Configurability for material plots

[Victor-Schwan]

BEGINRELEASENOTES

• formatting and refactorization
• material_plots_2D.py:
  • angleDef is a choice parameter + thetaRad choice added
  • outputDir option added
• material_scan_2d.py:
  • argParser added (compactFile,outputFileBase, angleDef)
  • outputDir option added

ENDRELEASENOTES

[tmadlener]

these are the changes with the formatting factored out

[andresailer]

these are the changes with the formatting factored out

Now can we have those changes on top of the original style 🤪

[tmadlener]

I am afraid that is not something I can do in 5 minutes ;)

Should we format all python sources and make a PR that we can then ignore in blames?

[andresailer]

Should we format all python sources and make a PR that we can then ignore in blames?

But something that reduces the amount of changes. Because #396 is going to be painful enough to merge.

[tmadlener]

Rebased this after #398 has been merged. Please check again that I didn't mess up in the process.

--inputFile is now available as well as the original --fname which now simply has inputFile as destination.

[tmadlener]

Thanks for checking.

[tmadlener]

This technically changes the default geometry from IDEA_o1_v02.xml to an ILD geometry. However, it also looks like the previous version didn't work without editing the file?

Technically this should be a positional (required) argument, since the script can't do anything useful without it. However, at this point --compactFile is also pretty much a convention for several scripts.

For me these changes are OK.

[Victor-Schwan]

I don't care which detector is the default as long as the path is absolute or relative to $K4GEO or $k4geo_DIR. Originally the path was relative and Daniel and I had trouble getting it to work. I think this should work out of the box without assuming a current working dir

[andresailer]

Originally which path was relative?

[tmadlener]

The path to the IDEA_o1_v02.xml compact file.

edit: I.e. the script only work if called in the directory where that file was, IIUC.

[andresailer]

Ahh, ok, the detector file was just hard coded. I guess the compactFile will accept relative or absolute paths.

[tmadlener]

Yes, exactly. And yes, absolute and relative paths will work. It is directly passed to the GeoSvc.detectors.

[tmadlener]

I am OK with this. Anything else from your side @andresailer?

[armin-ilg]

Thank you for the work @Victor-Schwan, it works fine for me! I think it would be nice to apply the same improvements to the 1D mat. budget scripts (in the same place) and update the scripts/check_material_budget.sh script accordingly. What do you think @andresailer @tmadlener ?

[tmadlener]

In the long run, I think harmonizing all of these scripts would be a good idea. Whether we do it in this PR or in follow up PRs depends a bit on the availability of @Victor-Schwan. I the bash script does not work any longer at the moment, we should fix it. @armin-ilg do you have some estimates how long this runs, and whether we could put this into a CI workflow (maybe in a slightly slimmed down version)?

[Victor-Schwan]

I agree that it is a good idea to harmonize as much as possible. I'd prefer to do the harmonization in another PR. In principle, one could also turn the scripts in the scripts folder into python scripts to have the same --compactFile interface (keeping a bash script calling them not to change the ui). Is the linked script affected by the changes? The main change is the different default detector but the script forwards a compact file anyway...

[armin-ilg]

In the long run, I think harmonizing all of these scripts would be a good idea. Whether we do it in this PR or in follow up PRs depends a bit on the availability of @Victor-Schwan. I the bash script does not work any longer at the moment, we should fix it. @armin-ilg do you have some estimates how long this runs, and whether we could put this into a CI workflow (maybe in a slightly slimmed down version)?

On lxplus the plots take O(5) minutes (running all angle defs), so that could definitely go into a CI workflow