James E Keenan wrote:
Will Coleda wrote:
I wrote a program (tools/dev/branch_status.pl) to give us some summary
information about existing branches, and put the output up at
Coke:
This was not passing t/codingstd/perlcritic.t. However, by applying the
following patch, I was able to get it to pass:
I should clarify that the file was passing perlcritic.t but was
generating a non-numeric value warning:
$ perl t/codingstd/perlcritic.t tools/dev/branch_status.pl
1..1
Argument "5.010_000" isn't numeric in subroutine entry at
/usr/local/lib/perl5/site_perl/5.10.0/Perl/Critic/Document.pm line 139.
ok 1 - Test::Perl::Critic for "tools/dev/branch_status.pl"
Index: tools/dev/branch_status.pl
===================================================================
--- tools/dev/branch_status.pl (revision 38625)
+++ tools/dev/branch_status.pl (working copy)
@@ -8,11 +8,9 @@
=cut
-## Modern::Perl (doesn't pass perlcritic)
-use 5.010_000;
+use v5.10.0;
use strict;
use warnings;
-use feature();
use XML::Twig;
use Perl6::Form;
I think this is more a bug in Perl::Critic than in perlcritic.t. Cf.:
https://rt.cpan.org/Ticket/Display.html?id=45892.
I subsequently got a reply from Eliott Shank++ in this ticket:
"Yes, version.pm returns "5.010_000" for this case, which, outside of
code compilation, isn't considered numeric by perl. The opinion has
been that using underscores in version numbers indicates a developer
version, which perl 5.010000 is not."
I confirmed that 'use 5.010000;' would enable the file to pass
t/codingstd/perlcritic.t as well.
Thank you very much.
kid51
_______________________________________________
http://lists.parrot.org/mailman/listinfo/parrot-dev
