_ Git - youtube-dl/blob - 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/(?P<id>[0-9a-zA-Z]+)'
  19     _TEST = {
  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     _GEO_BYPASS = False
  33
  34     def _real_extract(self, url):
  35         video_id = self._match_id(url)
  36
  37         try:
  38             headers = self.geo_verification_headers()
  39             headers.update({
  40                 'Content-Type': 'application/json; charset=utf-8',
  41             })
  42             video_data = self._download_json(
  43                 'https://www.go90.com/api/view/items/' + video_id, video_id,
  44                 headers=headers, data=b'{"client":"web","device_type":"pc"}')
  45         except ExtractorError as e:
  46             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400:
  47                 message = self._parse_json(e.cause.read().decode(), None)['error']['message']
  48                 if 'region unavailable' in message:
  49                     self.raise_geo_restricted(countries=['US'])
  50                 raise ExtractorError(message, expected=True)
  51             raise
  52
  53         if video_data.get('requires_drm'):
  54             raise ExtractorError('This video is DRM protected.', expected=True)
  55         main_video_asset = video_data['main_video_asset']
  56
  57         episode_number = int_or_none(video_data.get('episode_number'))
  58         series = None
  59         season = None
  60         season_id = None
  61         season_number = None
  62         for metadata in video_data.get('__children', {}).get('Item', {}).values():
  63             if metadata.get('type') == 'show':
  64                 series = metadata.get('title')
  65             elif metadata.get('type') == 'season':
  66                 season = metadata.get('title')
  67                 season_id = metadata.get('id')
  68                 season_number = int_or_none(metadata.get('season_number'))
  69
  70         title = episode = video_data.get('title') or series
  71         if series and series != title:
  72             title = '%s - %s' % (series, title)
  73
  74         thumbnails = []
  75         formats = []
  76         subtitles = {}
  77         for asset in video_data.get('assets'):
  78             if asset.get('id') == main_video_asset:
  79                 for source in asset.get('sources', []):
  80                     source_location = source.get('location')
  81                     if not source_location:
  82                         continue
  83                     source_type = source.get('type')
  84                     if source_type == 'hls':
  85                         m3u8_formats = self._extract_m3u8_formats(
  86                             source_location, video_id, 'mp4',
  87                             'm3u8_native', m3u8_id='hls', fatal=False)
  88                         for f in m3u8_formats:
  89                             mobj = re.search(r'/hls-(\d+)-(\d+)K', f['url'])
  90                             if mobj:
  91                                 height, tbr = mobj.groups()
  92                                 height = int_or_none(height)
  93                                 f.update({
  94                                     'height': f.get('height') or height,
  95                                     'width': f.get('width') or int_or_none(height / 9.0 * 16.0 if height else None),
  96                                     'tbr': f.get('tbr') or int_or_none(tbr),
  97                                 })
  98                         formats.extend(m3u8_formats)
  99                     elif source_type == 'dash':
 100                         formats.extend(self._extract_mpd_formats(
 101                             source_location, video_id, mpd_id='dash', fatal=False))
 102                     else:
 103                         formats.append({
 104                             'format_id': source.get('name'),
 105                             'url': source_location,
 106                             'width': int_or_none(source.get('width')),
 107                             'height': int_or_none(source.get('height')),
 108                             'tbr': int_or_none(source.get('bitrate')),
 109                         })
 110
 111                 for caption in asset.get('caption_metadata', []):
 112                     caption_url = caption.get('source_url')
 113                     if not caption_url:
 114                         continue
 115                     subtitles.setdefault(caption.get('language', 'en'), []).append({
 116                         'url': caption_url,
 117                         'ext': determine_ext(caption_url, 'vtt'),
 118                     })
 119             elif asset.get('type') == 'image':
 120                 asset_location = asset.get('location')
 121                 if not asset_location:
 122                     continue
 123                 thumbnails.append({
 124                     'url': asset_location,
 125                     'width': int_or_none(asset.get('width')),
 126                     'height': int_or_none(asset.get('height')),
 127                 })
 128         self._sort_formats(formats)
 129
 130         return {
 131             'id': video_id,
 132             'title': title,
 133             'formats': formats,
 134             'thumbnails': thumbnails,
 135             'description': video_data.get('short_description'),
 136             'like_count': int_or_none(video_data.get('like_count')),
 137             'timestamp': parse_iso8601(video_data.get('released_at')),
 138             'series': series,
 139             'episode': episode,
 140             'season': season,
 141             'season_id': season_id,
 142             'season_number': season_number,
 143             'episode_number': episode_number,
 144             'subtitles': subtitles,
 145             'age_limit': parse_age_limit(video_data.get('rating')),
 146         }