From 09747ba7663a9c6f89530c7ffbd95cb4776db6bf Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan Date: Sat, 19 Aug 2017 22:27:53 +0800 Subject: [PATCH] [liveleak] Support another liveleak embedding pattern (closes #13336) --- ChangeLog | 3 +++ youtube_dl/extractor/extractors.py | 5 ++++- youtube_dl/extractor/generic.py | 23 +++++++++++++++----- youtube_dl/extractor/liveleak.py | 35 +++++++++++++++++++++++++----- 4 files changed, 55 insertions(+), 11 deletions(-) diff --git a/ChangeLog b/ChangeLog index 6c32747c4..4c7997b2e 100644 --- a/ChangeLog +++ b/ChangeLog @@ -4,6 +4,9 @@ Core * [cda] Fix extraction (closes #13935) * [utils] Fix unescapeHTML for misformed string like "&a"" (#13935) +Extractors ++ [liveleak] Support another liveleak embedding pattern (#13336) + version 2017.08.18 diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index bda6826f1..17048fd6e 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -537,7 +537,10 @@ from .limelight import ( LimelightChannelListIE, ) from .litv import LiTVIE -from .liveleak import LiveLeakIE +from .liveleak import ( + LiveLeakIE, + LiveLeakEmbedIE, +) from .livestream import ( LivestreamIE, LivestreamOriginalIE, diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index d2fb2627d..49b00b87e 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -1519,14 +1519,27 @@ class GenericIE(InfoExtractor): # LiveLeak embed { 'url': 'http://www.wykop.pl/link/3088787/', - 'md5': 'ace83b9ed19b21f68e1b50e844fdf95d', + 'md5': '7619da8c820e835bef21a1efa2a0fc71', 'info_dict': { 'id': '874_1459135191', 'ext': 'mp4', 'title': 'Man shows poor quality of new apartment building', 'description': 'The wall is like a sand pile.', 'uploader': 'Lake8737', - } + }, + 'add_ie': [LiveLeakIE.ie_key()], + }, + # Another LiveLeak embed pattern (#13336) + { + 'url': 'https://milo.yiannopoulos.net/2017/06/concealed-carry-robbery/', + 'info_dict': { + 'id': '2eb_1496309988', + 'ext': 'mp4', + 'title': 'Thief robs place where everyone was armed', + 'description': 'md5:694d73ee79e535953cf2488562288eee', + 'uploader': 'brazilwtf', + }, + 'add_ie': [LiveLeakIE.ie_key()], }, # Duplicated embedded video URLs { @@ -2757,9 +2770,9 @@ class GenericIE(InfoExtractor): self._proto_relative_url(instagram_embed_url), InstagramIE.ie_key()) # Look for LiveLeak embeds - liveleak_url = LiveLeakIE._extract_url(webpage) - if liveleak_url: - return self.url_result(liveleak_url, 'LiveLeak') + liveleak_urls = LiveLeakIE._extract_urls(webpage) + if liveleak_urls: + return self.playlist_from_matches(liveleak_urls, video_id, video_title) # Look for 3Q SDN embeds threeqsdn_url = ThreeQSDNIE._extract_url(webpage) diff --git a/youtube_dl/extractor/liveleak.py b/youtube_dl/extractor/liveleak.py index b2247a84d..d23eaa355 100644 --- a/youtube_dl/extractor/liveleak.py +++ b/youtube_dl/extractor/liveleak.py @@ -75,12 +75,10 @@ class LiveLeakIE(InfoExtractor): }] @staticmethod - def _extract_url(webpage): - mobj = re.search( - r']+src="https?://(?:\w+\.)?liveleak\.com/ll_embed\?(?:.*?)i=(?P[\w_]+)(?:.*)', + def _extract_urls(webpage): + return re.findall( + r']+src="(https?://(?:\w+\.)?liveleak\.com/ll_embed\?[^"]*[if]=[\w_]+[^"]+)"', webpage) - if mobj: - return 'http://www.liveleak.com/view?i=%s' % mobj.group('id') def _real_extract(self, url): video_id = self._match_id(url) @@ -131,3 +129,30 @@ class LiveLeakIE(InfoExtractor): }) return info_dict + + +class LiveLeakEmbedIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?liveleak\.com/ll_embed\?.*?\b(?P[if])=(?P[\w_]+)' + + # See generic.py for actual test cases + _TESTS = [{ + 'url': 'https://www.liveleak.com/ll_embed?i=874_1459135191', + 'only_matching': True, + }, { + 'url': 'https://www.liveleak.com/ll_embed?f=ab065df993c1', + 'only_matching': True, + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + kind, video_id = mobj.group('kind', 'id') + + if kind == 'f': + webpage = self._download_webpage(url, video_id) + liveleak_url = self._search_regex( + r'logourl\s*:\s*(?P[\'"])(?P%s)(?P=q1)' % LiveLeakIE._VALID_URL, + webpage, 'LiveLeak URL', group='url') + elif kind == 'i': + liveleak_url = 'http://www.liveleak.com/view?i=%s' % video_id + + return self.url_result(liveleak_url, ie=LiveLeakIE.ie_key())