Mailing-List: contact dev-help@apr.apache.org; run by ezmlm
Precedence: bulk
Errors-To: <wrowe@rowe-clan.net>
Message-Id: <5.1.0.14.2.20020718123947.02bc35b8@pop3.rowe-clan.net>
Date: Thu, 18 Jul 2002 12:52:11 -0500
To: kfogel@collab.net
From: "William A. Rowe, Jr." <wrowe@rowe-clan.net>
Subject: Re: Okay, two questions for Bill Rowe
Cc: SVN Dev List <dev@subversion.tigris.org>, dev@apr.apache.org
In-Reply-To: <85fzyh54zv.fsf@newton.ch.collab.net>
References: <86ptxmbbfb.fsf@kepler.ch.collab.net>
 <86ptxmbbfb.fsf@kepler.ch.collab.net>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed

At 12:18 PM 7/18/2002, Karl Fogel wrote:
>Bill,
>
>    - What's the status of the new apr-iconv library?

The new apr-iconv library needs to be taught to consume apr.  I'll
do the Win32 work if someone will tackle the build side of Unix.

Note the end app [SVN] shouldn't be aware of it.  We need one
more macro, something like APR_HAS_APR_ICONV inside of
the apu_private or apu header.

>    - What's status on the move of apr_xlate* to apr-util?

Finished, with the exception of the APR_CHECK_ICONV_INBUF
m4 macro that tests for the iconv prototype differences, and one
more hack to use our apr_* namespace protected entry points
from xlate.c if APR_HAS_APR_ICONV is toggled.  We also need
that symbol set up in apu or apu_private.

The first macro sticks the #define APR_ICONV_INBUF_CONST 1
into apr_private.h which we don't see inside of apu_private.h, so
we can't pick it up inside of apr-util/xlate/xlate.c.

If someone who groks m4 much better than I would get this inserted
into apu_private.h from apu_private.h.in ... we would be done with the
new unix / using installed xlate problems.

>I'm asking for the Subversion Alpha release -- we'll need to know if
>we can base the release on HEAD of apr and friends, or if we should
>use last Monday's APR and ship the GNU iconv library with our Win32
>distributions.

You all know the ASF position on shipping GNU code, just be aware that
will be a fork from any Apache releases.  I'm trying to get into a position
where -we- build mod_charset_lite on Win32 and are in a position to ship
that example module.

>We'd also been thinking of making an interim release today.  It looks
>like for that we'd *have* to go with the last-Monday-etc solution
>above, right?

Unless our unix problem with APR_CHECK_ICONV_INBUF is fixed, yes.
If that gets fixed, you should be good for cvs HEAD on APR.

Bill