HTML reader: Revise treatment of li with id attribute.

Previously we always added an empty div before the list
item, but this created problems with spacing in tight
lists.  Now we do this:

If the list item contents begin with a Plain block,
we modify the Plain block by adding a Span around
its contents.

Otherwise, we add a Div around the contents of the
list item (instead of adding an empty Div to the
beginning, as before).

Closes #3596.

2 files changed, 67 insertions, 2 deletions

diff --git a/src/Text/Pandoc/Readers/HTML.hs b/src/Text/Pandoc/Readers/HTML.hs
index 5251962f2..14b051539 100644
--- a/src/Text/Pandoc/Readers/HTML.hs
+++ b/src/Text/Pandoc/Readers/HTML.hs
@@ -260,8 +260,12 @@ pBulletList = try $ do
 pListItem :: PandocMonad m => TagParser m a -> TagParser m Blocks
 pListItem nonItem = do
   TagOpen _ attr <- lookAhead $ pSatisfy (~== TagOpen "li" [])
-  let liDiv = maybe mempty (\x -> B.divWith (x, [], []) mempty) (lookup "id" attr)
-  (liDiv <>) <$> pInTags "li" block <* skipMany nonItem
+  let addId ident bs = case B.toList bs of
+                           (Plain ils:xs) -> B.fromList (Plain
+                                [Span (ident, [], []) ils] : xs)
+                           _ -> B.divWith (ident, [], []) bs
+  (maybe id addId (lookup "id" attr)) <$>
+    pInTags "li" block <* skipMany nonItem
 
 parseListStyleType :: String -> ListNumberStyle
 parseListStyleType "lower-roman" = LowerRoman
diff --git a/test/command/3596.md b/test/command/3596.md
new file mode 100644
index 000000000..a064ca632
--- /dev/null
+++ b/test/command/3596.md
@@ -0,0 +1,61 @@
+```
+% pandoc -f html -t markdown-raw_html-bracketed_spans-native_spans
+<ul>
+<li>foo</li>
+<li id="id">bar</li>
+<li>baz</li>
+</ul>
+^D
+-   foo
+-   bar
+-   baz
+```
+
+```
+% pandoc -f html -t markdown-raw_html-bracketed_spans-native_spans
+<ul>
+<li>foo</li>
+<li id="id">bar<ul><li>subbar</li></ul></li>
+<li>baz</li>
+</ul>
+^D
+-   foo
+-   bar
+    -   subbar
+-   baz
+```
+
+
+```
+% pandoc -f html -t markdown
+<ul>
+<li>foo</li>
+<li id="id">bar</li>
+<li>baz</li>
+</ul>
+^D
+-   foo
+-   [bar]{#id}
+-   baz
+```
+
+
+```
+% pandoc -f html -t markdown
+<ul>
+<li><p>foo</p></li>
+<li id="id"><p>bar</p></li>
+<li><p>baz</p></li>
+</ul>
+^D
+-   foo
+
+-   <div id="id">
+
+    bar
+
+    </div>
+
+-   baz
+
+```