Hi Rene,
You are right : I mean version 3.02/05, sorry for the confusion.
However, the object I am writing (i.e. AliTrack) doesn't have a problem with
ownership. I also mentioned to you that for the ownership question of
TObjArrays I sometimes would need the functionality of TRefArray, but sometimes
I also need the TObjArray with ownership functionality. This depends on the
parameters the user has set for the instance of the object.
For this it would be very convenient if TRefArray would be identical to a
TObjArray when the SetOwner flag is set. From you I understood that this
is currently not the case (please correct me if I am wrong), therefore I am
now writing my own copy constructors for my classes, in which I can determine
myself whether I just copy the pointer addresses in the TObjArrays or really
also make copies of the objects referred to in the TObjArrays.
It is in this process that I saw that with the new schema evolution my private
copy constructors work correctly, whereas the TObject::Clone() invokes
a crash. As I said, I am still investigating and as soon as I fond something
I'll let you know.
Cheers,
Nick.
Rene Brun wrote:
>
> Hi Nick,
>
> I am afraid that you are just confusing people:
> - 3.02/06 is not yet available. Current version is 3.02/05
> - Your problems are not related to I/O, but to ownership in TObjArrays.
> - I gave to you a recipee yesterday for the question of ownersip.
>
> Rene Brun
>
> Nick van Eijndhoven wrote:
> >
> > Dear ROOTers,
> > I am also experiencing problems with the I/O in version 3.02/06.
> > Somehow writing of instances of one of my private classes fails with
> > 3.02/06 using the new schema evolution, whereas with the old streamer
> > method (i.e. no "+" in the Linkdef.h) everything works correctly.
> > I get the impression that the problem(s) are related to TObject::Clone(),
> > which uses the I/O streamers as far as I understnad.
> > I am still investigating things, but I also think that it would be
> > better to wait a bit before making 3.02/06 the pro version.
> >
> > Cheers,
> > Nick.
> >
> > --
> > Dr. Nick van Eijndhoven mailto:nick@phys.uu.nl http://www.phys.uu.nl/~nick
> > --------------------------------------------------------------------------
> > Org.: Utrecht University/Nikhef, Department of Subatomic Physics
> > Address: Princetonplein 5, NL-3584 CC Utrecht, The Netherlands
> > Phone: +31-30-2532331(direct) +31-30-2531492(secr.) Fax: +31-30-2518689
> > NIKHEF: +31-20-5922178(direct) +31-20-5922000(secr.) Fax: +31-20-5925155
> > CERN: +41-22-7679751(direct) +41-22-7675857(secr.) Fax: +41-22-7679480
> > Offices: Buys Ballot lab. 710 (Utrecht) N241 (Nikhef) B23 1-020 (CERN)
> > --------------------------------------------------------------------------
> >
> >
> > Valeri Fine wrote:
> > >
> > > Hello Rene,
> > >
> > > > We intend to release Root version 3.02/06 this week. This will be the PRO
> > > > release.
> > >
> > > He have discovered several problems with the last version of ROOT from CVS
> > >
> > > 1. We can not read back our files written half year ago (with ROOT 3.00.06)
> > > We have not had any problem to read them with ROOT versions 3.02.00 / 01
> > > 2. We can not use TChain to read several TTrees' in row (See Bill Love's report)
> > > The macro crashes occasionally
> > > 3. We can not read the new files with TVolume objects,
> > > but we can read the old versions of there files though.
> > >
> > > We are working to understand the reasons of these new problems and will appreciate you
> > > will wait a little with your release.
> > >
> > > Have you any idea what has happened with ROOT i/o recently, that may have affected the
> > > backward compatibility ?
> > >
> > > With my regards, Valeri
> > > -----
> > > Dr.Valeri Fine
> > > STAR/US Atlas E-mail: fine@bnl.gov
> > > Brookhaven National Lab Phone: +1 631 344 7806
> > > Upton, NY 11973-5000 FAX: +1 631 344 4206
> > > USA
This archive was generated by hypermail 2b29 : Tue Jan 01 2002 - 17:51:11 MET