[project @ 2001-02-18 14:45:15 by qrczak]

Recent Unicode and future ISO-10646 finally decided that the character
code space ends at U+10FFFF. Let ghc follow the rules: maxBound::Char
is now '\x10FFFF', etc.

4 files changed, 28 insertions, 55 deletions

diff --git a/ghc/compiler/parser/Lex.lhs b/ghc/compiler/parser/Lex.lhs
index 8fdd6adada..130eef8b69 100644
--- a/ghc/compiler/parser/Lex.lhs
+++ b/ghc/compiler/parser/Lex.lhs
@@ -742,7 +742,7 @@ lex_escape cont buf
 			    [] -> charError buf'
 
 after_charnum cont i buf
-  = if i >= 0 && i <= 0x7FFFFFFF
+  = if i >= 0 && i <= 0x10FFFF
 	then cont (fromInteger i) buf
 	else charError buf
 
diff --git a/ghc/compiler/utils/UnicodeUtil.lhs b/ghc/compiler/utils/UnicodeUtil.lhs
index 64062dd3c7..51ca35881f 100644
--- a/ghc/compiler/utils/UnicodeUtil.lhs
+++ b/ghc/compiler/utils/UnicodeUtil.lhs
@@ -24,20 +24,7 @@ stringToUtf8 (c:s)
                         chr (0x80 + c `div`       0x40 `mod` 0x40) :
                         chr (0x80 + c                  `mod` 0x40) :
                         stringToUtf8 s
-    | c <= 0x1FFFFF   = chr (0xF0 + c `div`    0x40000           ) :
-                        chr (0x80 + c `div`     0x1000 `mod` 0x40) :
-                        chr (0x80 + c `div`       0x40 `mod` 0x40) :
-                        chr (0x80 + c                  `mod` 0x40) :
-                        stringToUtf8 s
-    | c <= 0x3FFFFFF  = chr (0xF8 + c `div`  0x1000000           ) :
-                        chr (0x80 + c `div`    0x40000 `mod` 0x40) :
-                        chr (0x80 + c `div`     0x1000 `mod` 0x40) :
-                        chr (0x80 + c `div`       0x40 `mod` 0x40) :
-                        chr (0x80 + c                  `mod` 0x40) :
-                        stringToUtf8 s
-    | c <= 0x7FFFFFFF = chr (0xFC + c `div` 0x40000000           ) :
-                        chr (0x80 + c `div`  0x1000000 `mod` 0x40) :
-                        chr (0x80 + c `div`    0x40000 `mod` 0x40) :
+    | c <= 0x10FFFF   = chr (0xF0 + c `div`    0x40000           ) :
                         chr (0x80 + c `div`     0x1000 `mod` 0x40) :
                         chr (0x80 + c `div`       0x40 `mod` 0x40) :
                         chr (0x80 + c                  `mod` 0x40) :
diff --git a/ghc/lib/std/PrelBase.lhs b/ghc/lib/std/PrelBase.lhs
index 4c0bcbe82d..7a933e07be 100644
--- a/ghc/lib/std/PrelBase.lhs
+++ b/ghc/lib/std/PrelBase.lhs
@@ -1,5 +1,5 @@
 % -----------------------------------------------------------------------------
-% $Id: PrelBase.lhs,v 1.39 2000/10/03 08:43:05 simonpj Exp $
+% $Id: PrelBase.lhs,v 1.40 2001/02/18 14:45:15 qrczak Exp $
 %
 % (c) The University of Glasgow, 1992-2000
 %
@@ -435,12 +435,9 @@ instance Ord Char where
   (C# c1) <  (C# c2) = c1 `ltChar#` c2
 
 chr :: Int -> Char
-chr (I# i) | i >=# 0#
-#if INT_SIZE_IN_BYTES > 4
-             && i <=# 0x7FFFFFFF#
-#endif
-             = C# (chr# i)
-	   | otherwise = error ("Prelude.chr: bad argument")
+chr (I# i) | i >=# 0# && i <=# 0x10FFFF#
+	     = C# (chr# i)
+	   | otherwise = error "Prelude.chr: bad argument"
 
 unsafeChr :: Int -> Char
 unsafeChr (I# i) =  C# (chr# i)
@@ -687,34 +684,23 @@ unpackCStringUtf8# addr
   = unpack 0#
   where
     unpack nh
-      | ch `eqChar#` '\0'# = []
+      | ch `eqChar#` '\0'#   = []
       | ch `leChar#` '\x7F'# = C# ch : unpack (nh +# 1#)
-      | ch `leChar#` '\xDF'# = C# (chr# ((ord# ch                                  `iShiftL#`  6#) +#
-                                         (ord# (indexCharOffAddr# addr (nh +# 1#))) -# 0x3080#))
-                               : unpack (nh +# 2#)
-      | ch `leChar#` '\xEF'# = C# (chr# ((ord# ch                                  `iShiftL#` 12#) +#
-                                         (ord# (indexCharOffAddr# addr (nh +# 1#)) `iShiftL#`  6#) +#
-                                         (ord# (indexCharOffAddr# addr (nh +# 2#))) -# 0xE2080#))
-                               : unpack (nh +# 3#)
-      | ch `leChar#` '\xF7'# = C# (chr# ((ord# ch                                  `iShiftL#` 18#) +#
-                                         (ord# (indexCharOffAddr# addr (nh +# 1#)) `iShiftL#` 12#) +#
-                                         (ord# (indexCharOffAddr# addr (nh +# 2#)) `iShiftL#`  6#) +#
-                                         (ord# (indexCharOffAddr# addr (nh +# 3#))) -# 0x3C82080#))
-                               : unpack (nh +# 4#)
-      | ch `leChar#` '\xFB'# = C# (chr# ((ord# ch -# 0xF8#                         `iShiftL#` 24#) +#
-                                         (ord# (indexCharOffAddr# addr (nh +# 1#)) `iShiftL#` 18#) +#
-                                         (ord# (indexCharOffAddr# addr (nh +# 2#)) `iShiftL#` 12#) +#
-                                         (ord# (indexCharOffAddr# addr (nh +# 3#)) `iShiftL#`  6#) +#
-                                         (ord# (indexCharOffAddr# addr (nh +# 4#))) -# 0x2082080#))
-                               : unpack (nh +# 5#)
-      | otherwise           = C# (chr# (((ord# ch -# 0xFC#)                        `iShiftL#` 30#) +#
-                                        ((ord# (indexCharOffAddr# addr (nh +# 1#)) -# 0x80#)
-                                                                                   `iShiftL#` 24#) +#
-                                         (ord# (indexCharOffAddr# addr (nh +# 2#)) `iShiftL#` 18#) +#
-                                         (ord# (indexCharOffAddr# addr (nh +# 3#)) `iShiftL#` 12#) +#
-                                         (ord# (indexCharOffAddr# addr (nh +# 4#)) `iShiftL#`  6#) +#
-                                         (ord# (indexCharOffAddr# addr (nh +# 5#))) -# 0x2082080#))
-                               : unpack (nh +# 6#)
+      | ch `leChar#` '\xDF'# =
+          C# (chr# ((ord# ch                                  -# 0xC0#) `iShiftL#`  6# +#
+                    (ord# (indexCharOffAddr# addr (nh +# 1#)) -# 0x80#))) :
+          unpack (nh +# 2#)
+      | ch `leChar#` '\xEF'# =
+          C# (chr# ((ord# ch                                  -# 0xE0#) `iShiftL#` 12# +#
+                    (ord# (indexCharOffAddr# addr (nh +# 1#)) -# 0x80#) `iShiftL#`  6# +#
+                    (ord# (indexCharOffAddr# addr (nh +# 2#)) -# 0x80#))) :
+          unpack (nh +# 3#)
+      | otherwise            =
+          C# (chr# ((ord# ch                                  -# 0xF0#) `iShiftL#` 18# +#
+                    (ord# (indexCharOffAddr# addr (nh +# 1#)) -# 0x80#) `iShiftL#` 12# +#
+                    (ord# (indexCharOffAddr# addr (nh +# 2#)) -# 0x80#) `iShiftL#`  6# +#
+                    (ord# (indexCharOffAddr# addr (nh +# 3#)) -# 0x80#))) :
+          unpack (nh +# 4#)
       where
 	ch = indexCharOffAddr# addr nh
 
diff --git a/ghc/lib/std/PrelEnum.lhs b/ghc/lib/std/PrelEnum.lhs
index 62f3167cec..0104e46f9b 100644
--- a/ghc/lib/std/PrelEnum.lhs
+++ b/ghc/lib/std/PrelEnum.lhs
@@ -1,5 +1,5 @@
 % -----------------------------------------------------------------------------
-% $Id: PrelEnum.lhs,v 1.12 2000/08/07 23:37:23 qrczak Exp $
+% $Id: PrelEnum.lhs,v 1.13 2001/02/18 14:45:15 qrczak Exp $
 %
 % (c) The University of Glasgow, 1992-2000
 %
@@ -180,11 +180,11 @@ instance Enum Ordering where
 \begin{code}
 instance  Bounded Char  where
     minBound =  '\0'
-    maxBound =  '\x7FFFFFFF'
+    maxBound =  '\x10FFFF'
 
 instance  Enum Char  where
     succ (C# c#)
-       | not (ord# c# ==# 0x7FFFFFFF#) = C# (chr# (ord# c# +# 1#))
+       | not (ord# c# ==# 0x10FFFF#) = C# (chr# (ord# c# +# 1#))
        | otherwise	        = error ("Prelude.Enum.Char.succ: bad argument")
     pred (C# c#)
        | not (ord# c# ==# 0#)   = C# (chr# (ord# c# -# 1#))
@@ -194,7 +194,7 @@ instance  Enum Char  where
     fromEnum = ord
 
     {-# INLINE enumFrom #-}
-    enumFrom (C# x) = eftChar (ord# x) 0x7FFFFFFF#
+    enumFrom (C# x) = eftChar (ord# x) 0x10FFFF#
 	-- Blarg: technically I guess enumFrom isn't strict!
 
     {-# INLINE enumFromTo #-}
@@ -235,13 +235,13 @@ eftCharList x y | x ># y    = []
 
 -- For enumFromThenTo we give up on inlining
 efdCharFB c n x1 x2
-  | delta >=# 0# = go_up_char_fb c n x1 delta 0x7FFFFFFF#
+  | delta >=# 0# = go_up_char_fb c n x1 delta 0x10FFFF#
   | otherwise    = go_dn_char_fb c n x1 delta 0#
   where
     delta = x2 -# x1
 
 efdCharList x1 x2
-  | delta >=# 0# = go_up_char_list x1 delta 0x7FFFFFFF#
+  | delta >=# 0# = go_up_char_list x1 delta 0x10FFFF#
   | otherwise    = go_dn_char_list x1 delta 0#
   where
     delta = x2 -# x1