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