Mailing-List: contact dev-help@apr.apache.org; run by ezmlm
Precedence: bulk
Date: Thu, 5 Feb 2004 18:51:49 -0800
From: Greg Stein <gstein@lyra.org>
To: dev@apr.apache.org
Subject: Re: why gen-uri-delims in Python?
Message-ID: <20040205185149.C16331@lyra.org>
Mail-Followup-To: dev@apr.apache.org
References: <40228B20.5020604@attglobal.net> <20040205122755.B15359@lyra.org>
 <4022F40A.2000409@attglobal.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <4022F40A.2000409@attglobal.net>;
 from trawick@attglobal.net on Thu, Feb 05, 2004 at 08:55:22PM -0500

On Thu, Feb 05, 2004 at 08:55:22PM -0500, Jeff Trawick wrote:
> Greg Stein wrote:
>...
> > Seems like we need to keep that around, or find another solution. For
> > example, we could generate an ASCII table and an EBCDIC table, and then
> > use ./configure to select the appropriate table.
> 
> shrug...  the old way, or copy from uri_delims.h.{ascii|ebcdic} at configure 
> time; need something that uses a trivial operation that can be easily done in 
> Win32 build or NetWare build or some other odd build

All righty. Are we only talking about ASCII and EBCDIC? There isn't
anything else that will sneak up on us, right?

And aren't Win32 and NetWare always going to use the ASCII form? In which
case, we just hard-wire their build systems to point to the ASCII version.

Once I understand the set of options here, I'll make it happen since I
broke it in the first place :-)

To be honest, I'm thinking that we just hand-write the two tables into
apr_uri.c and use a #define to select the right one. We should already
have a symbol somewhere saying whether we're using EBCDIC. Then we can get
rid of all this delimiter generation stuff. These tables are quite
constant... :-)

Cheers,
-g

-- 
Greg Stein, http://www.lyra.org/