4 files changed, 84 insertions, 8 deletions
diff --git a/Text/Pandoc/ODT.hs b/Text/Pandoc/ODT.hs
index b0bfc6693..5b900bf03 100644
--- a/Text/Pandoc/ODT.hs
+++ b/Text/Pandoc/ODT.hs
@@ -43,11 +43,11 @@ import Network.URI ( isURI )
 import qualified Data.ByteString as B ( writeFile, pack )
 import Data.ByteString.Internal ( c2w )
 import Prelude hiding ( writeFile, readFile )
-#ifdef _UTF8
-import System.IO.UTF8
 import System.IO ( stderr )
+#ifdef _UTF8STRING
+import System.IO.UTF8
 #else
-import System.IO
+import Text.Pandoc.UTF8
 #endif
 
 -- | Produce an ODT file from OpenDocument XML.
diff --git a/Text/Pandoc/Shared.hs b/Text/Pandoc/Shared.hs
index bc0791d77..6f37f633b 100644
--- a/Text/Pandoc/Shared.hs
+++ b/Text/Pandoc/Shared.hs
@@ -116,11 +116,11 @@ import Network.URI ( parseURI, URI (..), isAllowedInURI )
 import System.FilePath ( (</>), (<.>) )
 import System.IO.Error ( catch, ioError, isAlreadyExistsError )
 import System.Directory
-import Prelude hiding ( putStrLn )
-#ifdef _UTF8
+import Prelude hiding ( putStrLn, writeFile, readFile, getContents )
+#ifdef _UTF8STRING
 import System.IO.UTF8
 #else
-import System.IO
+import Text.Pandoc.UTF8
 #endif
 
 --
diff --git a/Text/Pandoc/TH.hs b/Text/Pandoc/TH.hs
index 30f5c86e4..376dfa00d 100644
--- a/Text/Pandoc/TH.hs
+++ b/Text/Pandoc/TH.hs
@@ -39,10 +39,10 @@ import Language.Haskell.TH.Syntax (Lift (..))
 import qualified Data.ByteString as B
 import Data.ByteString.Internal ( w2c )
 import Prelude hiding ( readFile )
-#ifdef _UTF8
+#ifdef _UTF8STRING
 import System.IO.UTF8
 #else
-import System.IO
+import Text.Pandoc.UTF8
 #endif
 
 -- | Insert contents of text file into a template.
diff --git a/Text/Pandoc/UTF8.hs b/Text/Pandoc/UTF8.hs
new file mode 100644
index 000000000..f8d041db7
--- /dev/null
+++ b/Text/Pandoc/UTF8.hs
@@ -0,0 +1,76 @@
+-- | Functions for IO using UTF-8 encoding.
+-- 
+-- The basic encoding and decoding functions are taken from
+-- <http://www.cse.ogi.edu/~hallgren/Talks/LHiH/base/lib/UTF8.hs>.
+-- (c) 2003, OGI School of Science & Engineering, Oregon Health and
+-- Science University.  
+--
+-- From the Char module supplied with HBC.
+-- Modified by Martin Norbaeck to pass illegal UTF-8 sequences unchanged.
+-- Modified by John MacFarlane to use [Word8] and export IO functions.
+
+module Text.Pandoc.UTF8 ( 
+              putStrLn
+            , putStr
+            , hPutStrLn
+            , hPutStr
+            , getContents
+            , readFile
+            , writeFile
+            ) where
+import Data.Word
+import System.IO ( Handle )
+import qualified Data.ByteString.Lazy as BS
+import Prelude hiding ( putStrLn, putStr, getContents, readFile, writeFile )
+
+putStrLn :: String -> IO ()
+putStrLn =  BS.putStrLn . BS.pack . toUTF8
+
+putStr :: String -> IO ()
+putStr = BS.putStr . BS.pack . toUTF8
+
+hPutStrLn :: Handle -> String -> IO ()
+hPutStrLn h =  BS.hPut h . BS.pack . toUTF8 . (++ "\n")
+
+hPutStr :: Handle -> String -> IO ()
+hPutStr h = BS.hPut h . BS.pack . toUTF8
+
+readFile :: FilePath -> IO String 
+readFile p = BS.readFile p >>= return . fromUTF8 . BS.unpack
+
+writeFile :: FilePath -> String -> IO ()
+writeFile p = BS.writeFile p . BS.pack . toUTF8
+
+getContents :: IO String
+getContents = BS.getContents >>= return . fromUTF8 . BS.unpack
+
+-- | Take a list of bytes in UTF-8 encoding and decode it into a Unicode string.
+fromUTF8 :: [Word8] -> String
+fromUTF8 [] = ""
+fromUTF8 (0xef : 0xbb : 0xbf :cs) = fromUTF8 cs -- skip BOM (byte order marker)
+fromUTF8 (c:c':cs) | 0xc0 <= c  && c  <= 0xdf && 
+                       0x80 <= c' && c' <= 0xbf =
+	toEnum ((fromEnum c `mod` 0x20) * 0x40 + fromEnum c' `mod` 0x40) : fromUTF8 cs
+fromUTF8 (c:c':c'':cs) | 0xe0 <= c   && c   <= 0xef && 
+		                   0x80 <= c'  && c'  <= 0xbf &&
+                           0x80 <= c'' && c'' <= 0xbf =
+	toEnum ((fromEnum c `mod` 0x10 * 0x1000) + (fromEnum c' `mod` 0x40) * 0x40 + fromEnum c'' `mod` 0x40) : fromUTF8 cs
+fromUTF8 (c:cs) = toEnum (fromEnum c) : fromUTF8 cs
+
+-- | Take a Unicode string and encode it as a list of bytes in UTF-8 encoding.
+toUTF8 :: String -> [Word8]
+toUTF8 "" = []
+toUTF8 (c:cs) =
+	if c > '\x0000' && c < '\x0080' then
+	    toEnum (fromEnum c) : toUTF8 cs
+	else if c < toEnum 0x0800 then
+	    let i = fromEnum c
+	    in  toEnum (0xc0 + i `div` 0x40) : 
+	        toEnum (0x80 + i `mod` 0x40) : 
+		toUTF8 cs
+	else
+	    let i = fromEnum c
+	    in  toEnum (0xe0 + i `div` 0x1000) : 
+	        toEnum (0x80 + (i `mod` 0x1000) `div` 0x40) : 
+		toEnum (0x80 + i `mod` 0x40) : 
+		toUTF8 cs