Re: cpan test redux...better, but... still shouldn't be on CPAN

"mem" is particularly bad name for a pragma, "mem" has nothing to do with memory; markloaded or markused sound better (and like they should be part of Module::Loaded)

P is a particularly bad name for a module, its a single letter top level namespace . So it exports a function of the same name P ( and Pe), doesn't make P a good module name -- If you need it available from a commandline consider your own Toolkit

PAUSE: pause_namingmodules On The Naming of Modules

CPAN Testers Reports: Report for P-1.1.4 shows a different failure mode (bugs in your code)

So now I'm getting 'mem' being both present and non-present in the same test. This doesn't make sense.

Funny, and then what happened?

If your test suite isn't telling you what's going on, improve it

I did contact the tester with all the fails ... and asked if they could think of anything that might be causing this... will have to see if they have ideas or even respond (they might be busy or something)...

When I was a tester, I did not appreciate being contacted like this; cpantesters isn't perlmonks/stackoverflow ... they run your test suite and submit reports automatically, most everything else is up to you; If your tests don't tell you enough about what the problem is, write better tests "Can't CPAN Testers read? I have detailed the format I require for bug reports!"

Just something to keep in mind because excessive "spam" from authors can drive some cpantesters away , its partially why I quit cpantesters

Got email back from user. They said it looked like the program under test ...

I just read this part of your post, well /dev/null isn't portable for starters ...

ddumper^{Basic debugging checklist} Basic debugging checklist item 4 works for debugging even in test suites

Thats all the interest I've managed to muster, I'm fickle lately