Re: [Boost-users] [serialization] serialization of a class as pointer and non-pointer

Subject: Re: [Boost-users] [serialization] serialization of a class as pointer and non-pointer
From: François Mauger (mauger_at_[hidden])
Date: 2011-10-11 12:10:12

Hi Daniela,

 >>>On 10/10/2011 11:10, Daniela Lambertz wrote:
> Hello,
>
> I have a class Energy which is first serialized as a vector of pointers.
> Then, it is serialized multiple times as a non-pointer attribute. Here,
> the issue occurs.
> The first non-pointer Energy is serialized normally and all following
> non-pointer Energies are serialized as pointers to the first one. This
> leads to wrong values.

You don't give details on the reason why you (try to) use the lib
in this way but it looks a design error (see the "a) don't do that"
by robert)

maybe you could investigate the use of some shared pointers
and reimplement part of your work.

a workaround that *could* work is to use 2 archives to confine the 
memory tracking mechanism in 2 independant serialization regions:
  - the first one for the pointer-ed stuff
  - the other one for the non-pointer-ed one
but it means some additionnal management and care.

however, re-design should be considered first.

regards

frc
--
> When I tried another class which is not serialized as a pointer anywhere
> in the code, it worked fine.
> (I'm using non-intrusive serialization and I split the serialize-method
> into save and load.)
>
> Now in the Serialization tutorial, I found that the adresses of objects
> with the tracking trait "track_selectively" are tracked if "an object of
> this type is anywhere in the program serialized through a pointer".
> Can this be the reason for my Energy-problem? If yes, is there a way to
> avoid this behaviour?
>
> So far, it only works correctly if I transform this Energy attribute
> into a pointer in the save-method and serialize the pointer. But as my
> program is still in development, the problem might also turn up with
> other classes in the future and so it would be nice to have a general
> solution for this.
>
> Please tell me, if you need example code.
>
> Hope you can help me. Thanks in advance!
>
> Best regards
> Daniela Lambertz
>
> SMS schreiben mit WEB.DE FreeMail - einfach, schnell und
> kostenguenstig. Jetzt gleich testen! *http://f.web.de/?mc=021192*
>
>
>
> _______________________________________________
> Boost-users mailing list
> Boost-users_at_[hidden]
> http://lists.boost.org/mailman/listinfo.cgi/boost-users
-- 
François Mauger
   Groupe "Interactions Fondamentales et Nature du Neutrino"
   NEMO-3/SuperNEMO Collaboration
LPC Caen-CNRS/IN2P3-UCBN-ENSICAEN
Département de Physique -- Université de Caen Basse-Normandie
Adresse/address:
   Laboratoire de Physique Corpusculaire de Caen (UMR 6534)
   ENSICAEN
   6, Boulevard du Marechal Juin
   14050 CAEN Cedex
   FRANCE
Courriel/e-mail: mauger_at_[hidden]
Tél./phone:      [phone]
Fax:             02 31 45 25 49 / ([phone]