From 0a5b1295b7c1aa6395b65ee137087c540b37b32b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Sun, 7 Jan 2018 00:31:53 +0700 Subject: [PATCH] [motherless:group] Relax entry extraction and add a fallback scenario --- youtube_dl/extractor/motherless.py | 29 +++++++++++++++++++++-------- 1 file changed, 21 insertions(+), 8 deletions(-) diff --git a/youtube_dl/extractor/motherless.py b/youtube_dl/extractor/motherless.py index 4adac691c..e24396e79 100644 --- a/youtube_dl/extractor/motherless.py +++ b/youtube_dl/extractor/motherless.py @@ -148,14 +148,27 @@ class MotherlessGroupIE(InfoExtractor): else super(MotherlessGroupIE, cls).suitable(url)) def _extract_entries(self, webpage, base): - return [ - self.url_result( - compat_urlparse.urljoin(base, video_path), - MotherlessIE.ie_key(), video_title=title) - for video_path, title in orderedSet(re.findall( - r'href="(/[^"]+)"[^>]+>\s+]+alt="[^-]+-\s([^"]+)"', - webpage)) - ] + entries = [] + for mobj in re.finditer( + r'href="(?P/[^"]+)"[^>]*>(?:\s*]+alt="[^-]+-\s(?P[^"]+)")?', + webpage): + video_url = compat_urlparse.urljoin(base, mobj.group('href')) + if not MotherlessIE.suitable(video_url): + continue + video_id = MotherlessIE._match_id(video_url) + title = mobj.group('title') + entries.append(self.url_result( + video_url, ie=MotherlessIE.ie_key(), video_id=video_id, + video_title=title)) + # Alternative fallback + if not entries: + entries = [ + self.url_result( + compat_urlparse.urljoin(base, '/' + video_id), + ie=MotherlessIE.ie_key(), video_id=video_id) + for video_id in orderedSet(re.findall( + r'data-codename=["\']([A-Z0-9]+)', webpage))] + return entries def _real_extract(self, url): group_id = self._match_id(url)