Front page | perl.perl5.porters |
Postings from March 2016
On Thu, May 28, 2015 at 11:09:09AM +0100, Dave Mitchell wrote:
> On Wed, May 27, 2015 at 09:30:22PM +0100, Paul "LeoNerd" Evans wrote:
> > What's the conclusion here? Do I have to fix any code in ListUtils.xs
> > then?
>
> Well I think the provisional conclusion is that lc() et al are at fault.
> As to whether List::Util is doing anything wrong, the two differences
> between the first() implementation and perl's grep() implementation are:
>
> 1) grep temporarily increases the ref count of each SV it aliases to $_,
> via:
>
> # define DEFSV_set(sv) \
> (SvREFCNT_dec(GvSV(PL_defgv)), GvSV(PL_defgv) = SvREFCNT_inc(sv))
>
> 2) grep makes a mortal copy of each PADTMP it encounters.
>
> Either of those would make List::Util work with the bad lc/uc/etc in 5.20.0
> and 5.22.0.
>
> The second of those was to fix
>
> #78194: Referencing a PADTMP twice produces two copies
>
> (which I don't fully understand from a cursory read of the ticket).
>
> The first, if nothing else, avoids crashes if the thing being iterated
> is freed elsewhere, e.g.
>
> use List::Util qw{ first };
> use Devel::Peek;
> my $a = "A";
> @a = (1);
> @b = first { shift @a; Dump $_ } @a;
>
> which outputs:
>
> SV = UNKNOWN(0xff) (0x1b19038) at 0x1af8f88
> REFCNT = 0
> FLAGS = ()
I've just pushed a fix as smoke-me/davem/lc_fix. I propose that this
gets merged before 5.24.0 (despite the code freeze).
Here's the commit message:
commit 8603ce1e371b120cca063739d86827130f272354
Author: David Mitchell <davem@iabyn.com>
AuthorDate: Sun Mar 20 17:12:13 2016 +0000
Commit: David Mitchell <davem@iabyn.com>
CommitDate: Sun Mar 20 17:34:08 2016 +0000
stop lc() etc accidentally modifying in-place.
As an optimisation, [ul]c() and [ul]cfirst() sometimes modify their
argument in-place rather than returning a modified copy.
This should only be done when there is no possibility that the arg is
going to be reused. However, this fails:
use List::Util qw{ first };
my %hash = ( ASD => 1, ZXC => 2, QWE => 3, TYU => 4);
print first { lc $_ eq 'qwe' } keys %hash;
which prints "qwe" rather than "QWE".
Bascally everything in perl that sets $_ or $a/$b and calls a code block
or function, such as map, grep, for and, sort, either copies any PADTMPs,
turns off SvTEMP, and/or bumps the reference count.
List::Util doesn't do this, and it is likely that other CPAN modules
which do "set $_ and call a block" don't either.
This has been failing since 5.20.0: perl has been in-placing if the arg is
(SvTEMP && RC==1 && !mg) (due to v5.19.7-112-g5cd5e2d).
Make the optimisation critera stricter by always copying SvTEMPs.
It still allows the optimisation if the arg is a PADTMP - I don't know
whether this is unsafe too.
Perhaps we can think of something better after 5.24?
--
A power surge on the Bridge is rapidly and correctly diagnosed as a faulty
capacitor by the highly-trained and competent engineering staff.
-- Things That Never Happen in "Star Trek" #9
Thread Previous
|
Thread Next