Org reader: parse LaTeX-style MathML entities

Org supports special symbols which can be included using LaTeX syntax,
but are actually MathML entities.  Examples for this are
`\nbsp` (non-breaking space), `\Aacute` (the letter A with accent acute)
or `\copy` (the copyright sign ©).

This fixes #1657.

1 files changed, 8 insertions, 1 deletions

diff --git a/src/Text/Pandoc/Readers/Org.hs b/src/Text/Pandoc/Readers/Org.hs
index a6ebf65dc..1ddfeab87 100644
--- a/src/Text/Pandoc/Readers/Org.hs
+++ b/src/Text/Pandoc/Readers/Org.hs
@@ -42,6 +42,7 @@ import           Text.Pandoc.Parsing hiding ( F, unF, askF, asksF, runF
 import           Text.Pandoc.Readers.LaTeX (inlineCommand, rawLaTeXInline)
 import           Text.Pandoc.Shared (compactify', compactify'DL)
 import           Text.TeXMath (readTeX, writePandoc, DisplayType(..))
+import qualified Text.TeXMath.Readers.MathML.EntityMap as MathMLEntityMap
 
 import           Control.Applicative ( Applicative, pure
                                      , (<$>), (<$), (<*>), (<*), (*>) )
@@ -1431,7 +1432,8 @@ simpleSubOrSuperString = try $
 inlineLaTeX :: OrgParser (F Inlines)
 inlineLaTeX = try $ do
   cmd <- inlineLaTeXCommand
-  maybe mzero returnF $ parseAsMath cmd `mplus` parseAsInlineLaTeX cmd
+  maybe mzero returnF $
+     parseAsMath cmd `mplus` parseAsMathMLSym cmd `mplus` parseAsInlineLaTeX cmd
  where
    parseAsMath :: String -> Maybe Inlines
    parseAsMath cs = B.fromList <$> texMathToPandoc cs
@@ -1439,6 +1441,11 @@ inlineLaTeX = try $ do
    parseAsInlineLaTeX :: String -> Maybe Inlines
    parseAsInlineLaTeX cs = maybeRight $ runParser inlineCommand state "" cs
 
+   parseAsMathMLSym :: String -> Maybe Inlines
+   parseAsMathMLSym cs = B.str <$> MathMLEntityMap.getUnicode (clean cs)
+    -- dropWhileEnd would be nice here, but it's not available before base 4.5
+    where clean = reverse . dropWhile (`elem` "{}") . reverse . drop 1
+
    state :: ParserState
    state = def{ stateOptions = def{ readerParseRaw = True }}