From 8e59ba46e26979cc11fa71e3f67aebbe6da4e8d6 Mon Sep 17 00:00:00 2001
From: Simon Marlow <simonmar@microsoft.com>
Date: Wed, 1 Mar 2006 11:35:36 +0000
Subject: [PATCH] Add support for Data.Char.generalCategory to libcompat

this is so that the stage1 compiler has proper support for Unicode.
Should fix these errors:

  lexical error in string/character literal at character '\8759'

when building the stage2 compiler.
---
 ghc/compiler/parser/Lexer.x      |   10 ++++---
 ghc/lib/compat/Compat/Unicode.hs |   57 ++++++++++++++++++++++++++++++++++++++
 ghc/lib/compat/Makefile          |    3 ++
 ghc/lib/compat/cbits/unicode.c   |    3 ++
 4 files changed, 69 insertions(+), 4 deletions(-)
 create mode 100644 ghc/lib/compat/Compat/Unicode.hs
 create mode 100644 ghc/lib/compat/cbits/unicode.c

diff --git a/ghc/compiler/parser/Lexer.x b/ghc/compiler/parser/Lexer.x
index 90fbf7a..6193c76 100644
--- a/ghc/compiler/parser/Lexer.x
+++ b/ghc/compiler/parser/Lexer.x
@@ -47,6 +47,12 @@ import DATA_BITS
 import Data.Char
 import Ratio
 --import TRACE
+
+#if __GLASGOW_HASKELL__ >= 605
+import Data.Char 	( GeneralCategory(..), generalCategory )
+#else
+import Compat.Unicode	( GeneralCategory(..), generalCategory )
+#endif
 }
 
 $unispace    = \x05
@@ -1182,9 +1188,6 @@ alexGetChar (AI loc ofs s)
 	other_graphic	= '\x6'
 
 	adj_c 
-#if __GLASGOW_HASKELL__ < 605
-	  = c  -- no Unicode support
-#else
 	  | c <= '\x06' = non_graphic
 	  | c <= '\xff' = c
 	  | otherwise = 
@@ -1213,7 +1216,6 @@ alexGetChar (AI loc ofs s)
 		  OtherSymbol           -> symbol
 		  Space                 -> space
 		  _other		-> non_graphic
-#endif
 
 -- This version does not squash unicode characters, it is used when
 -- lexing strings.
diff --git a/ghc/lib/compat/Compat/Unicode.hs b/ghc/lib/compat/Compat/Unicode.hs
new file mode 100644
index 0000000..4765511
--- /dev/null
+++ b/ghc/lib/compat/Compat/Unicode.hs
@@ -0,0 +1,57 @@
+{-# OPTIONS -cpp #-}
+module Compat.Unicode (
+    GeneralCategory(..), generalCategory,
+  ) where
+
+#if __GLASGOW_HASKELL__ > 604
+
+import Data.Char (GeneralCategory(..), generalCategory)
+
+#else
+
+import Foreign.C	( CInt )
+import Data.Char	( ord )
+
+-- | Unicode General Categories (column 2 of the UnicodeData table)
+-- in the order they are listed in the Unicode standard.
+
+data GeneralCategory
+        = UppercaseLetter       -- Lu  Letter, Uppercase
+        | LowercaseLetter       -- Ll  Letter, Lowercase
+        | TitlecaseLetter       -- Lt  Letter, Titlecase
+        | ModifierLetter        -- Lm  Letter, Modifier
+        | OtherLetter           -- Lo  Letter, Other
+        | NonSpacingMark        -- Mn  Mark, Non-Spacing
+        | SpacingCombiningMark  -- Mc  Mark, Spacing Combining
+        | EnclosingMark         -- Me  Mark, Enclosing
+        | DecimalNumber         -- Nd  Number, Decimal
+        | LetterNumber          -- Nl  Number, Letter
+        | OtherNumber           -- No  Number, Other
+        | ConnectorPunctuation  -- Pc  Punctuation, Connector
+        | DashPunctuation       -- Pd  Punctuation, Dash
+        | OpenPunctuation       -- Ps  Punctuation, Open
+        | ClosePunctuation      -- Pe  Punctuation, Close
+        | InitialQuote          -- Pi  Punctuation, Initial quote
+        | FinalQuote            -- Pf  Punctuation, Final quote
+        | OtherPunctuation      -- Po  Punctuation, Other
+        | MathSymbol            -- Sm  Symbol, Math
+        | CurrencySymbol        -- Sc  Symbol, Currency
+        | ModifierSymbol        -- Sk  Symbol, Modifier
+        | OtherSymbol           -- So  Symbol, Other
+        | Space                 -- Zs  Separator, Space
+        | LineSeparator         -- Zl  Separator, Line
+        | ParagraphSeparator    -- Zp  Separator, Paragraph
+        | Control               -- Cc  Other, Control
+        | Format                -- Cf  Other, Format
+        | Surrogate             -- Cs  Other, Surrogate
+        | PrivateUse            -- Co  Other, Private Use
+        | NotAssigned           -- Cn  Other, Not Assigned
+        deriving (Eq, Ord, Enum, Read, Show, Bounded)
+
+-- | Retrieves the general Unicode category of the character.
+generalCategory :: Char -> GeneralCategory
+generalCategory c = toEnum (wgencat (fromIntegral (ord c)))
+
+foreign import ccall unsafe "u_gencat"
+  wgencat :: CInt -> Int
+#endif
diff --git a/ghc/lib/compat/Makefile b/ghc/lib/compat/Makefile
index 06c6103..ae2f4ed 100644
--- a/ghc/lib/compat/Makefile
+++ b/ghc/lib/compat/Makefile
@@ -68,6 +68,9 @@ Distribution/ParseUtils.$(way_)o :  $(FPTOOLS_TOP)/libraries/Cabal/Distribution/
 Distribution/Compiler.$(way_)o :  $(FPTOOLS_TOP)/libraries/Cabal/Distribution/Compiler.hs
 Distribution/Version.$(way_)o :  $(FPTOOLS_TOP)/libraries/Cabal/Distribution/Version.hs
 Language/Haskell/Extension.$(way_)o :  $(FPTOOLS_TOP)/libraries/Cabal/Language/Haskell/Extension.hs
+cbits/unicode.o : $(FPTOOLS_TOP)/libraries/base/cbits/WCsubst.c $(FPTOOLS_TOP)/libraries/base/include/WCsubst.h
+
+SRC_CC_OPTS += -I$(FPTOOLS_TOP)/libraries/base/cbits -I$(FPTOOLS_TOP)/libraries/base/include
 
 # Make the #includes in the stubs independent of the current location
 SRC_HC_OPTS += -I$(FPTOOLS_TOP)/libraries
diff --git a/ghc/lib/compat/cbits/unicode.c b/ghc/lib/compat/cbits/unicode.c
new file mode 100644
index 0000000..0e0d1c5
--- /dev/null
+++ b/ghc/lib/compat/cbits/unicode.c
@@ -0,0 +1,3 @@
+#if __GLASGOW_HASKELL__ < 604
+#include "WCsubst.c"
+#endif
-- 
1.7.10.4