Front page | perl.perl5.porters |
Postings from August 2013
On Thu Jul 04 07:59:28 2013, demerphq wrote:
> Found in perl 5.14.2 and in bleadperl, note the warning about constant
> in void context:
>
> $ ./perl -Ilib -lwe'BEGIN{eval "sub C() { q() }";} if ($x) { print 1 }
> elsif (C) { print 2 }'
> Useless use of a constant ("") in void context at -e line 1.
> Name "main::x" used only once: possible typo at -e line 1.
>
> Changing the constant to a 0 avoids the warning:
>
> $ ./perl -Ilib -lwe'BEGIN{eval "sub C() { 0 }";} if ($x) { print 1 }
> elsif (C) { print 2 }'
> Name "main::x" used only once: possible typo at -e line 1.
>
> Rafael points out that you dont need the constant for it:
>
> $ ./perl -Ilib -lwe'if ($x) { print 1 } elsif ("") { print 2 }'
> Useless use of a constant ("") in void context at -e line 1.
> Name "main::x" used only once: possible typo at -e line 1.
>
> And he explains that this is due to an optimization. Apparently we
> should special case *any* false constant, not just 1 and 0.
The newSTATEOP in the ELSIF block in perly.y stops newLOGOP from seeing
the constant on the lhs (just from reading code; unconfirmed).
I suspect this is actually a regression, caused by:
commit f6f3a1fea2bda9d33f9cc5367b630a21738c9def
Author: Rafael Garcia-Suarez <rgarciasuarez@gmail.com>
Date: Fri Apr 18 10:42:17 2008 +0000
Fix the line-number-in-elsif longstanding bug.
This patch does two things :
- newSTATEOP now nullifies the state op it
just created if OPf_SPECIAL is passed to it
in flags
- the parser now inserts a nullified stateop
in the expression block of an elsif
p4raw-id: //depot/perl@33710
I’m running a bisect now.
--
Father Chrysostomos
---
via perlbug: queue: perl5 status: open
https://rt.perl.org:443/rt3/Ticket/Display.html?id=118753
Thread Next