Hi Volker,
Thanks for reporting this problem when splitlevel > 2.
Now fixed in the CVS version.
Rene Brun
Volker Hejny wrote:
>
> Hallo,
>
> some time ago I asked about a problem in splitting of TLorentzVector.
> In the release notes of 3.01/06 it is written that this problem is
> solved now. However, I have still the following problem even using
> the current cvs version. Consider a class like the following:
>
> class test: public TLorentzVector {
> public:
> TVector3 fV;
> Double_t fTime;
>
> test() {};
> virtual ~test() {};
>
> ClassDef(test,1)
> };
>
> I compiled this and loaded it into root. When adding a branch to a
> tree (e.g. with Branch("test","test",&address,32000,99)), the components
> of the base TLorentzVector are read as 0. The new components in the
> derived class are processed correctly. Have I missed something, is
> this a new problem or is it just due to lack of knowledge from my side?
>
> (I tested exactly the class above using a *.cc file containing only
> the include and the ClassImp(test) statement.)
>
> Best regards,
> Volker
>
> --
> Dr. Volker Hejny Tel: 02461/616853 **
> Institut f. Kernphysik Fax: 02461/613930 **
> ---------------------------------------------------------------- ** ** ---
> Forschungszentrum Juelich GmbH, D-52425 Juelich **
This archive was generated by hypermail 2b29 : Tue Jan 01 2002 - 17:50:56 MET