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|smashcast)\.tv/(?:[^/]+/)*videos?/(?P<id>[0-9]+)'
  20     _TESTS = [{
  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': r'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         'url': 'https://www.smashcast.tv/hitboxlive/videos/203213',
  43         'only_matching': True,
  44     }]
  45
  46     def _extract_metadata(self, url, video_id):
  47         thumb_base = 'https://edge.sf.hitbox.tv'
  48         metadata = self._download_json(
  49             '%s/%s' % (url, video_id), video_id, 'Downloading metadata JSON')
  50
  51         date = 'media_live_since'
  52         media_type = 'livestream'
  53         if metadata.get('media_type') == 'video':
  54             media_type = 'video'
  55             date = 'media_date_added'
  56
  57         video_meta = metadata.get(media_type, [])[0]
  58         title = video_meta.get('media_status')
  59         alt_title = video_meta.get('media_title')
  60         description = clean_html(
  61             video_meta.get('media_description') or
  62             video_meta.get('media_description_md'))
  63         duration = float_or_none(video_meta.get('media_duration'))
  64         uploader = video_meta.get('media_user_name')
  65         views = int_or_none(video_meta.get('media_views'))
  66         timestamp = parse_iso8601(video_meta.get(date), ' ')
  67         categories = [video_meta.get('category_name')]
  68         thumbs = [{
  69             'url': thumb_base + video_meta.get('media_thumbnail'),
  70             'width': 320,
  71             'height': 180
  72         }, {
  73             'url': thumb_base + video_meta.get('media_thumbnail_large'),
  74             'width': 768,
  75             'height': 432
  76         }]
  77
  78         return {
  79             'id': video_id,
  80             'title': title,
  81             'alt_title': alt_title,
  82             'description': description,
  83             'ext': 'mp4',
  84             'thumbnails': thumbs,
  85             'duration': duration,
  86             'uploader': uploader,
  87             'view_count': views,
  88             'timestamp': timestamp,
  89             'categories': categories,
  90         }
  91
  92     def _real_extract(self, url):
  93         video_id = self._match_id(url)
  94
  95         player_config = self._download_json(
  96             'https://www.smashcast.tv/api/player/config/video/%s' % video_id,
  97             video_id, 'Downloading video JSON')
  98
  99         formats = []
 100         for video in player_config['clip']['bitrates']:
 101             label = video.get('label')
 102             if label == 'Auto':
 103                 continue
 104             video_url = video.get('url')
 105             if not video_url:
 106                 continue
 107             bitrate = int_or_none(video.get('bitrate'))
 108             if determine_ext(video_url) == 'm3u8':
 109                 if not video_url.startswith('http'):
 110                     continue
 111                 formats.append({
 112                     'url': video_url,
 113                     'ext': 'mp4',
 114                     'tbr': bitrate,
 115                     'format_note': label,
 116                     'protocol': 'm3u8_native',
 117                 })
 118             else:
 119                 formats.append({
 120                     'url': video_url,
 121                     'tbr': bitrate,
 122                     'format_note': label,
 123                 })
 124         self._sort_formats(formats)
 125
 126         metadata = self._extract_metadata(
 127             'https://www.smashcast.tv/api/media/video', video_id)
 128         metadata['formats'] = formats
 129
 130         return metadata
 131
 132
 133 class HitboxLiveIE(HitboxIE):
 134     IE_NAME = 'hitbox:live'
 135     _VALID_URL = r'https?://(?:www\.)?(?:hitbox|smashcast)\.tv/(?P<id>[^/?#&]+)'
 136     _TESTS = [{
 137         'url': 'http://www.hitbox.tv/dimak',
 138         'info_dict': {
 139             'id': 'dimak',
 140             'ext': 'mp4',
 141             'description': 'md5:c9f80fa4410bc588d7faa40003fc7d0e',
 142             'timestamp': int,
 143             'upload_date': compat_str,
 144             'title': compat_str,
 145             'uploader': 'Dimak',
 146         },
 147         'params': {
 148             # live
 149             'skip_download': True,
 150         },
 151     }, {
 152         'url': 'https://www.smashcast.tv/dimak',
 153         'only_matching': True,
 154     }]
 155
 156     @classmethod
 157     def suitable(cls, url):
 158         return False if HitboxIE.suitable(url) else super(HitboxLiveIE, cls).suitable(url)
 159
 160     def _real_extract(self, url):
 161         video_id = self._match_id(url)
 162
 163         player_config = self._download_json(
 164             'https://www.smashcast.tv/api/player/config/live/%s' % video_id,
 165             video_id)
 166
 167         formats = []
 168         cdns = player_config.get('cdns')
 169         servers = []
 170         for cdn in cdns:
 171             # Subscribe URLs are not playable
 172             if cdn.get('rtmpSubscribe') is True:
 173                 continue
 174             base_url = cdn.get('netConnectionUrl')
 175             host = re.search(r'.+\.([^\.]+\.[^\./]+)/.+', base_url).group(1)
 176             if base_url not in servers:
 177                 servers.append(base_url)
 178                 for stream in cdn.get('bitrates'):
 179                     label = stream.get('label')
 180                     if label == 'Auto':
 181                         continue
 182                     stream_url = stream.get('url')
 183                     if not stream_url:
 184                         continue
 185                     bitrate = int_or_none(stream.get('bitrate'))
 186                     if stream.get('provider') == 'hls' or determine_ext(stream_url) == 'm3u8':
 187                         if not stream_url.startswith('http'):
 188                             continue
 189                         formats.append({
 190                             'url': stream_url,
 191                             'ext': 'mp4',
 192                             'tbr': bitrate,
 193                             'format_note': label,
 194                             'rtmp_live': True,
 195                         })
 196                     else:
 197                         formats.append({
 198                             'url': '%s/%s' % (base_url, stream_url),
 199                             'ext': 'mp4',
 200                             'tbr': bitrate,
 201                             'rtmp_live': True,
 202                             'format_note': host,
 203                             'page_url': url,
 204                             'player_url': 'http://www.hitbox.tv/static/player/flowplayer/flowplayer.commercial-3.2.16.swf',
 205                         })
 206         self._sort_formats(formats)
 207
 208         metadata = self._extract_metadata(
 209             'https://www.smashcast.tv/api/media/live', video_id)
 210         metadata['formats'] = formats
 211         metadata['is_live'] = True
 212         metadata['title'] = self._live_title(metadata.get('title'))
 213
 214         return metadata