_ 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 ..utils import (
   8     int_or_none,
   9     parse_iso8601,
  10 )
  11
  12
  13 class Go90IE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:www\.)?go90\.com/videos/(?P<id>[0-9a-zA-Z]+)'
  15     _TEST = {
  16         'url': 'https://www.go90.com/videos/84BUqjLpf9D',
  17         'md5': 'efa7670dbbbf21a7b07b360652b24a32',
  18         'info_dict': {
  19             'id': '84BUqjLpf9D',
  20             'ext': 'mp4',
  21             'title': 'Inside The Utah Coalition Against Pornography Convention',
  22             'description': 'VICE\'s Karley Sciortino meets with activists who discuss the state\'s strong anti-porn stance. Then, VICE Sports explains NFL contracts.',
  23             'timestamp': 1491868800,
  24             'upload_date': '20170411',
  25         }
  26     }
  27
  28     def _real_extract(self, url):
  29         video_id = self._match_id(url)
  30         video_data = self._download_json(
  31             'https://www.go90.com/api/view/items/' + video_id,
  32             video_id, headers={
  33                 'Content-Type': 'application/json; charset=utf-8',
  34             }, data=b'{"client":"web","device_type":"pc"}')
  35         title = video_data['title']
  36         main_video_asset = video_data['main_video_asset']
  37
  38         thumbnails = []
  39         formats = []
  40         for asset in video_data.get('assets'):
  41             if asset.get('id') == main_video_asset:
  42                 for source in asset.get('sources', []):
  43                     source_location = source.get('location')
  44                     if not source_location:
  45                         continue
  46                     source_type = source.get('type')
  47                     if source_type == 'hls':
  48                         m3u8_formats = self._extract_m3u8_formats(
  49                             source_location, video_id, 'mp4',
  50                             'm3u8_native', m3u8_id='hls', fatal=False)
  51                         for f in m3u8_formats:
  52                             mobj = re.search(r'/hls-(\d+)-(\d+)K', f['url'])
  53                             if mobj:
  54                                 height, tbr = mobj.groups()
  55                                 height = int_or_none(height)
  56                                 f.update({
  57                                     'height': f.get('height') or height,
  58                                     'width': f.get('width') or int_or_none(height / 9.0 * 16.0 if height else None),
  59                                     'tbr': f.get('tbr') or int_or_none(tbr),
  60                                 })
  61                         formats.extend(m3u8_formats)
  62                     elif source_type == 'dash':
  63                         formats.extend(self._extract_mpd_formats(
  64                             source_location, video_id, mpd_id='dash', fatal=False))
  65                     else:
  66                         formats.append({
  67                             'format_id': source.get('name'),
  68                             'url': source_location,
  69                             'width': int_or_none(source.get('width')),
  70                             'height': int_or_none(source.get('height')),
  71                             'tbr': int_or_none(source.get('bitrate')),
  72                         })
  73             elif asset.get('type') == 'image':
  74                 asset_location = asset.get('location')
  75                 if not asset_location:
  76                     continue
  77                 thumbnails.append({
  78                     'url': asset_location,
  79                     'width': int_or_none(asset.get('width')),
  80                     'height': int_or_none(asset.get('height')),
  81                 })
  82         self._sort_formats(formats)
  83
  84         return {
  85             'id': video_id,
  86             'title': title,
  87             'formats': formats,
  88             'thumbnails': thumbnails,
  89             'description': video_data.get('short_description'),
  90             'like_count': int_or_none(video_data.get('like_count')),
  91             'timestamp': parse_iso8601(video_data.get('released_at')),
  92         }