X-Git-Url: http://git.megacz.com/?a=blobdiff_plain;f=GHC%2FIO%2FEncoding.hs;h=5d8ecb4c7062894a5f0b11b6abc6062dc395f061;hb=54a0b4928967a350be687e17ff3576f2af6e21fb;hp=bb976e3cce6cc398751ed5bf406e32e4b8790a59;hpb=8afc9fecd586d3c4f7ef9c69fb1686a79e5f441d;p=ghc-base.git

diff --git a/GHC/IO/Encoding.hs b/GHC/IO/Encoding.hs
index bb976e3..5d8ecb4 100644
--- a/GHC/IO/Encoding.hs
+++ b/GHC/IO/Encoding.hs
@@ -1,4 +1,6 @@
-{-# OPTIONS_GHC -fno-implicit-prelude -funbox-strict-fields #-}
+{-# LANGUAGE CPP, NoImplicitPrelude, PatternGuards #-}
+{-# OPTIONS_GHC -funbox-strict-fields #-}
+
 -----------------------------------------------------------------------------
 -- |
 -- Module      :  GHC.IO.Encoding
@@ -16,7 +18,7 @@
 module GHC.IO.Encoding (
   BufferCodec(..), TextEncoding(..), TextEncoder, TextDecoder,
   latin1, latin1_encode, latin1_decode,
-  utf8, 
+  utf8, utf8_bom,
   utf16, utf16le, utf16be,
   utf32, utf32le, utf32be, 
   localeEncoding,
@@ -30,6 +32,9 @@ import GHC.IO.Encoding.Types
 import GHC.Word
 #if !defined(mingw32_HOST_OS)
 import qualified GHC.IO.Encoding.Iconv  as Iconv
+#else
+import qualified GHC.IO.Encoding.CodePage as CodePage
+import Text.Read (reads)
 #endif
 import qualified GHC.IO.Encoding.Latin1 as Latin1
 import qualified GHC.IO.Encoding.UTF8   as UTF8
@@ -43,45 +48,88 @@ import GHC.IO.Exception
 
 -- -----------------------------------------------------------------------------
 
-latin1, utf8, utf16, utf16le, utf16be, utf32, utf32le, utf32be, localeEncoding
-  :: TextEncoding
-
 -- | The Latin1 (ISO8859-1) encoding.  This encoding maps bytes
 -- directly to the first 256 Unicode code points, and is thus not a
--- complete Unicode encoding.
+-- complete Unicode encoding.  An attempt to write a character greater than
+-- '\255' to a 'Handle' using the 'latin1' encoding will result in an error.
+latin1  :: TextEncoding
 latin1 = Latin1.latin1_checked
 
--- | The UTF-8 unicode encoding
+-- | The UTF-8 Unicode encoding
+utf8  :: TextEncoding
 utf8 = UTF8.utf8
 
--- | The UTF-16 unicode encoding (a byte-order-mark should be used to
+-- | The UTF-8 Unicode encoding, with a byte-order-mark (BOM; the byte
+-- sequence 0xEF 0xBB 0xBF).  This encoding behaves like 'utf8',
+-- except that on input, the BOM sequence is ignored at the beginning
+-- of the stream, and on output, the BOM sequence is prepended.
+--
+-- The byte-order-mark is strictly unnecessary in UTF-8, but is
+-- sometimes used to identify the encoding of a file.
+--
+utf8_bom  :: TextEncoding
+utf8_bom = UTF8.utf8_bom
+
+-- | The UTF-16 Unicode encoding (a byte-order-mark should be used to
 -- indicate endianness).
+utf16  :: TextEncoding
 utf16 = UTF16.utf16
 
--- | The UTF-16 unicode encoding (litte-endian)
+-- | The UTF-16 Unicode encoding (litte-endian)
+utf16le  :: TextEncoding
 utf16le = UTF16.utf16le
 
--- | The UTF-16 unicode encoding (big-endian)
+-- | The UTF-16 Unicode encoding (big-endian)
+utf16be  :: TextEncoding
 utf16be = UTF16.utf16be
 
--- | The UTF-32 unicode encoding (a byte-order-mark should be used to
+-- | The UTF-32 Unicode encoding (a byte-order-mark should be used to
 -- indicate endianness).
+utf32  :: TextEncoding
 utf32 = UTF32.utf32
 
--- | The UTF-32 unicode encoding (litte-endian)
+-- | The UTF-32 Unicode encoding (litte-endian)
+utf32le  :: TextEncoding
 utf32le = UTF32.utf32le
 
--- | The UTF-32 unicode encoding (big-endian)
+-- | The UTF-32 Unicode encoding (big-endian)
+utf32be  :: TextEncoding
 utf32be = UTF32.utf32be
 
--- | The text encoding of the current locale
+-- | The Unicode encoding of the current locale
+localeEncoding  :: TextEncoding
 #if !defined(mingw32_HOST_OS)
 localeEncoding = Iconv.localeEncoding
 #else
-localeEncoding = Latin1.latin1
+localeEncoding = CodePage.localeEncoding
 #endif
 
--- | Acquire the named text encoding
+-- | Look up the named Unicode encoding.  May fail with 
+--
+--  * 'isDoesNotExistError' if the encoding is unknown
+--
+-- The set of known encodings is system-dependent, but includes at least:
+--
+--  * @UTF-8@
+--
+--  * @UTF-16@, @UTF-16BE@, @UTF-16LE@
+--
+--  * @UTF-32@, @UTF-32BE@, @UTF-32LE@
+--
+-- On systems using GNU iconv (e.g. Linux), there is additional
+-- notation for specifying how illegal characters are handled:
+--
+--  * a suffix of @\/\/IGNORE@, e.g. @UTF-8\/\/IGNORE@, will cause 
+--    all illegal sequences on input to be ignored, and on output
+--    will drop all code points that have no representation in the
+--    target encoding.
+--
+--  * a suffix of @\/\/TRANSLIT@ will choose a replacement character
+--    for illegal sequences or code points.
+--
+-- On Windows, you can access supported code pages with the prefix
+-- @CP@; for example, @\"CP1250\"@.
+--
 mkTextEncoding :: String -> IO TextEncoding
 #if !defined(mingw32_HOST_OS)
 mkTextEncoding = Iconv.mkTextEncoding
@@ -93,8 +141,10 @@ mkTextEncoding "UTF-16BE" = return utf16be
 mkTextEncoding "UTF-32"   = return utf32
 mkTextEncoding "UTF-32LE" = return utf32le
 mkTextEncoding "UTF-32BE" = return utf32be
+mkTextEncoding ('C':'P':n)
+    | [(cp,"")] <- reads n = return $ CodePage.codePageEncoding cp
 mkTextEncoding e = ioException
-     (IOError Nothing InvalidArgument "mkTextEncoding"
+     (IOError Nothing NoSuchThing "mkTextEncoding"
           ("unknown encoding:" ++ e)  Nothing Nothing)
 #endif