Fixed bug in Markdown raw HTML parsing.
This was a regression, with the rewrite of `htmlInBalanced`
(from `Text.Pandoc.Readers.HTML`) in 1.17.
It caused newlines to be omitted in raw HTML blocks.
Closes #2804.
commitb1ffdf3b01c5acec18ee5f776841f478eb7b7810
authorJohn MacFarlane <jgm@berkeley.edu>
dateTue 22 Mar 2016, 16:56 (11 years ago)
parentdb20790
src/Text/Pandoc/Readers/HTML.hs+1−1
@@ -939,7 +939,7 @@ htmlInBalanced f = try $ do
939939 (TagClose _ : TagPosition er ec : _) -> do
940940 let ls = er - sr
941941 let cs = ec - sc
942− lscontents <- concat <$> count ls anyLine
942+ lscontents <- unlines <$> count ls anyLine
943943 cscontents <- count cs anyChar
944944 (_,closetag) <- htmlTag (~== TagClose tn)
945945 return (lscontents ++ cscontents ++ closetag)