Bug #123866 for XML-XSH: Can't use 'defined(@array)'

Tue Dec 12 13:24:01 2017 SREZIC [...] cpan.org - Ticket created

Subject:

Can't use 'defined(@array)'

The test suite fails with newer perls (>= 5.22): ... PERL_DL_NONLAZY=1 "/usr/perl5.22.3p/bin/perl" "-MExtUtils::Command::MM" "-MTest::Harness" "-e" "undef *Test::Harness::Switches; test_harness(0, 'blib/lib', 'blib/arch')" t/*.t Can't use 'defined(@array)' (Maybe you should just omit the defined()?) at /home/cpansand/.cpan/build/2017121212/XML-XSH-1.8.5-3/blib/lib/XML/XSH/Completion.pm line 111. Compilation failed in require at /home/cpansand/.cpan/build/2017121212/XML-XSH-1.8.5-3/blib/lib/XML/XSH.pm line 10. BEGIN failed--compilation aborted at /home/cpansand/.cpan/build/2017121212/XML-XSH-1.8.5-3/blib/lib/XML/XSH.pm line 10. Compilation failed in require at t/01basic.t line 164. BEGIN failed--compilation aborted at t/01basic.t line 164. # Failed test 1 in t/01basic.t at line 163 # t/01basic.t line 163 is: END { ok(0) unless $loaded; } t/01basic.t .. Dubious, test returned 255 (wstat 65280, 0xff00) Failed 67/67 subtests ...

Tue Dec 12 13:26:51 2017 CHOROBA [...] cpan.org - Correspondence added

The distribution is deprecated and not maintained. The last change added a skip for all tests when AUTOMATED_TESTING is true to avoid further messages from the testers. Ch.

Tue Dec 12 13:26:51 2017 The RT System itself - Status changed from 'new' to 'open'

Tue Dec 12 13:26:52 2017 CHOROBA [...] cpan.org - Status changed from 'open' to 'rejected'

Tue Dec 12 13:26:52 2017 CHOROBA [...] cpan.org - Fixed in 1.8.5 added

Tue Dec 12 13:34:58 2017 SREZIC [...] cpan.org - Correspondence added

On 2017-12-12 13:26:51, CHOROBA wrote: Show quoted text

> The distribution is deprecated and not maintained. The last change > added a skip for all tests when AUTOMATED_TESTING is true to avoid > further messages from the testers.

This won't work. AUTOMATED_TESTING is IMHO a misused variable, and for example I am not setting it in my CPAN.pm sessions --- which is configured to create and send test reports, regardless whether I use it as a "smoker" or not. The only difference about "smoking vs non-smoking" is whether I am about to use the tested module or not. I think the CPAN Testers Admin configuration page has a setting for disabling sending mails about specific CPAN modules. (Or maybe fix the problem, even though the module is deprecated, it seems like a very minor change...) Regards, Slaven

Tue Dec 12 14:04:10 2017 CHOROBA [...] cpan.org - Correspondence added

And another small change next time... I'd rather delete the distribution from CPAN completely. Ch. On Tue Dec 12 13:34:58 2017, SREZIC wrote: Show quoted text

> On 2017-12-12 13:26:51, CHOROBA wrote:

> > The distribution is deprecated and not maintained. The last change > > added a skip for all tests when AUTOMATED_TESTING is true to avoid > > further messages from the testers.

> > This won't work. AUTOMATED_TESTING is IMHO a misused variable, and for > example I am not setting it in my CPAN.pm sessions --- which is > configured to create and send test reports, regardless whether I use > it as a "smoker" or not. The only difference about "smoking vs non- > smoking" is whether I am about to use the tested module or not. > > I think the CPAN Testers Admin configuration page has a setting for > disabling sending mails about specific CPAN modules. > > (Or maybe fix the problem, even though the module is deprecated, it > seems like a very minor change...) > > Regards, > Slaven

Tue Dec 12 23:49:44 2017 ANDK [...] cpan.org - Correspondence added

If you really have no further interest in a package, you can transfer your first-come permissions to ADOPTME. See https://github.com/andk/pause/blob/master/doc/operating-model.md for more about how the PAUSE operates nowadays. This is probably a superior or at least a good second-best or accompanying way to mark a module as deprecated than telling people in tickets that the module is deprecated. Other options would be to edit the Changes file or to mention the deprecation right in the NAME section of the module. Both would help to spot the fact on metacpan.org. And/or to throw an exception on running the Makefile.PL. The latter would tell the enduser about the intentions of the current maintainer before he considers filing a bug report. The advantage of the ADOPTME variant is you do not have to upload any further version.