[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: ready for another Debian upload (was: Re: PSPP back in Debian)
From: |
bojo42 |
Subject: |
Re: ready for another Debian upload (was: Re: PSPP back in Debian) |
Date: |
Mon, 05 Mar 2012 01:41:21 +0100 |
Done, new version ready for push at mentors (with removed hurd-i386
build dependency).
The ftpmasters also removed all former packages with successful builds
on testsuite failures like i requested (or to be precisly they just
removed all pspp packages from sid ;)
So when the all the fixes will work as expected we'll have a pretty
cleaned up package in Debian, that will migrate to Wheezy after ten
days. Wasn't the shortest journey, but now we quite have the best PSPP
in Debian ever ;) Just hope we get still get it into Ubuntu for 12.04
LTS, would be a nice reconquered userbase for your all your newer work.
Cheers and a late good night from Europe.
Bojo42
Am Samstag, den 03.03.2012, 05:47 +0000 schrieb John Darrington:
> On Fri, Mar 02, 2012 at 09:24:26PM -0800, Ben Pfaff wrote:
>
> I fixed all the ia64 test failures tonight.
>
> bojo42, please generate a new Debian package from the tip of
> master when you have a chance. With a little luck, I think PSPP
> might pass "make check" on all the Debian architectures on the
> next try.
>
> Thanks,
>
> Ben.
> --
> Ben Pfaff
> http://benpfaff.org
>
>
> If I could make a suggestion: It seems that the hurd-i386 architecture
> is being omitted for the lack of a postgres server on hurd. Postgres
> is lacking because Hurd doesn't have SysV IPC - and is not likely to
> for some time. However, it seems a shame that PSPP should not have
> a Hurd package for such a trivial reason.
>
> There is no reason that PSPP actually needs a Postgres server to run
> or build. It's only needed for that one test. Users on Hurd can still
> make use of PSPP's postgres features (they'll just have to connect to
> a database running on another machine).
>
> Is it possible to adjust the build dependencies such that Hurd-i386
> does not depend on Postgresql? That way, I beleive the test will
> just be skipped in the make check stage.
>
> J'
>