Respect indent when parsing Org bullet lists

Fixes issue with top-level bullet list parsing.
Previously we would use `many1 spaceChars` rather than respecting
the list's indent level. We also permitted `*` bullets on unindented
lists, which should unambiguously parse as `header 1`.
Combined, this meant headers at a different indent level were
being unwittingly slurped into preceding bullet lists, as per
Issue #1650.

2 files changed, 37 insertions, 2 deletions

diff --git a/src/Text/Pandoc/Readers/Org.hs b/src/Text/Pandoc/Readers/Org.hs
index 5c00a1b27..199c9f2de 100644
--- a/src/Text/Pandoc/Readers/Org.hs
+++ b/src/Text/Pandoc/Readers/Org.hs
@@ -832,8 +832,9 @@ definitionList = fmap B.definitionList . fmap compactify'DL . sequence
                  <$> many1 (definitionListItem bulletListStart)
 
 bulletList :: OrgParser (F Blocks)
-bulletList = fmap B.bulletList . fmap compactify' . sequence
-             <$> many1 (listItem bulletListStart)
+bulletList = try $ do n <- lookAhead bulletListStart
+                      fmap B.bulletList . fmap compactify' . sequence
+                        <$> many1 (listItem (bulletListCont n))
 
 orderedList :: OrgParser (F Blocks)
 orderedList = fmap B.orderedList . fmap compactify' . sequence
@@ -850,6 +851,13 @@ bulletListStart :: OrgParser Int
 bulletListStart = genericListStart bulletListMarker
   where bulletListMarker = pure <$> oneOf "*-+"
 
+-- parses bullet list marker at a known indent level
+bulletListCont :: Int -> OrgParser Int
+bulletListCont n
+  -- Unindented lists are legal, but they can't use '*' bullets
+  | n <= 1 = oneOf "+-" >> return n
+  | otherwise = count (n-1) spaceChar >> oneOf "+-*" >> return n
+
 orderedListStart :: OrgParser Int
 orderedListStart = genericListStart orderedListMarker
   -- Ordered list markers allowed in org-mode
diff --git a/tests/Tests/Readers/Org.hs b/tests/Tests/Readers/Org.hs
index 92ec8155b..f27388e46 100644
--- a/tests/Tests/Readers/Org.hs
+++ b/tests/Tests/Readers/Org.hs
@@ -622,6 +622,33 @@ tests =
                        ]
                      ]
 
+      , "Bullet List with Decreasing Indent" =:
+           ("  - Discovery\n\
+            \ - Human After All\n") =?>
+           mconcat [ bulletList [ plain "Discovery" ]
+                   , bulletList [ plain ("Human" <> space <> "After" <> space <> "All")]
+                   ]
+
+      , "Header follows Bullet List" =:
+          ("  - Discovery\n\
+           \  - Human After All\n\
+           \* Homework") =?>
+          mconcat [ bulletList [ plain "Discovery"
+                               , plain ("Human" <> space <> "After" <> space <> "All")
+                               ]
+                  , header 1 "Homework"
+                  ]
+
+      , "Bullet List Unindented with trailing Header" =:
+          ("- Discovery\n\
+           \- Homework\n\
+           \* NotValidListItem") =?>
+          mconcat [ bulletList [ plain "Discovery"
+                               , plain "Homework"
+                               ]
+                  , header 1 "NotValidListItem"
+                  ]
+
       , "Simple Ordered List" =:
           ("1. Item1\n" ++
            "2. Item2\n") =?>