From 887c6acdf231e6846e0dbb63b644fc7dc3378700 Mon Sep 17 00:00:00 2001 From: Philipp Hagemeister Date: Mon, 18 Nov 2013 13:28:26 +0100 Subject: [PATCH] Support multiple embedded YouTube URLs (Fixes #1787) --- youtube_dl/extractor/generic.py | 31 +++++++++++++++++-------------- 1 file changed, 17 insertions(+), 14 deletions(-) diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index c7552fddb..e1d6a2a01 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -162,6 +162,16 @@ def _real_extract(self, url): raise ExtractorError(u'Failed to download URL: %s' % url) self.report_extraction(video_id) + + # it's tempting to parse this further, but you would + # have to take into account all the variations like + # Video Title - Site Name + # Site Name | Video Title + # Video Title - Tagline | Site Name + # and so on and so forth; it's just not practical + video_title = self._html_search_regex(r'(.*)', + webpage, u'video title', default=u'video', flags=re.DOTALL) + # Look for BrightCove: bc_url = BrightcoveIE._extract_brightcove_url(webpage) if bc_url is not None: @@ -177,11 +187,13 @@ def _real_extract(self, url): return self.url_result(surl, 'Vimeo') # Look for embedded YouTube player - mobj = re.search( - r']+?src=(["\'])(?Phttps?://(?:www\.)?youtube.com/embed/.+?)\1', webpage) - if mobj: - surl = unescapeHTML(mobj.group(u'url')) - return self.url_result(surl, 'Youtube') + matches = re.findall( + r']+?src=(["\'])(?P(?:https?:)?//(?:www\.)?youtube.com/embed/.+?)\1', webpage) + if matches: + urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Youtube') + for tuppl in matches] + return self.playlist_result( + urlrs, playlist_id=video_id, playlist_title=video_title) # Look for Bandcamp pages with custom domain mobj = re.search(r']*?content="(.*?bandcamp\.com.*?)"', webpage) @@ -226,15 +238,6 @@ def _real_extract(self, url): video_extension = os.path.splitext(video_id)[1][1:] video_id = os.path.splitext(video_id)[0] - # it's tempting to parse this further, but you would - # have to take into account all the variations like - # Video Title - Site Name - # Site Name | Video Title - # Video Title - Tagline | Site Name - # and so on and so forth; it's just not practical - video_title = self._html_search_regex(r'(.*)', - webpage, u'video title', default=u'video', flags=re.DOTALL) - # video uploader is domain name video_uploader = self._search_regex(r'(?:https?://)?([^/]*)/.*', url, u'video uploader')