Re: [ROOT] TGeant3TGeo::IsRootGeometrySupported returns false?

From: Rene Brun (brun@pcbrun.cern.ch)
Date: Wed Dec 22 2004 - 10:16:02 MET


Hi Christian,

On Wed, 22 Dec 2004, Christian Holm Christensen wrote:

> Hi Rene,
> 
> On Tue, 2004-12-21 at 16:42 +0100, Rene Brun wrote:
> > Hi Christian,
> > 
> > You are right. I have added this function in the class TGeantTGeo.
> 
> Great, Thanks. 
> 
> I was wondering, whould it be possible to split out the GUI classes from
> the libgeant321.so library, so that one can use that library directly in
> batch jobs?  

We had to reintroduce these classes following some user requests.
I am personally in favour of removing these classes. To be seen with the
next iteration.
Meanwhile, simply specify teh option "-b" when you run your application.

> 
> Also, I think the SUBROUTINEs G3LUND, G3LUNDI and  the lines 
> 
>       ELSE IF(CHCOMM.EQ.'GCLUND') THEN
>          IADD=GCADDI(IFLUND)
> 
> in comad/gcomad.F should be taken out of the library.  It seems a bit
> harsh to require that one load the libPythia6 library before using
> TGeant3.  After all, the event generation is moved to
> TVirtualMCApplication, and the SUBROUTINEs are not exported. 

I will look into this after the holidays.
I still have to document the new system of classes TGeant3/TGeantTGeo
as discussed during the recent VMC workshop at CERN.

> 
> Finally, why are the dummy routines not linked directly into the
> library?   Or perhaps the libgeant321.so should link explicitly to the
> libdummies.so library. 

These routines cannot be in the library. When the class TGeant3f77 is 
used (case of a geant3 fortran application using libgeant321) the 
application implements these routines.

Rene

>  > My 2 cents. 
> 
> Yours,
> 
> 



This archive was generated by hypermail 2b29 : Sun Jan 02 2005 - 05:50:10 MET