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