I think most of these changes are fine. I had the non-persistent pointers already i my own versions. We are now writing objects of this kind to file (root), so it becomes crucial to either fix object defs and/or write appropriate streamer functions. I guess when we come to next root versions some of this may help, but until then. Unless the streamer fct is written the OLD files cannot be read!!!!!! > Dear all, > > I hope you can get this message at BNL, I'm sending the message to > Flemming as well in case there's still some problems. > > After my previous mail about some memory leak, I've cleaned up a bit some > track classes : > > - BrMatchedTrack : > fHelix is now a pointer to BrHelix > modified methods in consequence (getter and setter) > removed some useless private members (fExitX, Y, Z and > fEntranceX, Y, Z), instead I use a BrVector3D fEntrance and fExit. > The setters and getters GetEntranceX, Y, Z etc. use now this > BrVector3D. Therefore, there shouldn't be any bad consequence on > existing codes. If so, sorry and tell me asap. > > - BrModuleMatchTrack : > did the same, replaced helix object by a pointer to BrHelix and > modified methods consequently. > > - BrSpectrometerTrack.cxx : > remove fHelix (pointer to BrHelix) in the destructor of > BrMatchedTrack. > > I've tagged this brat version 1.12.3 so that you can still use the > previous one. > > In case you face some problems with the version I've committed, tell me! > It compiled well but couldn't make some quick tests. > > Djam > > :o) ****************************************************** )o: > |ME : Djamel Ouerdane EMAIL : ouerdane@nbi.dk | > |OFFICE : Tb1 @ NBI PHONE : +45 353 252 69 (office) | > |INST : Niels Bohr Institute, | > | Blegdamsvej 17, 2100 København Ø, Danmark | > :oD ****************************************************** \o: > >
This archive was generated by hypermail 2b29 : Mon Dec 18 2000 - 22:04:58 EST