GSI Forum
GSI Helmholtzzentrum für Schwerionenforschung

Home » PANDA » PandaRoot » Bugs, Fixes, Releases » Heavy Development on genfit
Heavy Development on genfit [message #4246] Tue, 15 May 2007 23:36 Go to next message
Sebastian Neubert is currently offline  Sebastian Neubert
Messages: 282
Registered: March 2006
Location: Munich
first-grade participant

From: *e18.physik.tu-muenchen.de
Hi everybody!

We are doing brain surgery on genfit at the moment to prepare for the arrival of GEANE and to provide a somewhat easier interface.

Please do NOT UPDATE genfit and recotasks until I tell you so.

This is inconvenient. But as (to my knowledge) we do not have something like a development branch I have to do it this way.

I would be happy about a more sofisticated svn-management than just everybody working on the head.

Cheers!


Sebastian Neubert
Technische Universität München
Department Physik E18
sneubert@e18.physik.tu-muenchen.de
tel: +49-8928912592
Re: Heavy Development on genfit [message #4247 is a reply to message #4246] Wed, 16 May 2007 00:25 Go to previous messageGo to next message
Johan Messchendorp is currently offline  Johan Messchendorp
Messages: 693
Registered: April 2007
Location: University of Groningen
first-grade participant

From: *xs4all.nl

... this could indeed be a good motivation to start with a "stable" and "development" branch in svn. Any opinions/experiences in that direction? And could that be easily implemented and activated (question to Mohammad....)?

Johan.



Johan Messchendorp
University of Groningen/KVI
Zernikelaan 25
NL-9747 AA Groningen
The Netherlands
tel. +31-503633558
fax +31-503634003
Re: Heavy Development on genfit [message #4257 is a reply to message #4247] Wed, 16 May 2007 15:49 Go to previous messageGo to next message
Florian Uhlig is currently offline  Florian Uhlig
Messages: 424
Registered: May 2007
first-grade participant
From: *gsi.de
I think it is a good idea to have a stable version of the code.
On the other side i don't understand why there is need to put
code back to SVN which is known not to work.
Is it not possible to test and develop the code localy and when
everything is fine put it back to SVN?

Ciao

Florian
Re: Heavy Development on genfit [message #4258 is a reply to message #4257] Wed, 16 May 2007 15:53 Go to previous messageGo to next message
Sebastian Neubert is currently offline  Sebastian Neubert
Messages: 282
Registered: March 2006
Location: Munich
first-grade participant

From: *e18.physik.tu-muenchen.de
Hi Florian!

No this is not possible.
I want to track the changes I make to the code in small steps. For this a version management system is there.
If you require people to only check in running code they will "never" check in.

However I agree that with present setup of svn it is very bad to commit something that does not even compile. Confused

Sebastian.


Sebastian Neubert
Technische Universität München
Department Physik E18
sneubert@e18.physik.tu-muenchen.de
tel: +49-8928912592
Re: Heavy Development on genfit [message #4259 is a reply to message #4258] Wed, 16 May 2007 15:55 Go to previous messageGo to next message
Sebastian Neubert is currently offline  Sebastian Neubert
Messages: 282
Registered: March 2006
Location: Munich
first-grade participant

From: *e18.physik.tu-muenchen.de
Hi!

I have just checked in a state (rev 754) where tpc/ recotasks/ and genfit/ all compile. So you can try to update.

However the operation is not finished and at the moment I have to debug because the kalman now is doing very strange things Sad

Stay tuned!

Sebastian.


Sebastian Neubert
Technische Universität München
Department Physik E18
sneubert@e18.physik.tu-muenchen.de
tel: +49-8928912592
Re: Heavy Development on genfit [message #4260 is a reply to message #4246] Wed, 16 May 2007 16:05 Go to previous messageGo to next message
Sebastian Neubert is currently offline  Sebastian Neubert
Messages: 282
Registered: March 2006
Location: Munich
first-grade participant

From: *e18.physik.tu-muenchen.de
Hi!

Another argument for checking in development code:

If somebody is to help me with the code I need to share my current status (containing the bug I need help with).

Sebastian.


Sebastian Neubert
Technische Universität München
Department Physik E18
sneubert@e18.physik.tu-muenchen.de
tel: +49-8928912592
icon1.gif  Re: Heavy Development on genfit [message #4261 is a reply to message #4259] Wed, 16 May 2007 16:08 Go to previous messageGo to next message
Florian Uhlig is currently offline  Florian Uhlig
Messages: 424
Registered: May 2007
first-grade participant
From: *gsi.de
Hi Sebastian

Thanks for your fast reply.
I understand that you want to check in the stuff bit by bit.
For the time beeing i will not compile the affected libraries.

Good luck for hunting the bugs.
Re: Heavy Development on genfit [message #4262 is a reply to message #4246] Wed, 16 May 2007 16:43 Go to previous messageGo to next message
Ralf Kliemt is currently offline  Ralf Kliemt
Messages: 507
Registered: May 2007
Location: GSI, Darmstadt
first-grade participant

From: 141.30.85*
Hi Sebasian,

A quick report:
Taking rev. 754 and compiling it I get the following:
/home/ralfk/Pandaroot/pandaroot/tpc/TpcSimplePatternRecoTask.cxx: In member function ‘virtual void TpcSimplePatternRecoTask::Exec(const Option_t*)’:
/home/ralfk/Pandaroot/pandaroot/tpc/TpcSimplePatternRecoTask.cxx:113: error: ‘class Track’ has no member named ‘addHitIndices’
/home/ralfk/Pandaroot/pandaroot/tpc/TpcSimplePatternRecoTask.cxx:122: error: ‘class Track’ has no member named ‘getHitIndices’
/home/ralfk/Pandaroot/pandaroot/tpc/TpcSimplePatternRecoTask.cxx:123: error: ‘class Track’ has no member named ‘getHitIndices’
/home/ralfk/Pandaroot/pandaroot/tpc/TpcSimplePatternRecoTask.cxx:157: error: ‘class Track’ has no member named ‘getDetIDs’

In the genfit/Track.h these accessors are commented.
Re: Heavy Development on genfit [message #4263 is a reply to message #4262] Wed, 16 May 2007 16:58 Go to previous messageGo to next message
Sebastian Neubert is currently offline  Sebastian Neubert
Messages: 282
Registered: March 2006
Location: Munich
first-grade participant

From: *e18.physik.tu-muenchen.de
Hi!

In this revision the TpcSimplePatternRecognitionTask should not be included in the makefile anymore. Probably I forgot to update the CMakelist.

Ah! I understand that this will get me in trouble. I am still using automake/conf. In such a case having two build systems is really nasty.

Hm...

Just take the "bad" classes out of the build please!

Sebastian.


Sebastian Neubert
Technische Universität München
Department Physik E18
sneubert@e18.physik.tu-muenchen.de
tel: +49-8928912592
Re: Heavy Development on genfit [message #4271 is a reply to message #4246] Fri, 18 May 2007 18:17 Go to previous messageGo to next message
Sebastian Neubert is currently offline  Sebastian Neubert
Messages: 282
Registered: March 2006
Location: Munich
first-grade participant

From: *e18.physik.tu-muenchen.de
Hi!

In genfit the Track object can now fill a TVirtualGeoTrack.
This is used in recotasks/TrackVisTask to visualize the tracks.

Unfortunately it was not possible to use the geometry provided by Cbm. There were segmentation violations when I tried this, that I could not debug. An Expert should look at this.

The Hit visualization has to be done yet.

Cheers! Sebastian.


Sebastian Neubert
Technische Universität München
Department Physik E18
sneubert@e18.physik.tu-muenchen.de
tel: +49-8928912592
Re: Heavy Development on genfit [message #4295 is a reply to message #4246] Tue, 22 May 2007 15:06 Go to previous messageGo to next message
Mohammad Al-Turany is currently offline  Mohammad Al-Turany
Messages: 518
Registered: April 2004
Location: GSI, Germany
first-grade participant
From: *gsi.de
Hi,

You should use TGeoTrack and not TVirtualGeoTrack, the later has no Draw implementation, it has only the dummy Draw implimentation of TObject!

regards

Mohammad
Re: Heavy Development on genfit [message #4297 is a reply to message #4295] Tue, 22 May 2007 15:53 Go to previous message
Sebastian Neubert is currently offline  Sebastian Neubert
Messages: 282
Registered: March 2006
Location: Munich
first-grade participant

From: *e18.physik.tu-muenchen.de
Hi Mohammad!

I tried TGeoTrack. It does not work. This is due to the special handling of GeoTracks in the GeoManager. You have to get a TVirtualGeoTrack-pointer from the GeoManager and then fill the object with data. Of course internally GeoManager will use TGeoTrack.

In the meanwhile I got it working. Didn't I post that?

Cheers! Sebastian.


Sebastian Neubert
Technische Universität München
Department Physik E18
sneubert@e18.physik.tu-muenchen.de
tel: +49-8928912592
Previous Topic: test
Next Topic: Addressing Hits by Array-index
Goto Forum:
  


Current Time: Mon Nov 25 08:48:59 CET 2024

Total time taken to generate the page: 0.00871 seconds