youtube_dl/extractor/go90.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_HTTPError
   8 from ..utils import (
   9     determine_ext,
  10     ExtractorError,
  11     int_or_none,
  12     parse_age_limit,
  13     parse_iso8601,
  14 )
  15
  16
  17 class Go90IE(InfoExtractor):
  18     _VALID_URL = r'https?://(?:www\.)?go90\.com/(?:videos|embed)/(?P<id>[0-9a-zA-Z]+)'
  19     _TESTS = [{
  20         'url': 'https://www.go90.com/videos/84BUqjLpf9D',
  21         'md5': 'efa7670dbbbf21a7b07b360652b24a32',
  22         'info_dict': {
  23             'id': '84BUqjLpf9D',
  24             'ext': 'mp4',
  25             'title': 'Daily VICE - Inside The Utah Coalition Against Pornography Convention',
  26             'description': 'VICE\'s Karley Sciortino meets with activists who discuss the state\'s strong anti-porn stance. Then, VICE Sports explains NFL contracts.',
  27             'timestamp': 1491868800,
  28             'upload_date': '20170411',
  29             'age_limit': 14,
  30         }
  31     }, {
  32         'url': 'https://www.go90.com/embed/261MflWkD3N',
  33         'only_matching': True,
  34     }]
  35     _GEO_BYPASS = False
  36
  37     def _real_extract(self, url):
  38         video_id = self._match_id(url)
  39
  40         try:
  41             headers = self.geo_verification_headers()
  42             headers.update({
  43                 'Content-Type': 'application/json; charset=utf-8',
  44             })
  45             video_data = self._download_json(
  46                 'https://www.go90.com/api/view/items/' + video_id, video_id,
  47                 headers=headers, data=b'{"client":"web","device_type":"pc"}')
  48         except ExtractorError as e:
  49             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400:
  50                 message = self._parse_json(e.cause.read().decode(), None)['error']['message']
  51                 if 'region unavailable' in message:
  52                     self.raise_geo_restricted(countries=['US'])
  53                 raise ExtractorError(message, expected=True)
  54             raise
  55
  56         if video_data.get('requires_drm'):
  57             raise ExtractorError('This video is DRM protected.', expected=True)
  58         main_video_asset = video_data['main_video_asset']
  59
  60         episode_number = int_or_none(video_data.get('episode_number'))
  61         series = None
  62         season = None
  63         season_id = None
  64         season_number = None
  65         for metadata in video_data.get('__children', {}).get('Item', {}).values():
  66             if metadata.get('type') == 'show':
  67                 series = metadata.get('title')
  68             elif metadata.get('type') == 'season':
  69                 season = metadata.get('title')
  70                 season_id = metadata.get('id')
  71                 season_number = int_or_none(metadata.get('season_number'))
  72
  73         title = episode = video_data.get('title') or series
  74         if series and series != title:
  75             title = '%s - %s' % (series, title)
  76
  77         thumbnails = []
  78         formats = []
  79         subtitles = {}
  80         for asset in video_data.get('assets'):
  81             if asset.get('id') == main_video_asset:
  82                 for source in asset.get('sources', []):
  83                     source_location = source.get('location')
  84                     if not source_location:
  85                         continue
  86                     source_type = source.get('type')
  87                     if source_type == 'hls':
  88                         m3u8_formats = self._extract_m3u8_formats(
  89                             source_location, video_id, 'mp4',
  90                             'm3u8_native', m3u8_id='hls', fatal=False)
  91                         for f in m3u8_formats:
  92                             mobj = re.search(r'/hls-(\d+)-(\d+)K', f['url'])
  93                             if mobj:
  94                                 height, tbr = mobj.groups()
  95                                 height = int_or_none(height)
  96                                 f.update({
  97                                     'height': f.get('height') or height,
  98                                     'width': f.get('width') or int_or_none(height / 9.0 * 16.0 if height else None),
  99                                     'tbr': f.get('tbr') or int_or_none(tbr),
 100                                 })
 101                         formats.extend(m3u8_formats)
 102                     elif source_type == 'dash':
 103                         formats.extend(self._extract_mpd_formats(
 104                             source_location, video_id, mpd_id='dash', fatal=False))
 105                     else:
 106                         formats.append({
 107                             'format_id': source.get('name'),
 108                             'url': source_location,
 109                             'width': int_or_none(source.get('width')),
 110                             'height': int_or_none(source.get('height')),
 111                             'tbr': int_or_none(source.get('bitrate')),
 112                         })
 113
 114                 for caption in asset.get('caption_metadata', []):
 115                     caption_url = caption.get('source_url')
 116                     if not caption_url:
 117                         continue
 118                     subtitles.setdefault(caption.get('language', 'en'), []).append({
 119                         'url': caption_url,
 120                         'ext': determine_ext(caption_url, 'vtt'),
 121                     })
 122             elif asset.get('type') == 'image':
 123                 asset_location = asset.get('location')
 124                 if not asset_location:
 125                     continue
 126                 thumbnails.append({
 127                     'url': asset_location,
 128                     'width': int_or_none(asset.get('width')),
 129                     'height': int_or_none(asset.get('height')),
 130                 })
 131         self._sort_formats(formats)
 132
 133         return {
 134             'id': video_id,
 135             'title': title,
 136             'formats': formats,
 137             'thumbnails': thumbnails,
 138             'description': video_data.get('short_description'),
 139             'like_count': int_or_none(video_data.get('like_count')),
 140             'timestamp': parse_iso8601(video_data.get('released_at')),
 141             'series': series,
 142             'episode': episode,
 143             'season': season,
 144             'season_id': season_id,
 145             'season_number': season_number,
 146             'episode_number': episode_number,
 147             'subtitles': subtitles,
 148             'age_limit': parse_age_limit(video_data.get('rating')),
 149         }