From 7e85a637700a5b134ac9381e811c0fc1da8cad8f Mon Sep 17 00:00:00 2001
From: simonmar <unknown>
Date: Wed, 27 Aug 2003 11:03:44 +0000
Subject: [PATCH] [project @ 2003-08-27 11:03:44 by simonmar] isSpace should
 only recognise Latin-1 whitespace characters, according to
 the report.

Perhaps we should have isUniSpace, too?
---
 GHC/Unicode.hsc |   23 +++++++++++------------
 1 file changed, 11 insertions(+), 12 deletions(-)

diff --git a/GHC/Unicode.hsc b/GHC/Unicode.hsc
index fc41060..7b84535 100644
--- a/GHC/Unicode.hsc
+++ b/GHC/Unicode.hsc
@@ -57,6 +57,16 @@ isPrint                 :: Char -> Bool
 -- | Selects white-space characters in the Latin-1 range.
 -- (In Unicode terms, this includes spaces and some control characters.)
 isSpace                 :: Char -> Bool
+-- isSpace includes non-breaking space
+-- Done with explicit equalities both for efficiency, and to avoid a tiresome
+-- recursion with GHC.List elem
+isSpace c		=  c == ' '	||
+			   c == '\t'	||
+			   c == '\n'	||
+			   c == '\r'	||
+			   c == '\f'	||
+			   c == '\v'	||
+			   c == '\xa0'
 
 -- | Selects alphabetic Unicode characters (letters) that are not lower-case.
 -- (In Unicode terms, this includes letters in upper and title cases,
@@ -119,7 +129,7 @@ type CInt = (#type int)
 isDigit    c = iswdigit (fromIntegral (ord c)) /= 0
 isAlpha    c = iswalpha (fromIntegral (ord c)) /= 0
 isAlphaNum c = iswalnum (fromIntegral (ord c)) /= 0
-isSpace    c = iswspace (fromIntegral (ord c)) /= 0
+--isSpace    c = iswspace (fromIntegral (ord c)) /= 0
 isControl  c = iswcntrl (fromIntegral (ord c)) /= 0
 isPrint    c = iswprint (fromIntegral (ord c)) /= 0
 isUpper    c = iswupper (fromIntegral (ord c)) /= 0
@@ -166,17 +176,6 @@ foreign import ccall unsafe "towupper"
 isControl c		=  c < ' ' || c >= '\DEL' && c <= '\x9f'
 isPrint c		=  not (isControl c)
 
--- isSpace includes non-breaking space
--- Done with explicit equalities both for efficiency, and to avoid a tiresome
--- recursion with GHC.List elem
-isSpace c		=  c == ' '	||
-			   c == '\t'	||
-			   c == '\n'	||
-			   c == '\r'	||
-			   c == '\f'	||
-			   c == '\v'	||
-			   c == '\xa0'
-
 -- The upper case ISO characters have the multiplication sign dumped
 -- randomly in the middle of the range.  Go figure.
 isUpper c		=  c >= 'A' && c <= 'Z' || 
-- 
1.7.10.4