From 70a2829fee4203ebeb399481304d289ff92adf29 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Tue, 16 Aug 2016 21:17:52 +0700 Subject: [PATCH] [xvideos] Fix HLS extraction (Closes #10356) --- youtube_dl/extractor/xvideos.py | 34 ++++++++++++++++----------------- 1 file changed, 17 insertions(+), 17 deletions(-) diff --git a/youtube_dl/extractor/xvideos.py b/youtube_dl/extractor/xvideos.py index 1dfe031ca..b2ef15119 100644 --- a/youtube_dl/extractor/xvideos.py +++ b/youtube_dl/extractor/xvideos.py @@ -42,24 +42,24 @@ def _real_extract(self, url): video_url = compat_urllib_parse_unquote(self._search_regex( r'flv_url=(.+?)&', webpage, 'video URL', default='')) if video_url: - formats.append({'url': video_url}) + formats.append({ + 'url': video_url, + 'format_id': 'flv', + }) - player_args = self._search_regex( - r'(?s)new\s+HTML5Player\((.+?)\)', webpage, ' html5 player', default=None) - if player_args: - for arg in player_args.split(','): - format_url = self._search_regex( - r'(["\'])(?Phttps?://.+?)\1', arg, 'url', - default=None, group='url') - if not format_url: - continue - ext = determine_ext(format_url) - if ext == 'mp4': - formats.append({'url': format_url}) - elif ext == 'm3u8': - formats.extend(self._extract_m3u8_formats( - format_url, video_id, 'mp4', - entry_protocol='m3u8_native', m3u8_id='hls', fatal=False)) + for kind, _, format_url in re.findall( + r'setVideo([^(]+)\((["\'])(http.+?)\2\)', webpage): + format_id = kind.lower() + if format_id == 'hls': + formats.extend(self._extract_m3u8_formats( + format_url, video_id, 'mp4', + entry_protocol='m3u8_native', m3u8_id='hls', fatal=False)) + elif format_id in ('urllow', 'urlhigh'): + formats.append({ + 'url': format_url, + 'format_id': '%s-%s' % (determine_ext(format_url, 'mp4'), format_id[3:]), + 'quality': -2 if format_id.endswith('low') else None, + }) self._sort_formats(formats)