HTML Reader: Added ability to read MathML formatted <math> blocks

1 files changed, 16 insertions, 0 deletions

diff --git a/src/Text/Pandoc/Readers/HTML.hs b/src/Text/Pandoc/Readers/HTML.hs
index 34aa381d2..2e8b56124 100644
--- a/src/Text/Pandoc/Readers/HTML.hs
+++ b/src/Text/Pandoc/Readers/HTML.hs
@@ -571,6 +571,22 @@ pRawHtmlInline = do
      then return $ B.rawInline "html" $ renderTags' [result]
      else return mempty
 
+mathMLToTeXMath :: String -> Either String String
+mathMLToTeXMath s = writeTeXMath <$> readMathML s
+
+pMath :: Bool -> TagParser Inlines
+pMath inCase = try $ do
+  open@(TagOpen _ attr) <- pSatisfy $ tagOpen (=="math") (const True)
+  unless (inCase) (guard (maybe False  (== mathMLNamespace) (lookup "xmlns" attr)))
+  contents <- manyTill pAnyTag (pSatisfy (~== TagClose "math"))
+  let math = mathMLToTeXMath $
+              (renderTags $ [open] ++ contents ++ [TagClose "math"])
+  let constructor =
+        maybe B.math (\x -> if (x == "inline") then B.math else B.displayMath)
+          (lookup "display" attr)
+  return $ either (const mempty)
+            (\x -> if null x then mempty else constructor x) math
+
 pInlinesInTags :: String -> (Inlines -> Inlines)
                -> TagParser Inlines
 pInlinesInTags tagtype f = extractSpaces f <$> pInTags tagtype inline