X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/453698570f26bebd37b39df8537d993b57d77b8b..97a8fc3ae80fb363c69c2e6b8c29b5373ac72aea:/youtube_dl/extractor/hitbox.py diff --git a/youtube_dl/extractor/hitbox.py b/youtube_dl/extractor/hitbox.py index 84bd7c0..e21ebb8 100644 --- a/youtube_dl/extractor/hitbox.py +++ b/youtube_dl/extractor/hitbox.py @@ -10,6 +10,7 @@ from ..utils import ( float_or_none, int_or_none, compat_str, + determine_ext, ) @@ -24,7 +25,7 @@ class HitboxIE(InfoExtractor): 'alt_title': 'hitboxlive - Aug 9th #6', 'description': '', 'ext': 'mp4', - 'thumbnail': 're:^https?://.*\.jpg$', + 'thumbnail': r're:^https?://.*\.jpg$', 'duration': 215.1666, 'resolution': 'HD 720p', 'uploader': 'hitboxlive', @@ -42,7 +43,8 @@ class HitboxIE(InfoExtractor): def _extract_metadata(self, url, video_id): thumb_base = 'https://edge.sf.hitbox.tv' metadata = self._download_json( - '%s/%s' % (url, video_id), video_id) + '%s/%s' % (url, video_id), video_id, + 'Downloading metadata JSON') date = 'media_live_since' media_type = 'livestream' @@ -87,21 +89,41 @@ class HitboxIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - metadata = self._extract_metadata( - 'https://www.hitbox.tv/api/media/video', - video_id) - player_config = self._download_json( 'https://www.hitbox.tv/api/player/config/video/%s' % video_id, - video_id) + video_id, 'Downloading video JSON') - clip = player_config.get('clip') - video_url = clip.get('url') - res = clip.get('bitrates', [])[0].get('label') + formats = [] + for video in player_config['clip']['bitrates']: + label = video.get('label') + if label == 'Auto': + continue + video_url = video.get('url') + if not video_url: + continue + bitrate = int_or_none(video.get('bitrate')) + if determine_ext(video_url) == 'm3u8': + if not video_url.startswith('http'): + continue + formats.append({ + 'url': video_url, + 'ext': 'mp4', + 'tbr': bitrate, + 'format_note': label, + 'protocol': 'm3u8_native', + }) + else: + formats.append({ + 'url': video_url, + 'tbr': bitrate, + 'format_note': label, + }) + self._sort_formats(formats) - metadata['resolution'] = res - metadata['url'] = video_url - metadata['protocol'] = 'm3u8' + metadata = self._extract_metadata( + 'https://www.hitbox.tv/api/media/video', + video_id) + metadata['formats'] = formats return metadata @@ -129,10 +151,6 @@ class HitboxLiveIE(HitboxIE): def _real_extract(self, url): video_id = self._match_id(url) - metadata = self._extract_metadata( - 'https://www.hitbox.tv/api/media/live', - video_id) - player_config = self._download_json( 'https://www.hitbox.tv/api/player/config/live/%s' % video_id, video_id) @@ -141,26 +159,48 @@ class HitboxLiveIE(HitboxIE): cdns = player_config.get('cdns') servers = [] for cdn in cdns: + # Subscribe URLs are not playable + if cdn.get('rtmpSubscribe') is True: + continue base_url = cdn.get('netConnectionUrl') - host = re.search('.+\.([^\.]+\.[^\./]+)/.+', base_url).group(1) + host = re.search(r'.+\.([^\.]+\.[^\./]+)/.+', base_url).group(1) if base_url not in servers: servers.append(base_url) for stream in cdn.get('bitrates'): label = stream.get('label') - if label != 'Auto': + if label == 'Auto': + continue + stream_url = stream.get('url') + if not stream_url: + continue + bitrate = int_or_none(stream.get('bitrate')) + if stream.get('provider') == 'hls' or determine_ext(stream_url) == 'm3u8': + if not stream_url.startswith('http'): + continue + formats.append({ + 'url': stream_url, + 'ext': 'mp4', + 'tbr': bitrate, + 'format_note': label, + 'rtmp_live': True, + }) + else: formats.append({ - 'url': '%s/%s' % (base_url, stream.get('url')), + 'url': '%s/%s' % (base_url, stream_url), 'ext': 'mp4', - 'vbr': stream.get('bitrate'), - 'resolution': label, + 'tbr': bitrate, 'rtmp_live': True, 'format_note': host, 'page_url': url, 'player_url': 'http://www.hitbox.tv/static/player/flowplayer/flowplayer.commercial-3.2.16.swf', }) - self._sort_formats(formats) + + metadata = self._extract_metadata( + 'https://www.hitbox.tv/api/media/live', + video_id) metadata['formats'] = formats metadata['is_live'] = True metadata['title'] = self._live_title(metadata.get('title')) + return metadata