Bug #118722 for Perl-Critic-Pulp: RequireFinalCut fires if =cut isn't preceded by a blank line

Fri Nov 11 12:48:04 2016 PETDANCE [...] cpan.org - Ticket created

Subject:

RequireFinalCut fires if =cut isn't preceded by a blank line

This code executes correctly: my $x = 199; =pod blah blah =cut print "$x\n"; There should be a blank line before the =cut, and pod2text complains properly: $ pod2text pod.pl pod.pl around line 2: =pod directives shouldn't be over one line long! Ignoring all 2 lines of content However, RequireFinalCut complains that there isn't a final =cut. I understand that there should be a blank line before the =cut, but the message should say that instead of saying that there isn't a =cut.

Sat Nov 12 02:46:53 2016 user42_kevin [...] yahoo.com.au - Correspondence added

Subject:	Re: [rt.cpan.org #118722] RequireFinalCut fires if =cut isn't preceded by a blank line
Date:	Sat, 12 Nov 2016 18:45:59 +1100
To:	"Andy Lester via RT" <bug-Perl-Critic-Pulp [...] rt.cpan.org>
From:	Kevin Ryde <user42_kevin [...] yahoo.com.au>

"Andy Lester via RT" <bug-Perl-Critic-Pulp@rt.cpan.org> writes: Show quoted text

> > I understand that there should be a blank line before the =cut,

Yes. Show quoted text

> but the message should say that instead of saying that there > isn't a =cut.

Hmm. I use Pod::Parser and it doesn't present such a =cut as a "cut without blank", rather gives text "blah blah =cut" arguments to the =pod command (which doesn't take arguments, per the warning message you saw). I'm not sure if it's a good idea for me to try to notice a cut which has not parsed. What do you think? I had intended letting podchecker cover parse problems (even possibly in a perlcritic policy if wanted within that framework). But maybe I should output parse error messages anyway so as not to silently let badness go by, even if they might be repeated a few times when successive policies parse. Hmm.

Sat Nov 12 02:46:53 2016 The RT System itself - Status changed from 'new' to 'open'

Sat Nov 12 23:58:59 2016 PETDANCE [...] cpan.org - Correspondence added

Not sure what to do. Not sure it's worth putting out two different error messages. Maybe modify the existing one to cover the cause of "=cut without a preceding newline"? All I know is that when I first got the error I was a little confused because clearly there's that =cut and so what's it complaining about?

Thu Nov 24 06:00:00 2016 user42_kevin [...] yahoo.com.au - Correspondence added

Subject:	Re: [rt.cpan.org #118722] RequireFinalCut fires if =cut isn't preceded by a blank line
Date:	Thu, 24 Nov 2016 21:58:08 +1100
To:	"Andy Lester via RT" <bug-Perl-Critic-Pulp [...] rt.cpan.org>
From:	Kevin Ryde <user42_kevin [...] yahoo.com.au>

"Andy Lester via RT" <bug-Perl-Critic-Pulp@rt.cpan.org> writes: Show quoted text

> > All I know is that when I first got the error I was a little confused > because clearly there's that =cut and so what's it complaining about?

On balance, yep. I uploaded a version 91 allowing for no blank line. As you say, that's how it looks, and how perl executes. podchecker or another policy can report whether pod without blanks is in fact a good idea. -- I played a country and western song backwards to listen for evil messages. It sounded just the same except it was about a fella whose wife stood by him while he brought in a successful harvest and his dog recovered from a long illness.

Thu May 14 03:21:41 2020 KRYDE [...] cpan.org - Status changed from 'open' to 'resolved'

Thu May 14 03:21:42 2020 KRYDE [...] cpan.org - Fixed in 91 added