('assets', {}).get('asset', []):
 101             asset_url = asset.get('value')
 102             if not asset_url:
 103                 continue
 104             format_id = asset.get('format')
 105             ext = determine_ext(asset_url)
 106             if ext == 'm3u8':
 107                 video_type = video_data.get('type')
 108                 data = {
 109                     'video_id': video_data['id'],
 110                     'video_type': video_type,
 111                     'brand': brand,
 112                     'device': '001',
 113                 }
 114                 if video_data.get('accesslevel') == '1':
 115                     requestor_id = site_info['requestor_id']
 116                     resource = self._get_mvpd_resource(
 117                         requestor_id, title, video_id, None)
 118                     auth = self._extract_mvpd_auth(
 119                         url, video_id, requestor_id, resource)
 120                     data.update({
 121                         'token': auth,
 122                         'token_type': 'ap',
 123                         'adobe_requestor_id': requestor_id,
 124                     })
 125                 else:
 126                     self._initialize_geo_bypass(['US'])
 127                 entitlement = self._download_json(
 128                     'https://api.entitlement.watchabc.go.com/vp2/ws-secure/entitlement/2020/authorize.json',
 129                     video_id, data=urlencode_postdata(data))
 130                 errors = entitlement.get('errors', {}).get('errors', [])
 131                 if errors:
 132                     for error in errors:
 133                         if error.get('code') == 1002:
 134                             self.raise_geo_restricted(
 135                                 error['message'], countries=['US'])
 136                     error_message = ', '.join([error['message'] for error in errors])
 137                     raise ExtractorError('%s said: %s' % (self.IE_NAME, error_message), expected=True)
 138                 asset_url += '?' + entitlement['uplynkData']['sessionKey']
 139                 formats.extend(self._extract_m3u8_formats(
 140                     asset_url, video_id, 'mp4', m3u8_id=format_id or 'hls', fatal=False))
 141             else:
 142                 f = {
 143                     'format_id': format_id,
 144                     'url': asset_url,
 145                     'ext': ext,
 146                 }
 147                 if re.search(r'(?:/mp4/source/|_source\.mp4)', asset_url):
 148                     f.update({
 149                         'format_id': ('%s-' % format_id if format_id else '') + 'SOURCE',
 150                         'preference': 1,
 151                     })
 152                 else:
 153                     mobj = re.search(r'/(\d+)x(\d+)/', asset_url)
 154                     if mobj:
 155                         height = int(mobj.group(2))
 156                         f.update({
 157                             'format_id': ('%s-' % format_id if format_id else '') + '%dP' % height,
 158                             'width': int(mobj.group(1)),
 159                             'height': height,
 160                         })
 161                 formats.append(f)
 162         self._sort_formats(formats)
 163
 164         subtitles = {}
 165         for cc in video_data.get('closedcaption', {}).get('src', []):
 166             cc_url = cc.get('value')
 167             if not cc_url:
 168                 continue
 169             ext = determine_ext(cc_url)
 170             if ext == 'xml':
 171                 ext = 'ttml'
 172             subtitles.setdefault(cc.get('lang'), []).append({
 173                 'url': cc_url,
 174                 'ext': ext,
 175             })
 176
 177         thumbnails = []
 178         for thumbnail in video_data.get('thumbnails', {}).get('thumbnail', []):
 179             thumbnail_url = thumbnail.get('value')
 180             if not thumbnail_url:
 181                 continue
 182             thumbnails.append({
 183                 'url': thumbnail_url,
 184                 'width': int_or_none(thumbnail.get('width')),
 185                 'height': int_or_none(thumbnail.get('height')),
 186             })
 187
 188         return {
 189             'id': video_id,
 190             'title': title,
 191             'description': video_data.get('longdescription') or video_data.get('description'),
 192             'duration': int_or_none(video_data.get('duration', {}).get('value'), 1000),
 193             'age_limit': parse_age_limit(video_data.get('tvrating', {}).get('rating')),
 194             'episode_number': int_or_none(video_data.get('episodenumber')),
 195             'series': video_data.get('show', {}).get('title'),
 196             'season_number': int_or_none(video_data.get('season', {}).get('num')),
 197             'thumbnails': thumbnails,
 198             'formats': formats,
 199             'subtitles': subtitles,
 200         }