apr-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Garrett Rooney <roo...@electricjellyfish.net>
Subject Re: Gump failures on apr-iconv and apr-util since 2005-11-10
Date Tue, 20 Dec 2005 22:49:58 GMT
On 12/20/05, William A. Rowe, Jr. <wrowe@rowe-clan.net> wrote:
> Garrett Rooney wrote:
> >
> > The problem started when the Gump people upgraded to a new version of
> > GCC.  There's even a bug filed about this issue, but nobody has gotten
> > motivated enough to fix it yet.  In addition to fixing the actual
> > code, it would be nice if we could list apr-iconv as an optional
> > dependency, not a required one, since on unix systems it's virtually
> > never used anyway.  I believe there is a way to make this happen, but
> > again, nobody has gotten around to changing the Gump configuration.
>
> For the purposes of gump, this seems to be a very good thing, binding to
> apr-iconv.  What the end user does, binding to apr-iconv or the clib's own
> iconv symbols, is a trivial difference.

So you want gump to build a configuration that virtually no users
actually use?  That seems totally useless to me.  Gump should be
building and testing the system as it is intended to be used, if this
was a win32 build, I'd be all for it testing apr-iconv, but it's a
unix build, it should use the support for the platform iconv just like
a normal user would.

-garrett

Mime
View raw message