cvs commit: fptools/libraries/base/GHC Unicode.hsfptools/libraries/base/cbits WCsubst.cfptools/libraries/base/include WCsubst.h HsBase.h

[Simon Marlow]

On 14 March 2005 14:08, Ross Paterson wrote:

> On Mon, Mar 14, 2005 at 04:18:08AM -0800, Simon Marlow wrote:
>> simonmar    2005/03/14 04:18:08 PST
>> 
>>   Modified files:
>>     libraries/base/GHC   Unicode.hs
>>     libraries/base/include HsBase.h
>>   Added files:
>>     libraries/base/cbits WCsubst.c
>>     libraries/base/include WCsubst.h
>>   Log:
>>   Add Dimitry Golubovsky <dimitry at golubovsky.org>'s Unicode character
>>   class implementation.  This will remove the dependency on libc's
>>   locale code and give us much more consistent support for Unicode
>>   across platforms.
> 
> Might it make sense for Data.Char to re-export GeneralCategory,
> generalCategory and toTitle?
>
> The new predicates could be moved to Data.Char.
 
Sure.  Will Hugs now be using the version of this code in
libraries/base?

> isDigit returns True for all decimal numbers (Nd), not just the
> documented [0-9].  Won't this break readDec and readHex?  If there
> is a predicate selecting Nd, you also need a function giving the
> value of a digit, as well as a predicate for ASCII digits.

Ah yes, I thought there was a good reason to have isDigit only recognise
0..9.  I'll make that change; we have isNumber to recognise the whole Nd
class.

> isPrint returns True for unassigned code points (Cn).

Dimitry?

Cheers,	
	Simon