From d413095f7e14f146f66d5f0cb828fb55aa4cb323 Mon Sep 17 00:00:00 2001 From: remitamine Date: Tue, 9 Feb 2016 17:15:41 +0100 Subject: [PATCH] [extractor/common] remove duplicated formats and subtiles in smil manifests --- youtube_dl/extractor/common.py | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index 9c0421de7..1143f6dbb 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -1186,6 +1186,7 @@ def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_para http_count = 0 m3u8_count = 0 + src_urls = [] videos = smil.findall(self._xpath_ns('.//video', namespace)) for video in videos: src = video.get('src') @@ -1222,6 +1223,9 @@ def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_para continue src_url = src if src.startswith('http') else compat_urlparse.urljoin(base, src) + if src_url in src_urls: + continue + src_urls.append(src_url) if proto == 'm3u8' or src_ext == 'm3u8': m3u8_formats = self._extract_m3u8_formats( @@ -1267,11 +1271,13 @@ def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_para return formats def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'): + urls = [] subtitles = {} for num, textstream in enumerate(smil.findall(self._xpath_ns('.//textstream', namespace))): src = textstream.get('src') - if not src: + if not src or src in urls: continue + urls.append(src) ext = textstream.get('ext') or determine_ext(src) if not ext: type_ = textstream.get('type')