From f88ebf3ebf49e00ffa12778caf6817cc34459e6a Mon Sep 17 00:00:00 2001
From: John MacFarlane <jgm@berkeley.edu>
Date: Tue, 6 Jul 2021 10:22:07 -0700
Subject: Markdown reader:  don't try to read contents in self-closing HTML
 tag.

Previously we had problems parsing raw HTML with self-closing
tags like `<col/>`. The problem was that pandoc would look
for a closing tag to close the markdown contents, but the
closing tag had, in effect, already been parsed by `htmlTag`.

This fixes the issue described in
<https://groups.google.com/d/msgid/pandoc-discuss/297bc662-7841-4423-bcbb-534e99bbba09n%40googlegroups.com>.
---
 src/Text/Pandoc/Readers/Markdown.hs | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

(limited to 'src')

diff --git a/src/Text/Pandoc/Readers/Markdown.hs b/src/Text/Pandoc/Readers/Markdown.hs
index 1e9867d07..2dc7ddf52 100644
--- a/src/Text/Pandoc/Readers/Markdown.hs
+++ b/src/Text/Pandoc/Readers/Markdown.hs
@@ -1121,6 +1121,7 @@ rawTeXBlock = do
 rawHtmlBlocks :: PandocMonad m => MarkdownParser m (F Blocks)
 rawHtmlBlocks = do
   (TagOpen tagtype _, raw) <- htmlTag isBlockTag
+  let selfClosing = "/>" `T.isSuffixOf` raw
   -- we don't want '<td>    text' to be a code block:
   skipMany spaceChar
   indentlevel <- (blankline >> length <$> many (char ' ')) <|> return 0
@@ -1134,7 +1135,9 @@ rawHtmlBlocks = do
                  gobbleAtMostSpaces indentlevel
                  notFollowedBy' closer
                  block
-  contents <- mconcat <$> many block'
+  contents <- if selfClosing
+                 then return mempty
+                 else mconcat <$> many block'
   result <-
     try
     (do gobbleAtMostSpaces indentlevel
-- 
cgit v1.2.3