This distribution has been tested as part of the cpan-testers
effort to test as many new uploads to CPAN as possible. See
http://testers.cpan.org/

Please cc any replies to cpan-testers@perl.org to keep other
test volunteers informed and to prevent any duplicate effort.

--
Dear Nathan Gray,

This is a computer-generated report for Class-Inflate-0.05
on perl-5.6.2, created automatically by CPAN-Reporter-1.02
and sent to the CPAN Testers mailing list.

If you have received this email directly, it is because the person testing
your distribution chose to send a copy to your CPAN email address; there
may be a delay before the official report is received and processed
by CPAN Testers.

Thank you for uploading your work to CPAN. However, it appears that
there were some problems with your distribution. If these results are
not what you expect, please consult "Notes for CPAN Authors" on
the CPAN Testers Wiki: http://cpantest.grango.org

Sections of this report:

On Wed, Oct 24, 2007 at 05:09:04PM +0100, david@cantrell.org.uk wrote:

Dear Nathan -- I see two things going on.

1) I think that CPAN.pm might have a bug where it's not finding
prerequisites because the PL failed. In this case, you have
Makefile.PL running Build.PL. There's some funky logic in CPAN.pm to
know when/where it can get prerequisites and maybe the use of
Module::Build::Compat is confusing it. (Andreas -- what about reading
YAML regardless of the result of the PL phase?)

2) Your Build.PL subclass has "use DBI" -- but that code gets loaded
by M::B to create your subclass. So you're guaranteed to fail if DBI
isn't installed. This isn't something that "build_requires" can
address -- build_requires are only satisfied after the PL file runs.
Instead, you have a few options:

* Use the bleeding_edge "configure_requires" option in META.yml, which
isn't widely supported yet, that is intended to get things installed
before calling Makefile.PL or Build.PL

* Change your "use DBI" to "require DBI" and put that inside your
ACTION_test subroutine so it isn't called until the test phase (when
presumably, your prerequisites pull it in).

On Thu, Oct 25, 2007 at 07:09:01AM -0400, David Golden wrote:

This sounds like a good solution.

If PL fails we never proceed any further, right? I think we should not
change that, it would just increase the number of possible paths
through the installation procedure. And variants need maintainance and
all that.

Or am I missing something?

On 10/26/07, Andreas J. Koenig wrote:

For Class-Inflate, prereqs weren't found by CPAN::Reporter calling
prereq_pm on the distribution object and I'm not sure why.

I'm suggesting that if a META.yml exists, then prereq_pm() should be
able to get prerequisites regardless of whether *.PL has run (unless
the dynamic flag is set in META.yml). I'm not sure that's the current
logic.

It's possible that the because Class-Inflate is using Makefile.Pl with
Module::Build::Compat, this is somehow confusing prereq_pm().

David

Adam invented configure_requires for a real reason. If there is a
requirement before the PL file can be run, only configure_requires is
a possible solution. Now we must tell people to use it whenever they
have exactly this problem. If they do not have this problem then they
shall write their Makefile/Build and let the rest of the chain do its
duty.