Bug #102448 for Perl-Critic-Pulp: Bug 100291

Mon Mar 02 07:45:33 2015 eda [...] waniasset.com - Ticket created

Subject:	Bug 100291 - ProhibitNullStatements
Date:	Mon, 2 Mar 2015 12:45:20 +0000
To:	"'bug-Perl-Critic-Pulp [...] rt.cpan.org'" <bug-Perl-Critic-Pulp [...] rt.cpan.org>
From:	Ed Avis <eda [...] waniasset.com>

Thanks for looking into the bug https://rt.cpan.org/Public/Bug/Display.html?id=100291 I am not able to update it on the web interface so I comment here. It says that version 89 has some code to detect 'try' and 'catch', but I find that the following test case use TryCatch; try { $dbh = db_connect } exit 1; still produces the error Null statement (stray semicolon) [ValuesAndExpressions::ProhibitNullStatements] -- Ed Avis <eda@waniasset.com> Show quoted text

______________________________________________________________________ This email has been scanned by the Symantec Email Security.cloud service. For more information please visit http://www.symanteccloud.com

______________________________________________________________________

Tue Mar 03 04:08:37 2015 user42_kevin [...] yahoo.com.au - Correspondence added

Subject:	Re: [rt.cpan.org #102448] Bug 100291 - ProhibitNullStatements
Date:	Tue, 03 Mar 2015 20:04:00 +1100
To:	"Ed Avis via RT" <bug-Perl-Critic-Pulp [...] rt.cpan.org>
From:	Kevin Ryde <user42_kevin [...] yahoo.com.au>

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

> > use TryCatch; > try { $dbh = db_connect } > exit 1;

Thanks. ppi gives that as one "statement". I'll look for where the try actually ends. (The wrong report treats the "exit" as part of the try.)

Tue Mar 03 04:08:37 2015 The RT System itself - Status changed from 'new' to 'open'

Sat Mar 07 03:33:04 2015 user42_kevin [...] yahoo.com.au - Correspondence added

Subject:	Re: [rt.cpan.org #102448] Bug 100291 - ProhibitNullStatements
Date:	Sat, 07 Mar 2015 19:26:45 +1100
To:	"Ed Avis via RT" <bug-Perl-Critic-Pulp [...] rt.cpan.org>
From:	Kevin Ryde <user42_kevin [...] yahoo.com.au>

I applied a fix for version 90. It's imperfect since PPI puts everything after a try{} as part of that one statement. I imagine that will affect other bits of perlcritic too in subtle ways. Could think about lobbying PPI for some hook to let syntax-mangling modules declare their style or supply code which parses. Though alas I haven't had good response from PPI even for simple things. If extra PPI compound statement classes were needed then parts of perlcritic might have to recognise them, but at least expressions vs code vs etc would be parsed.

Sat Mar 07 11:18:14 2015 eda [...] waniasset.com - Correspondence added

Subject:	Re: [rt.cpan.org #102448] Bug 100291 - ProhibitNullStatements
Date:	Sat, 7 Mar 2015 16:18:01 +0000
To:	"bug-Perl-Critic-Pulp [...] rt.cpan.org" <bug-Perl-Critic-Pulp [...] rt.cpan.org>
From:	Ed Avis <eda [...] waniasset.com>

Thanks. I think it is time for try and catch to be added as first-class primitives to the Perl 5 language. Perhaps if the current work on adding decent OO (making Moose built in, effectively) is a success then the perl5-porters will look at try and catch next. Failing that, PPI could parse them as a special case. That sounds saner than the tar pit of letting modules hook into PPI for their own particular syntax mangles. Show quoted text

______________________________________________________________________ This email has been scanned by the Symantec Email Security.cloud service. For more information please visit http://www.symanteccloud.com

______________________________________________________________________

Wed Mar 11 02:16:17 2015 user42_kevin [...] yahoo.com.au - Correspondence added

Subject:	Re: [rt.cpan.org #102448] Bug 100291 - ProhibitNullStatements
Date:	Wed, 11 Mar 2015 17:08:43 +1100
To:	"Ed Avis via RT" <bug-Perl-Critic-Pulp [...] rt.cpan.org>
From:	Kevin Ryde <user42_kevin [...] yahoo.com.au>

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

> > (making Moose built in, effectively)

Hope not. I tried it but didn't enjoy it. Show quoted text

> Failing that, PPI could parse them as a special case.

Could do. I think only one of the three modules has a "finally" block, if I'm not mistaken, which difference might require something subtle.

Sun Jun 14 05:05:29 2015 KRYDE [...] cpan.org - Correspondence added

Believe the ProhibitNullStatements part of try/catch is reasonable now (though I'm starting to lose track of what else I'm supposed to be doing :-).

Sun Jun 14 05:05:31 2015 KRYDE [...] cpan.org - Status changed from 'open' to 'resolved'

Bug #102448 for Perl-Critic-Pulp: Bug 100291 - ProhibitNullStatements