qpid-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Rupert Smith" <rupertlssm...@googlemail.com>
Subject Re: Apr version for C++ Build.
Date Wed, 07 Feb 2007 10:42:23 GMT
Yes, README.rhel3 is out of date. It referes to scripts that are no longer
there. It was usefull to me when doing a build on rh3 though, because the
other READMEs didn't mention make 3.8 which I had to upgrade to. README-dist
should probably mention make 3.8 perhaps in a "How to Build on RH3 "

The other thing wrong with the build instructions is that README-dist is not
included in the source distribution and README-dev is. In fact, you need to
read both in order to do a build and figure out how the instructions in both
these files need to be merged together. Might be better just to have a
single set of instructions in one file, with an optional section at the
begining decribing steps to be taken to build from a checkout and then
common steps to build from a source distribution. If this sounds like a good
idea, I volunteer to submit a patch for it while it is still fresh in my


On 2/6/07, Alan Conway <aconway@redhat.com> wrote:
> On Tue, 2007-02-06 at 17:26 +0000, Gordon Sim wrote:
> > Rupert Smith wrote:
> > > What version of apr does the C++ build use? In README-dist it claims
> > > 1.2.2but in
> > > README.rhel3 it claims 1.2.7. I guess either will do, but seems
> sensible to
> > > have a consistent version. I used 1.2.7.
> >
> > Can't answer the question authoritatively I'm afraid. I also use 1.2.7.
> >
> > The README.rhel3 was initially created prior to the autoconf based build
> >   and perhaps should no longer be there at all.
> There hasn't been any attempt to identify the earliest versions that
> qpid can work with, so the versions mentioned in the READMEs are usually
> just whatever was on the persons machine at the time they wrote the
> README. It may well be the case that older versions work (definitely is
> for some libs.)
> I didn't do the rhel3 README so I don't know if there's a particular
> reason it mentions a higher version, but choosing the higher version in
> case of doubt is a good plan.
> Cheers,
> Alan.

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message