SQL::Statement unusable in major Linux distributions

[Paul Beardsell]

I don't believe that the Debian packagers would have taken SQL::Statement
1.22 if they had known about CPAN bug 50788.  At CPAN 50788 was repeatedly
suppressed and the maintainer refused to conduct the supplied test on 1.22.

Debian: I think 1.22 should be removed from Debian testing pending a fix.
There may be lots of nice extra functionality there, but the basics are
broken.  A SQL engine where a simple UPDATE fails is not fit for purpose.

Every other distro is in the same unenviable position as the Debian distro
compilers.  I will continue to report non-distro specific bugs to CPAN.

Paul Beardsell
Paul at Beardsell.com


2009/10/27 Jens Rehsack <rehsack at googlemail.com>

> 2009/10/26 Paul Beardsell <paul at beardsell.com>:
> > 2009/10/26 Jonathan Yu <jonathan.i.yu at gmail.com>
> >>
> >> Hi:
> >>
> >> I'm not sure why all this discussion is happening on the Debian list.
> >> I've not seen much that has been discussed that is Debian-specific.
> >>
> >> Note that unstable has version 1.22 of this module, which is the most
> >> recent version as upstream. Debian users can always install older
> >> versions of the package if they so choose (see the apt-get manpage).
> >>
> > I agree, not Debian specific, but reporting of bugs in so-called "old"
> > versions of libsql-statement-perl [SQL::Statement]] (even though the bugs
> > are in in stable here) are strongly discouraged by the package maintainer
> at
> > CPAN.
>
> Paul has caught the point. As the maintainers of the CPAN module we
> are not responsible for sloppy update cycles for what reason ever in any
> distributions (Linux and others), so we decline to accept bugs for module
> releases which are depreciated from our point of view. On the other hand,
> we're happy to work together with the packaging maintainers of any
> distribution to help choosing stable versions (in case of S:S I think,
> the packager saw the failed tests of 1.17 .. 1.20 and didn't took
> deeper). Same could be for 1.24 of Sys::Filesystem (I added new tests
> to get feedback via SmokeTest - but this doesn't tell any about whether
> the reported errors where in the module since ages).
>
> Paul now tries to figure out, what would be the very best way to report
> such a problem next time and I think he hopes for a statement from
> the Debian perl package maintainers.
>
> /Jens
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/pkg-perl-maintainers/attachments/20091027/def52da1/attachment.htm>