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 .common import InfoExtractor
   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(InfoExtractor):
  17     _BRANDS = {
  18         'abc': '001',
  19         'freeform': '002',
  20         'watchdisneychannel': '004',
  21         'watchdisneyjunior': '008',
  22         'watchdisneyxd': '009',
  23     }
  24     _VALID_URL = r'https?://(?:(?P<sub_domain>%s)\.)?go\.com/(?:[^/]+/)*(?:vdka(?P<id>\w+)|season-\d+/\d+-(?P<display_id>[^/?#]+))' % '|'.join(_BRANDS.keys())
  25     _TESTS = [{
  26         'url': 'http://abc.go.com/shows/castle/video/most-recent/vdka0_g86w5onx',
  27         'info_dict': {
  28             'id': '0_g86w5onx',
  29             'ext': 'mp4',
  30             'title': 'Sneak Peek: Language Arts',
  31             'description': 'md5:7dcdab3b2d17e5217c953256af964e9c',
  32         },
  33         'params': {
  34             # m3u8 download
  35             'skip_download': True,
  36         },
  37     }, {
  38         'url': 'http://abc.go.com/shows/after-paradise/video/most-recent/vdka3335601',
  39         'only_matching': True,
  40     }]
  41
  42     def _real_extract(self, url):
  43         sub_domain, video_id, display_id = re.match(self._VALID_URL, url).groups()
  44         if not video_id:
  45             webpage = self._download_webpage(url, display_id)
  46             video_id = self._search_regex(
  47                 # There may be inner quotes, e.g. data-video-id="'VDKA3609139'"
  48                 # from http://freeform.go.com/shows/shadowhunters/episodes/season-2/1-this-guilty-blood
  49                 r'data-video-id=["\']*VDKA(\w+)', webpage, 'video id')
  50         brand = self._BRANDS[sub_domain]
  51         video_data = self._download_json(
  52             'http://api.contents.watchabc.go.com/vp2/ws/contents/3000/videos/%s/001/-1/-1/-1/%s/-1/-1.json' % (brand, video_id),
  53             video_id)['video'][0]
  54         title = video_data['title']
  55
  56         formats = []
  57         for asset in video_data.get('assets', {}).get('asset', []):
  58             asset_url = asset.get('value')
  59             if not asset_url:
  60                 continue
  61             format_id = asset.get('format')
  62             ext = determine_ext(asset_url)
  63             if ext == 'm3u8':
  64                 video_type = video_data.get('type')
  65                 if video_type == 'lf':
  66                     entitlement = self._download_json(
  67                         'https://api.entitlement.watchabc.go.com/vp2/ws-secure/entitlement/2020/authorize.json',
  68                         video_id, data=urlencode_postdata({
  69                             'video_id': video_data['id'],
  70                             'video_type': video_type,
  71                             'brand': brand,
  72                             'device': '001',
  73                         }))
  74                     errors = entitlement.get('errors', {}).get('errors', [])
  75                     if errors:
  76                         error_message = ', '.join([error['message'] for error in errors])
  77                         raise ExtractorError('%s said: %s' % (self.IE_NAME, error_message), expected=True)
  78                     asset_url += '?' + entitlement['uplynkData']['sessionKey']
  79                 formats.extend(self._extract_m3u8_formats(
  80                     asset_url, video_id, 'mp4', m3u8_id=format_id or 'hls', fatal=False))
  81             else:
  82                 formats.append({
  83                     'format_id': format_id,
  84                     'url': asset_url,
  85                     'ext': ext,
  86                 })
  87         self._sort_formats(formats)
  88
  89         subtitles = {}
  90         for cc in video_data.get('closedcaption', {}).get('src', []):
  91             cc_url = cc.get('value')
  92             if not cc_url:
  93                 continue
  94             ext = determine_ext(cc_url)
  95             if ext == 'xml':
  96                 ext = 'ttml'
  97             subtitles.setdefault(cc.get('lang'), []).append({
  98                 'url': cc_url,
  99                 'ext': ext,
 100             })
 101
 102         thumbnails = []
 103         for thumbnail in video_data.get('thumbnails', {}).get('thumbnail', []):
 104             thumbnail_url = thumbnail.get('value')
 105             if not thumbnail_url:
 106                 continue
 107             thumbnails.append({
 108                 'url': thumbnail_url,
 109                 'width': int_or_none(thumbnail.get('width')),
 110                 'height': int_or_none(thumbnail.get('height')),
 111             })
 112
 113         return {
 114             'id': video_id,
 115             'title': title,
 116             'description': video_data.get('longdescription') or video_data.get('description'),
 117             'duration': int_or_none(video_data.get('duration', {}).get('value'), 1000),
 118             'age_limit': parse_age_limit(video_data.get('tvrating', {}).get('rating')),
 119             'episode_number': int_or_none(video_data.get('episodenumber')),
 120             'series': video_data.get('show', {}).get('title'),
 121             'season_number': int_or_none(video_data.get('season', {}).get('num')),
 122             'thumbnails': thumbnails,
 123             'formats': formats,
 124             'subtitles': subtitles,
 125         }