first implementation of StripCluster EDM class
A EDM class with no dependency on Athena for describing the strip clusters in FASER. It currently stores local position, local covariance matrix, a pointer to global position and identifier. Code does depend on Eigen vector and matrix classes. The identifier class will have to be coded, it's just an integer for the time being. Feedback welcome :-)
Merge request reports
Activity
Do you think it would be wise to introduce a namespace for these classes? Or add them to xAOD::?
One reason I ask is that there is already an ATLAS Identifier class:
This is a generic 64-bit value that relies on a helper class (there can be one for each specific type of identifier) to interpret it. Since it looks like you want yours to hold a strip ID (is that right?) maybe the ATLAS class can be re-used?
I guess the Id field is just a unique index for the object - do we need that? Things might live in containers and be mainly accessed by ElementLinks.
Not trying to be pedantic, but just want to understand...
Hi @dcasper,
indeed, we were discussing the possibility to have all our objects in the xAOD namespace. I've also discussed this with some experts who designed the xAOD format a few years back. They agreed with me that for a much simpler experiment like FASER, who is starting their software from scratch, using xAOD everywhere already during the offline reconstruction is advantageous. We will adapt our classes accordingly. This has the advantage that we can also ROOT browse all potential file formats we will have (likely only one). Using the ATLAS identified seems reasonable on a second look. It doesn't have any dependencies we don't want. I think we first thought it to be too complicated, but looking now most methods are just simple logical definitions.
Best, Roland
mentioned in commit a745ed07
