develooper Front page | perl.perl5.porters | Postings from May 2013

Re: Storable refactoring, was Re: [perl #118139] Storable in DESTROY blocks

Thread Previous | Thread Next
From:
Leon Timmermans
Date:
May 29, 2013 17:40
Subject:
Re: Storable refactoring, was Re: [perl #118139] Storable in DESTROY blocks
Message ID:
CAHhgV8gNPveKhq9SqqUg2UD0c2kqCwaJo2d9jCTjvm+_bLZt3w@mail.gmail.com
On Wed, May 29, 2013 at 7:22 PM, Salvador Fandino <sfandino@yahoo.com> wrote:
> I have been working for some time in refactoring Storable. See
>
>   https://github.com/salva/perl5/tree/improve_storable
>
> And a wrapped version for CPAN distribution:
>
>   https://github.com/salva/p5-Storable
>
> It is still a work in progress and still has some bugs (more notably it
> doesn't work on 32 bit platforms where perl has been compiled with
> use64bitint set).
>
> Initially my plan was to send the patch to p5p once those bugs have been
> solved but as this bug report is here and people is already trying to
> solve it, I think is a good time to tell the world about my work on
> Storable :-)
>
> With my changes, the context is created when store or freeze functions
> are called on the C stack and passed from function to function when
> recurring. The temporal structures are mortalized so that they are freed
> when the code comes back to Perl-land.
>
> I might have introduced new bugs but IMO, now, the code is pretty much
> simpler and bugs should also be much easier to fix than before.

Sounds good.

> If you agree that having that into blead is a good idea, I think that
> the way to follow is to release it to CPAN as a development version
> first so that it can be tested (I have been testing it myself with all
> the latest perls on the 5.6 5.8, 5.10, 5.12, 5.14, 5.16, 5.18 and 5.19
> series with and without threads)

A CPAN smoke may be helpful :-)

Leon

Thread Previous | Thread Next


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