add decodeW8

3 files changed, 28 insertions, 8 deletions

diff --git a/Command/Sync.hs b/Command/Sync.hs
index 2a27bf870..630ceb053 100644
--- a/Command/Sync.hs
+++ b/Command/Sync.hs
@@ -267,7 +267,7 @@ mergeFile file key
 			++ takeExtension file
 		
 shortHash :: String -> String
-shortHash = take 4 . md5s . encodeFilePath
+shortHash = take 4 . md5s . md5FilePath
 
 changed :: Remote -> Git.Ref -> Annex Bool
 changed remote b = do
diff --git a/Locations.hs b/Locations.hs
index be5fd6c2f..d68d5fb5f 100644
--- a/Locations.hs
+++ b/Locations.hs
@@ -248,12 +248,12 @@ hashDirMixed :: Hasher
 hashDirMixed k = addTrailingPathSeparator $ take 2 dir </> drop 2 dir
 	where
 		dir = take 4 $ display_32bits_as_dir =<< [a,b,c,d]
-		ABCD (a,b,c,d) = md5 $ encodeFilePath $ key2file k
+		ABCD (a,b,c,d) = md5 $ md5FilePath $ key2file k
 
 hashDirLower :: Hasher
 hashDirLower k = addTrailingPathSeparator $ take 3 dir </> drop 3 dir
 	where
-		dir = take 6 $ md5s $ encodeFilePath $ key2file k
+		dir = take 6 $ md5s $ md5FilePath $ key2file k
 
 {- modified version of display_32bits_as_hex from Data.Hash.MD5
  -   Copyright (C) 2001 Ian Lynagh 
diff --git a/Utility/FileSystemEncoding.hs b/Utility/FileSystemEncoding.hs
index d027ede48..98f2f7755 100644
--- a/Utility/FileSystemEncoding.hs
+++ b/Utility/FileSystemEncoding.hs
@@ -5,7 +5,13 @@
  - Licensed under the GNU GPL version 3 or higher.
  -}
 
-module Utility.FileSystemEncoding where
+module Utility.FileSystemEncoding (
+	fileEncoding,
+	withFilePath,
+	md5FilePath,
+	decodeW8,
+	encodeW8	
+) where
 
 import qualified GHC.Foreign as GHC
 import qualified GHC.IO.Encoding as Encoding
@@ -31,19 +37,28 @@ withFilePath :: FilePath -> (CString -> IO a) -> IO a
 withFilePath fp f = Encoding.getFileSystemEncoding
 	>>= \enc -> GHC.withCString enc fp f
 
-{- Encodes a FilePath into a Md5.Str, applying the filesystem encoding.
+{- Encodes a FilePath into a String, applying the filesystem encoding.
+ -
+ - There are very few things it makes sense to do with such an encoded
+ - string. It's not a legal filename; it should not be displayed.
+ - So this function is not exported, but instead used by the few functions
+ - that can usefully consume it.
  -
  - This use of unsafePerformIO is belived to be safe; GHC's interface
  - only allows doing this conversion with CStrings, and the CString buffer
  - is allocated, used, and deallocated within the call, with no side
  - effects.
  -}
-{-# NOINLINE encodeFilePath #-}
-encodeFilePath :: FilePath -> MD5.Str
-encodeFilePath fp = MD5.Str $ unsafePerformIO $ do
+{-# NOINLINE _encodeFilePath #-}
+_encodeFilePath :: FilePath -> String
+_encodeFilePath fp = unsafePerformIO $ do
 	enc <- Encoding.getFileSystemEncoding
 	GHC.withCString enc fp $ GHC.peekCString Encoding.char8
 
+{- Encodes a FilePath into a Md5.Str, applying the filesystem encoding. -}
+md5FilePath :: FilePath -> MD5.Str
+md5FilePath = MD5.Str . _encodeFilePath
+
 {- Converts a [Word8] to a FilePath, encoding using the filesystem encoding.
  -
  - w82c produces a String, which may contain Chars that are invalid
@@ -55,3 +70,8 @@ encodeW8 :: [Word8] -> FilePath
 encodeW8 w8 = unsafePerformIO $ do
 	enc <- Encoding.getFileSystemEncoding
 	GHC.withCString Encoding.char8 (w82s w8) $ GHC.peekCString enc
+
+{- Useful when you want the actual number of bytes that will be used to
+ - represent the FilePath on disk. -}
+decodeW8 :: FilePath -> [Word8]
+decodeW8 = s2w8 . _encodeFilePath