From a04101db46e96176f0a751a2211f817b3b569c9d Mon Sep 17 00:00:00 2001
From: "Kyle J. McKay" <mackyle@gmail.com>
Date: Wed, 30 Sep 2020 23:26:54 -0700
Subject: [PATCH] Markdown.pl: minor empty tag improvments

Even though block tags such as "<p/>" should not appear in
valid XHTML documents, the internal validator (which is
enabled by default) will properly expand "<p/>" to "<p></p>".

However, the block formatting code fails to notice such
an empty tag block leading to it being wrapped in a spurious
"<p>...</p>" pair before it's expanded by the validation code.

Attempt to recognize some of these valid-for-xml-but-not-xhtml
blocks earlier to produce better output.

This is not a perfect fix, but it's an improvement.

It's really an odd edge case anyway that's unlikely to be
encountered very often.

Signed-off-by: Kyle J. McKay <mackyle@gmail.com>
---
 Markdown.pl | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/Markdown.pl b/Markdown.pl
index e1e0c6e..8864f93 100755
--- a/Markdown.pl
+++ b/Markdown.pl
@@ -936,6 +936,27 @@ sub _HashHTMLBlocks {
 		$g_html_blocks{$key} = $1;
 		"\n\n" . $key . "\n\n";
 	    }eigmx;
+
+    #
+    # Now match any empty block tags that should have been paired
+    #
+    $text =~ s{
+		(			# save in $1
+		    ^			# start of line (with /m)
+		    (?:\Q$idt\E)?	# optional lead in
+		    <(?:$block_tags_b)	# start tag = $2
+		    \b			# word break
+		    (?:[^<>])*?		#
+		    /?>			# the matching end tag
+		    [ ]*		# trailing spaces
+		    (?=\n+|\Z) # followed by a newline or end of document
+		)
+	    }{
+		my $key = block_id($1);
+		$g_html_blocks{$key} = $1;
+		"\n\n" . $key . "\n\n";
+	    }eigmx;
+
     # Special case just for <hr />. It was easier to make a special case than
     # to make the other regex more complicated.
     $text =~ s{