_ 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             '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/(?:[^/]+/)*(?:vdka(?P<id>\w+)|season-\d+/\d+-(?P<display_id>[^/?#]+))' % '|'.join(_SITE_INFO.keys())
  40     _GEO_COUNTRIES = ['US']
  41     _TESTS = [{
  42         'url': 'http://abc.go.com/shows/castle/video/most-recent/vdka0_g86w5onx',
  43         'info_dict': {
  44             'id': '0_g86w5onx',
  45             'ext': 'mp4',
  46             'title': 'Sneak Peek: Language Arts',
  47             'description': 'md5:7dcdab3b2d17e5217c953256af964e9c',
  48         },
  49         'params': {
  50             # m3u8 download
  51             'skip_download': True,
  52         },
  53     }, {
  54         'url': 'http://abc.go.com/shows/after-paradise/video/most-recent/vdka3335601',
  55         'only_matching': True,
  56     }]
  57
  58     def _real_extract(self, url):
  59         sub_domain, video_id, display_id = re.match(self._VALID_URL, url).groups()
  60         if not video_id:
  61             webpage = self._download_webpage(url, display_id)
  62             video_id = self._search_regex(
  63                 # There may be inner quotes, e.g. data-video-id="'VDKA3609139'"
  64                 # from http://freeform.go.com/shows/shadowhunters/episodes/season-2/1-this-guilty-blood
  65                 r'data-video-id=["\']*VDKA(\w+)', webpage, 'video id')
  66         site_info = self._SITE_INFO[sub_domain]
  67         brand = site_info['brand']
  68         video_data = self._download_json(
  69             'http://api.contents.watchabc.go.com/vp2/ws/contents/3000/videos/%s/001/-1/-1/-1/%s/-1/-1.json' % (brand, video_id),
  70             video_id)['video'][0]
  71         title = video_data['title']
  72
  73         formats = []
  74         for asset in video_data.get('assets', {}).get('asset', []):
  75             asset_url = asset.get('value')
  76             if not asset_url:
  77                 continue
  78             format_id = asset.get('format')
  79             ext = determine_ext(asset_url)
  80             if ext == 'm3u8':
  81                 video_type = video_data.get('type')
  82                 if video_type == 'lf':
  83                     data = {
  84                         'video_id': video_data['id'],
  85                         'video_type': video_type,
  86                         'brand': brand,
  87                         'device': '001',
  88                     }
  89                     if video_data.get('accesslevel') == '1':
  90                         requestor_id = site_info['requestor_id']
  91                         resource = self._get_mvpd_resource(
  92                             requestor_id, title, video_id, None)
  93                         auth = self._extract_mvpd_auth(
  94                             url, video_id, requestor_id, resource)
  95                         data.update({
  96                             'token': auth,
  97                             'token_type': 'ap',
  98                             'adobe_requestor_id': requestor_id,
  99                         })
 100                     entitlement = self._download_json(
 101                         'https://api.entitlement.watchabc.go.com/vp2/ws-secure/entitlement/2020/authorize.json',
 102                         video_id, data=urlencode_postdata(data), headers=self.geo_verification_headers())
 103                     errors = entitlement.get('errors', {}).get('errors', [])
 104                     if errors:
 105                         for error in errors:
 106                             if error.get('code') == 1002:
 107                                 self.raise_geo_restricted(
 108                                     error['message'], countries=self._GEO_COUNTRIES)
 109                         error_message = ', '.join([error['message'] for error in errors])
 110                         raise ExtractorError('%s said: %s' % (self.IE_NAME, error_message), expected=True)
 111                     asset_url += '?' + entitlement['uplynkData']['sessionKey']
 112                 formats.extend(self._extract_m3u8_formats(
 113                     asset_url, video_id, 'mp4', m3u8_id=format_id or 'hls', fatal=False))
 114             else:
 115                 formats.append({
 116                     'format_id': format_id,
 117                     'url': asset_url,
 118                     'ext': ext,
 119                 })
 120         self._sort_formats(formats)
 121
 122         subtitles = {}
 123         for cc in video_data.get('closedcaption', {}).get('src', []):
 124             cc_url = cc.get('value')
 125             if not cc_url:
 126                 continue
 127             ext = determine_ext(cc_url)
 128             if ext == 'xml':
 129                 ext = 'ttml'
 130             subtitles.setdefault(cc.get('lang'), []).append({
 131                 'url': cc_url,
 132                 'ext': ext,
 133             })
 134
 135         thumbnails = []
 136         for thumbnail in video_data.get('thumbnails', {}).get('thumbnail', []):
 137             thumbnail_url = thumbnail.get('value')
 138             if not thumbnail_url:
 139                 continue
 140             thumbnails.append({
 141                 'url': thumbnail_url,
 142                 'width': int_or_none(thumbnail.get('width')),
 143                 'height': int_or_none(thumbnail.get('height')),
 144             })
 145
 146         return {
 147             'id': video_id,
 148             'title': title,
 149             'description': video_data.get('longdescription') or video_data.get('description'),
 150             'duration': int_or_none(video_data.get('duration', {}).get('value'), 1000),
 151             'age_limit': parse_age_limit(video_data.get('tvrating', {}).get('rating')),
 152             'episode_number': int_or_none(video_data.get('episodenumber')),
 153             'series': video_data.get('show', {}).get('title'),
 154             'season_number': int_or_none(video_data.get('season', {}).get('num')),
 155             'thumbnails': thumbnails,
 156             'formats': formats,
 157             'subtitles': subtitles,
 158         }