HTML writer: override Text.XHtml's stringToHtml function,

so that characters below 0xff are not converted to numerical entity
references. Also convert '\160' to " ". This should aid readability
and editability of the HTML source. It does presuppose that the HTML
will be served as UTF-8.


git-svn-id: https://pandoc.googlecode.com/svn/trunk@1338 788f1e2b-df1e-0410-8736-df70ead52e1b

1 files changed, 15 insertions, 1 deletions

diff --git a/Text/Pandoc/Writers/HTML.hs b/Text/Pandoc/Writers/HTML.hs
index d0c45bec1..d31ebfb85 100644
--- a/Text/Pandoc/Writers/HTML.hs
+++ b/Text/Pandoc/Writers/HTML.hs
@@ -39,7 +39,7 @@ import Data.Char ( ord, toLower, isAlpha )
 import Data.List ( isPrefixOf, intersperse )
 import qualified Data.Set as S
 import Control.Monad.State
-import Text.XHtml.Transitional
+import Text.XHtml.Transitional hiding ( stringToHtml )
 
 data WriterState = WriterState
     { stNotes            :: [Html]       -- ^ List of notes
@@ -62,6 +62,20 @@ renderFragment opts = if writerWrapText opts
                          then renderHtmlFragment
                          else showHtmlFragment
 
+-- | Slightly modified version of Text.XHtml's stringToHtml.
+-- Only uses numerical entities for 0xff and greater.
+-- Adds  .
+stringToHtml :: String -> Html
+stringToHtml = primHtml . concatMap fixChar
+    where
+      fixChar '<' = "&lt;"
+      fixChar '>' = "&gt;"
+      fixChar '&' = "&amp;"
+      fixChar '"' = "&quot;"
+      fixChar '\160' = "&nbsp;"
+      fixChar c | ord c < 0xff = [c]
+      fixChar c = "&#" ++ show (ord c) ++ ";"
+
 -- | Convert Pandoc document to Html string.
 writeHtmlString :: WriterOptions -> Pandoc -> String
 writeHtmlString opts =