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/(?:[^/]+/)*(?:vdka(?P<id>\w+)|season-\d+/\d+-(?P<display_id>[^/?#]+))' % '|'.join(_SITE_INFO.keys())
  40     _TESTS = [{
  41         'url': 'http://abc.go.com/shows/castle/video/most-recent/vdka0_g86w5onx',
  42         'info_dict': {
  43             'id': '0_g86w5onx',
  44             'ext': 'mp4',
  45             'title': 'Sneak Peek: Language Arts',
  46             'description': 'md5:7dcdab3b2d17e5217c953256af964e9c',
  47         },
  48         'params': {
  49             # m3u8 download
  50             'skip_download': True,
  51         },
  52     }, {
  53         'url': 'http://abc.go.com/shows/after-paradise/video/most-recent/vdka3335601',
  54         'only_matching': True,
  55     }]
  56
  57     def _real_extract(self, url):
  58         sub_domain, video_id, display_id = re.match(self._VALID_URL, url).groups()
  59         if not video_id:
  60             webpage = self._download_webpage(url, display_id)
  61             video_id = self._search_regex(
  62                 # There may be inner quotes, e.g. data-video-id="'VDKA3609139'"
  63                 # from http://freeform.go.com/shows/shadowhunters/episodes/season-2/1-this-guilty-blood
  64                 r'data-video-id=["\']*VDKA(\w+)', webpage, 'video id')
  65         site_info = self._SITE_INFO[sub_domain]
  66         brand = site_info['brand']
  67         video_data = self._download_json(
  68             'http://api.contents.watchabc.go.com/vp2/ws/contents/3000/videos/%s/001/-1/-1/-1/%s/-1/-1.json' % (brand, video_id),
  69             video_id)['video'][0]
  70         title = video_data['title']
  71
  72         formats = []
  73         for asset in video_data.get('assets', {}).get('asset', []):
  74             asset_url = asset.get('value')
  75             if not asset_url:
  76                 continue
  77             format_id = asset.get('format')
  78             ext = determine_ext(asset_url)
  79             if ext == 'm3u8':
  80                 video_type = video_data.get('type')
  81                 data = {
  82                     'video_id': video_data['id'],
  83                     'video_type': video_type,
  84                     'brand': brand,
  85                     'device': '001',
  86                 }
  87                 if video_data.get('accesslevel') == '1':
  88                     requestor_id = site_info['requestor_id']
  89                     resource = self._get_mvpd_resource(
  90                         requestor_id, title, video_id, None)
  91                     auth = self._extract_mvpd_auth(
  92                         url, video_id, requestor_id, resource)
  93                     data.update({
  94                         'token': auth,
  95                         'token_type': 'ap',
  96                         'adobe_requestor_id': requestor_id,
  97                     })
  98                 else:
  99                     self._initialize_geo_bypass(['US'])
 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=['US'])
 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                 f = {
 116                     'format_id': format_id,
 117                     'url': asset_url,
 118                     'ext': ext,
 119                 }
 120                 if re.search(r'(?:/mp4/source/|_source\.mp4)', asset_url):
 121                     f.update({
 122                         'format_id': ('%s-' % format_id if format_id else '') + 'SOURCE',
 123                         'preference': 1,
 124                     })
 125                 else:
 126                     mobj = re.search(r'/(\d+)x(\d+)/', asset_url)
 127                     if mobj:
 128                         height = int(mobj.group(2))
 129                         f.update({
 130                             'format_id': ('%s-' % format_id if format_id else '') + '%dP' % height,
 131                             'width': int(mobj.group(1)),
 132                             'height': height,
 133                         })
 134                 formats.append(f)
 135         self._sort_formats(formats)
 136
 137         subtitles = {}
 138         for cc in video_data.get('closedcaption', {}).get('src', []):
 139             cc_url = cc.get('value')
 140             if not cc_url:
 141                 continue
 142             ext = determine_ext(cc_url)
 143             if ext == 'xml':
 144                 ext = 'ttml'
 145             subtitles.setdefault(cc.get('lang'), []).append({
 146                 'url': cc_url,
 147                 'ext': ext,
 148             })
 149
 150         thumbnails = []
 151         for thumbnail in video_data.get('thumbnails', {}).get('thumbnail', []):
 152             thumbnail_url = thumbnail.get('value')
 153             if not thumbnail_url:
 154                 continue
 155             thumbnails.append({
 156                 'url': thumbnail_url,
 157                 'width': int_or_none(thumbnail.get('width')),
 158                 'height': int_or_none(thumbnail.get('height')),
 159             })
 160
 161         return {
 162             'id': video_id,
 163             'title': title,
 164             'description': video_data.get('longdescription') or video_data.get('description'),
 165             'duration': int_or_none(video_data.get('duration', {}).get('value'), 1000),
 166             'age_limit': parse_age_limit(video_data.get('tvrating', {}).get('rating')),
 167             'episode_number': int_or_none(video_data.get('episodenumber')),
 168             'series': video_data.get('show', {}).get('title'),
 169             'season_number': int_or_none(video_data.get('season', {}).get('num')),
 170             'thumbnails': thumbnails,
 171             'formats': formats,
 172             'subtitles': subtitles,
 173         }