develooper Front page | perl.perl5.porters | Postings from February 2012

Re: Failure in porting/utils.t

Thread Previous
From:
Karl Williamson
Date:
February 19, 2012 07:42
Subject:
Re: Failure in porting/utils.t
Message ID:
4F411850.4090107@khwilliamson.com
On 02/18/2012 11:56 PM, Todd Rinaldo wrote:
>
> On Jan 29, 2012, at 3:54 PM, Karl Williamson wrote:
>
>> On 01/29/2012 02:50 PM, Father Chrysostomos wrote:
>>>
>>> On Jan 29, 2012, at 12:54 PM, Karl Williamson wrote:
>>>
>>>> On 01/29/2012 01:36 PM, Father Chrysostomos wrote:
>>>>>
>>>>> On Jan 29, 2012, at 11:10 AM, Karl Williamson wrote:
>>>>>
>>>>>> I am suddenly getting failures in porting/utils.t.  A sample is:
>>>>>>
>>>>>> # Failed at porting/utils.t line 81
>>>>>> #      got "Use of uninitialized value $mask in vec at ../lib/warnings.pm line 391.\nUse of uninitialized value $mask in bitwise or (|) at ../lib/warnings.pm line 397.\nUse of uninitialized value $mask in vec at ../lib/warnings.pm line 391.\nUse of uninitialized value $mask in bitwise or (|) at ../lib/warnings.pm line 397.\nUse of uninitialized value $mask in vec at ../lib/warnings.pm line 407.\nUse of uninitialized value $mask in bitwise and (&) at ../lib/warnings.pm line 419.\nPorting/bisect.pl syntax OK\n"
>>>>>>
>>>>>>
>>>>>> A bisect yields this:
>>>>>> commit 7e4f04509c6d4e8d2ed0e31eaf59004e5c930b39
>>>>>> Author: Father Chrysostomos<sprout@cpan.org>
>>>>>> Date:   Thu Jan 26 20:43:17 2012 -0800
>>>>>>
>>>>>>     Allow ${^WARNING_BITS} to turn off lexical warnings
>>>>>>
>>>>>>
>>>>>> However, I didn't get that until today, and I have downloaded blead several times since that commit was made, including earlier today.  So I don't understand how it is suddenly showing up.
>>>>>
>>>>> Do you need to re-run regen/warnings.pl?
>>>>>
>>>>> That’s the only thing I can think of.  Perhaps it got out of synch at your end somehow.
>>>>>
>>>>>
>>>>
>>>>
>>>> Nothing changes when I run that; so that doesn't explain things.
>>>
>>> Does line 389 of warnings.pm contain //, as shown below?
>>>
>>>      385 sub import
>>>      386 {
>>>      387     shift;
>>>      388
>>>      389     my $mask = ${^WARNING_BITS} // ($^W ? $Bits{all} : $NONE) ;
>>>      390
>>>      391     if (vec($mask, $Offsets{'all'}, 1)) {
>>>      392         $mask |= $Bits{'all'} ;
>>>      393         $mask |= $DeadBits{'all'} if vec($mask, $Offsets{'all'}+1, 1);
>>>      394     }
>>>      395
>>>      396     # Empty @_ is equivalent to @_ = 'all' ;
>>>      397     ${^WARNING_BITS} = @_ ? _bits($mask, @_) : $mask | $Bits{all} ;
>>>      398 }
>>>
>>>
>>
>>
>> Yes:
>>     my $mask = ${^WARNING_BITS} // ($^W ? $Bits{all} : $NONE) ;
>
> Ditto.
>
> I'm seeing this from warning also from common::sense. Did we ever get to the bottom of this?
>
> Todd
>


They magically stopped happening for me at some point.

Thread Previous


nntp.perl.org: Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at ask@perl.org | Group listing | About