Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/hitbox.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     clean_html,
   9     parse_iso8601,
  10     float_or_none,
  11     int_or_none,
  12     compat_str,
  13     determine_ext,
  14 )
  15
  16
  17 class HitboxIE(InfoExtractor):
  18     IE_NAME = 'hitbox'
  19     _VALID_URL = r'https?://(?:www\.)?hitbox\.tv/video/(?P<id>[0-9]+)'
  20     _TEST = {
  21         'url': 'http://www.hitbox.tv/video/203213',
  22         'info_dict': {
  23             'id': '203213',
  24             'title': 'hitbox @ gamescom, Sub Button Hype extended, Giveaway - hitbox News Update with Oxy',
  25             'alt_title': 'hitboxlive - Aug 9th #6',
  26             'description': '',
  27             'ext': 'mp4',
  28             'thumbnail': 're:^https?://.*\.jpg$',
  29             'duration': 215.1666,
  30             'resolution': 'HD 720p',
  31             'uploader': 'hitboxlive',
  32             'view_count': int,
  33             'timestamp': 1407576133,
  34             'upload_date': '20140809',
  35             'categories': ['Live Show'],
  36         },
  37         'params': {
  38             # m3u8 download
  39             'skip_download': True,
  40         },
  41     }
  42
  43     def _extract_metadata(self, url, video_id):
  44         thumb_base = 'https://edge.sf.hitbox.tv'
  45         metadata = self._download_json(
  46             '%s/%s' % (url, video_id), video_id,
  47             'Downloading metadata JSON')
  48
  49         date = 'media_live_since'
  50         media_type = 'livestream'
  51         if metadata.get('media_type') == 'video':
  52             media_type = 'video'
  53             date = 'media_date_added'
  54
  55         video_meta = metadata.get(media_type, [])[0]
  56         title = video_meta.get('media_status')
  57         alt_title = video_meta.get('media_title')
  58         description = clean_html(
  59             video_meta.get('media_description') or
  60             video_meta.get('media_description_md'))
  61         duration = float_or_none(video_meta.get('media_duration'))
  62         uploader = video_meta.get('media_user_name')
  63         views = int_or_none(video_meta.get('media_views'))
  64         timestamp = parse_iso8601(video_meta.get(date), ' ')
  65         categories = [video_meta.get('category_name')]
  66         thumbs = [
  67             {'url': thumb_base + video_meta.get('media_thumbnail'),
  68              'width': 320,
  69              'height': 180},
  70             {'url': thumb_base + video_meta.get('media_thumbnail_large'),
  71              'width': 768,
  72              'height': 432},
  73         ]
  74
  75         return {
  76             'id': video_id,
  77             'title': title,
  78             'alt_title': alt_title,
  79             'description': description,
  80             'ext': 'mp4',
  81             'thumbnails': thumbs,
  82             'duration': duration,
  83             'uploader': uploader,
  84             'view_count': views,
  85             'timestamp': timestamp,
  86             'categories': categories,
  87         }
  88
  89     def _real_extract(self, url):
  90         video_id = self._match_id(url)
  91
  92         player_config = self._download_json(
  93             'https://www.hitbox.tv/api/player/config/video/%s' % video_id,
  94             video_id, 'Downloading video JSON')
  95
  96         formats = []
  97         for video in player_config['clip']['bitrates']:
  98             label = video.get('label')
  99             if label == 'Auto':
 100                 continue
 101             video_url = video.get('url')
 102             if not video_url:
 103                 continue
 104             bitrate = int_or_none(video.get('bitrate'))
 105             if determine_ext(video_url) == 'm3u8':
 106                 if not video_url.startswith('http'):
 107                     continue
 108                 formats.append({
 109                     'url': video_url,
 110                     'ext': 'mp4',
 111                     'tbr': bitrate,
 112                     'format_note': label,
 113                     'protocol': 'm3u8_native',
 114                 })
 115             else:
 116                 formats.append({
 117                     'url': video_url,
 118                     'tbr': bitrate,
 119                     'format_note': label,
 120                 })
 121         self._sort_formats(formats)
 122
 123         metadata = self._extract_metadata(
 124             'https://www.hitbox.tv/api/media/video',
 125             video_id)
 126         metadata['formats'] = formats
 127
 128         return metadata
 129
 130
 131 class HitboxLiveIE(HitboxIE):
 132     IE_NAME = 'hitbox:live'
 133     _VALID_URL = r'https?://(?:www\.)?hitbox\.tv/(?!video)(?P<id>.+)'
 134     _TEST = {
 135         'url': 'http://www.hitbox.tv/dimak',
 136         'info_dict': {
 137             'id': 'dimak',
 138             'ext': 'mp4',
 139             'description': 'md5:c9f80fa4410bc588d7faa40003fc7d0e',
 140             'timestamp': int,
 141             'upload_date': compat_str,
 142             'title': compat_str,
 143             'uploader': 'Dimak',
 144         },
 145         'params': {
 146             # live
 147             'skip_download': True,
 148         },
 149     }
 150
 151     def _real_extract(self, url):
 152         video_id = self._match_id(url)
 153
 154         player_config = self._download_json(
 155             'https://www.hitbox.tv/api/player/config/live/%s' % video_id,
 156             video_id)
 157
 158         formats = []
 159         cdns = player_config.get('cdns')
 160         servers = []
 161         for cdn in cdns:
 162             # Subscribe URLs are not playable
 163             if cdn.get('rtmpSubscribe') is True:
 164                 continue
 165             base_url = cdn.get('netConnectionUrl')
 166             host = re.search('.+\.([^\.]+\.[^\./]+)/.+', base_url).group(1)
 167             if base_url not in servers:
 168                 servers.append(base_url)
 169                 for stream in cdn.get('bitrates'):
 170                     label = stream.get('label')
 171                     if label == 'Auto':
 172                         continue
 173                     stream_url = stream.get('url')
 174                     if not stream_url:
 175                         continue
 176                     bitrate = int_or_none(stream.get('bitrate'))
 177                     if stream.get('provider') == 'hls' or determine_ext(stream_url) == 'm3u8':
 178                         if not stream_url.startswith('http'):
 179                             continue
 180                         formats.append({
 181                             'url': stream_url,
 182                             'ext': 'mp4',
 183                             'tbr': bitrate,
 184                             'format_note': label,
 185                             'rtmp_live': True,
 186                         })
 187                     else:
 188                         formats.append({
 189                             'url': '%s/%s' % (base_url, stream_url),
 190                             'ext': 'mp4',
 191                             'tbr': bitrate,
 192                             'rtmp_live': True,
 193                             'format_note': host,
 194                             'page_url': url,
 195                             'player_url': 'http://www.hitbox.tv/static/player/flowplayer/flowplayer.commercial-3.2.16.swf',
 196                         })
 197         self._sort_formats(formats)
 198
 199         metadata = self._extract_metadata(
 200             'https://www.hitbox.tv/api/media/live',
 201             video_id)
 202         metadata['formats'] = formats
 203         metadata['is_live'] = True
 204         metadata['title'] = self._live_title(metadata.get('title'))
 205
 206         return metadata