_ 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 .common import InfoExtractor
   7 from ..utils import (
   8     int_or_none,
   9     determine_ext,
  10     parse_age_limit,
  11 )
  12
  13
  14 class GoIE(InfoExtractor):
  15     _BRANDS = {
  16         'abc': '001',
  17         'freeform': '002',
  18         'watchdisneychannel': '004',
  19         'watchdisneyjunior': '008',
  20         'watchdisneyxd': '009',
  21     }
  22     _VALID_URL = r'https?://(?:(?P<sub_domain>%s)\.)?go\.com/.*?vdka(?P<id>\w+)' % '|'.join(_BRANDS.keys())
  23     _TESTS = [{
  24         'url': 'http://abc.go.com/shows/castle/video/most-recent/vdka0_g86w5onx',
  25         'info_dict': {
  26             'id': '0_g86w5onx',
  27             'ext': 'mp4',
  28             'title': 'Sneak Peek: Language Arts',
  29             'description': 'md5:7dcdab3b2d17e5217c953256af964e9c',
  30         },
  31         'params': {
  32             # m3u8 download
  33             'skip_download': True,
  34         },
  35     }, {
  36         'url': 'http://abc.go.com/shows/after-paradise/video/most-recent/vdka3335601',
  37         'only_matching': True,
  38     }]
  39
  40     def _real_extract(self, url):
  41         sub_domain, video_id = re.match(self._VALID_URL, url).groups()
  42         video_data = self._download_json(
  43             'http://api.contents.watchabc.go.com/vp2/ws/contents/3000/videos/%s/001/-1/-1/-1/%s/-1/-1.json' % (self._BRANDS[sub_domain], video_id),
  44             video_id)['video'][0]
  45         title = video_data['title']
  46
  47         formats = []
  48         for asset in video_data.get('assets', {}).get('asset', []):
  49             asset_url = asset.get('value')
  50             if not asset_url:
  51                 continue
  52             format_id = asset.get('format')
  53             ext = determine_ext(asset_url)
  54             if ext == 'm3u8':
  55                 formats.extend(self._extract_m3u8_formats(
  56                     asset_url, video_id, 'mp4', m3u8_id=format_id or 'hls', fatal=False))
  57             else:
  58                 formats.append({
  59                     'format_id': format_id,
  60                     'url': asset_url,
  61                     'ext': ext,
  62                 })
  63         self._sort_formats(formats)
  64
  65         subtitles = {}
  66         for cc in video_data.get('closedcaption', {}).get('src', []):
  67             cc_url = cc.get('value')
  68             if not cc_url:
  69                 continue
  70             ext = determine_ext(cc_url)
  71             if ext == 'xml':
  72                 ext = 'ttml'
  73             subtitles.setdefault(cc.get('lang'), []).append({
  74                 'url': cc_url,
  75                 'ext': ext,
  76             })
  77
  78         thumbnails = []
  79         for thumbnail in video_data.get('thumbnails', {}).get('thumbnail', []):
  80             thumbnail_url = thumbnail.get('value')
  81             if not thumbnail_url:
  82                 continue
  83             thumbnails.append({
  84                 'url': thumbnail_url,
  85                 'width': int_or_none(thumbnail.get('width')),
  86                 'height': int_or_none(thumbnail.get('height')),
  87             })
  88
  89         return {
  90             'id': video_id,
  91             'title': title,
  92             'description': video_data.get('longdescription') or video_data.get('description'),
  93             'duration': int_or_none(video_data.get('duration', {}).get('value'), 1000),
  94             'age_limit': parse_age_limit(video_data.get('tvrating', {}).get('rating')),
  95             'episode_number': int_or_none(video_data.get('episodenumber')),
  96             'series': video_data.get('show', {}).get('title'),
  97             'season_number': int_or_none(video_data.get('season', {}).get('num')),
  98             'thumbnails': thumbnails,
  99             'formats': formats,
 100             'subtitles': subtitles,
 101         }