star-fst-l AT lists.bnl.gov
Subject: Star-fst-l mailing list
List archive
Re: [Star-fst-l] FST software issues in DEV library
- From: Te-Chuan Huang <tchuang.phys AT gmail.com>
- To: Xu Sun <sunxuhit AT gmail.com>
- Cc: Star-fst L <star-fst-l AT lists.bnl.gov>
- Subject: Re: [Star-fst-l] FST software issues in DEV library
- Date: Sat, 23 Apr 2022 00:28:52 +0800
Dear All,
I have send out the PR.
Dmitri asked where the geometry table (StarDb/Geometry/fst/fstSensorOnWedge.20211110.000001.C) used.
So, I check the StFstHitMaker which is the one got segmentation violation problem.
Then, I found below:
StFstHit *newHit = sensorHitCollection->hits()[idx];
double local[3];
double global[3];
local[0] = newHit->localPosition(0);
local[1] = newHit->localPosition(1);
local[2] = newHit->localPosition(2);
//Ye: simple transformation. Need to revisit
// The simple transformation will be updated with the geomtry table in database later
int sensorId = 1000 + ((int)newHit->getWedge() - 1) * kFstNumSensorsPerWedge + (int)newHit->getSensor();
TGeoHMatrix *geoMSensorOnGlobal = (TGeoHMatrix *) mSensorTransforms->FindObject(Form("R%04i", sensorId));
geoMSensorOnGlobal->LocalToMaster(local, global);
global[0] = local[0]*cos(local[1]);
global[1] = local[0]*sin(local[1]);
global[2] = local[2];
StThreeVectorF vecGlobal(global);
newHit->setPosition(vecGlobal); //set global position
In the comments, it says “updated with the geometry table in database”.
But it seems to me that the transformation from local to global coordinates was done twice (one using TGeoHMatrix and one using simple calculation).
Did I miss anything?
Cheers,
Te-Chuan
On Apr 22, 2022, at 8:46 AM, Xu Sun <sunxuhit AT gmail.com> wrote:Hi Te-Chuan,Thanks for the email.I just verified your change locally and they work fine.Please submit the PR when you have time.Best,XuOn Wed, Apr 20, 2022 at 12:18 PM Te-Chuan Huang <tchuang.phys AT gmail.com> wrote:Dear All,
After some investigations, I think I found the issues in DEV library.
There are two main issues causing FST codes failed to run.
First is on the chain options. The FST options are a little bit messed up:
There are duplicate options for FST.
And the StFstUtil should be loaded at very beginning along with the utilities of other detectors, for example the StIstUtil.
But, in DEV, it is the last one being loaded.
Second issue is in StarDb/Geometry/fst/fstSensorOnWedge.20211110.000001.C
The sensor id should be from 1000 to 1107, but in this macro, it was assigned from 1 to 108.
After these issues fixed, FST codes can be ran successfully in both 32bits and 64bits.
You can find the fixed codes at: /star/data01/pwg/tchuang/FST/Offline/Lib_dev/
If there is no other changes needed, I will make the PR on Github to fix these issues later this week.
Cheers,
Te-Chuan
_______________________________________________
Star-fst-l mailing list
Star-fst-l AT lists.bnl.gov
https://lists.bnl.gov/mailman/listinfo/star-fst-l
-
[Star-fst-l] FST software issues in DEV library,
Te-Chuan Huang, 04/20/2022
-
Re: [Star-fst-l] FST software issues in DEV library,
Xu Sun, 04/21/2022
- Re: [Star-fst-l] FST software issues in DEV library, Te-Chuan Huang, 04/22/2022
-
Re: [Star-fst-l] FST software issues in DEV library,
Xu Sun, 04/21/2022
Archive powered by MHonArc 2.6.24.