Trying to display LHCb event data

[andrewpap22]

Greetings,

So, as discussed on the Wednesday meeting, I've been trying to display some LHCb event data that I've gathered from C++ code of the detector itself. ( of course with the help of my supervisor @sponce )

I think that the format of the file is correct ( but of course I might be wrong since nothing is being displayed ).

here is my file (uploaded in .txt because GitHub doesn't support upload of .json in discussion) LHCb_event_data.txt

The ways I tried to display my file are the following:

• Load the file from the GUI interface directly
• Change the default LHCb file with mine in the corresponding typescript file.

The part of the code where I changed the file is the following:

/**
 * Directory Path: 
 * /phoenix/packages/phoenix-ng/projects/phoenix-app/src/assets/files/lhcb
 * So this is the path where I added my file, 
 * And then on the path: 
 * /phoenix/packages/phoenix-ng/projects/phoenix-app/src/app/sections/lhcb
 * The modified file: lhcb.component.ts
*/

// Line 60
.get("assets/files/lhcb/my_file.json")

Any help with what I'm doing wrong would be appreciated, and I'll be glad to help and contribute to the phoenix project as I progress with my knowledge and all...

Thank you in advance for your valuable help,

All the best,

Andreas

[EdwardMoyse]

Hi @andrewpap22 - firstly, you cannot load this file from the GUI directly because it's not in Phoenix format.

As an example of a file which is, have a look at :

which is this one:
https://github.com/HSF/phoenix/blob/master/packages/phoenix-ng/projects/phoenix-app/src/assets/files/event_data/ATLAS_calibration.json

We should add a LHCb loader to the GUI. #281

I'll try your second approach now.

[EdwardMoyse]

BTW we should improve reporting of errors in Phoenix, but if you open your console log in your browser you might have seen something. And if you open the info panel, you can get a hint that something is wrong because the event number and run number are undefined.

But we should definitely try to make this clearer.

[EdwardMoyse]

Incidentally, the earlier "Error: Invalid file format!" in the info panel is because the file was named .json.txt when I first tried to open it (I'm using Windows and I still struggle with the basics!)

[EdwardMoyse]

Okay, this file also seems to be a completely different json format from the file provided by @bcouturi, which is what the LHCb loader understands.

Here's the file Ben provided:
https://github.com/HSF/phoenix/blob/master/packages/phoenix-ng/projects/phoenix-app/src/assets/files/lhcb/00191749_0005296728.json

How did you generate this JSON file? Perhaps @bcouturi can help?

Otherwise, if you want to look at the ATLAS Phoenix dumper it's here:
https://gitlab.cern.ch/atlas/athena/-/tree/master/Event/DumpEventDataToJSON/src

It's obviously reading in ATLAS data, but I think it shouldn't be too hard to port it to LHCb.

[EdwardMoyse]

I'm very happy to have a chat over zoom or whatever, if this doesn't make sense!

[bcouturi]

Sorry just catching up as I was off yesterday. Andreas is redoing (cleanly) the whole chain so the idea is to abandon our custom export/import for something more standard if possible.

[andrewpap22]

Greetings,

So I managed to make the .json file proper in Phoenix format (of course with the valuable help of Dr. @sponce) and as you can see in the screenshot now we have a proper visualization for Long tracks and VPHits for the LHCb experiment as well! 😄

Now for our next steps I would like again to ask for your valuable help in trying to properly display as well the Calorimeters and Energy Deposits.

As you can see in our new created file below, (remove the .txt extention when you download the file)

tracksV2.json.txt

We have inside that file our Calorimeters (ECal and HCal) and they're not like the ATLAS ones where you have eta and phi, etc... we instead have XYZ points and of course the energy deposits as you have. We could convert the XYZ points to phi I guess but that would not make sense for the LHCb Clusters.

So a workaround would be I guess to write some code in the Phoenix library in the LHCb part of the project to understand and display properly our LHCb specific CaloClusters.

So my question is if you could "drive" me somehow on trying to start and write that particular code (I would be grateful if you could give me some directions to try and do that inside Phoenix, because it will be a great learning journey for me and obviously far than interesting) or if there is something else we can do here and of course again I want to invest my time in doing that! 😄

Thank you very much for your time and help!!!

[sponce]

And [x,y] would be [eta,phi] in the Cylinder case of course

[EdwardMoyse]

Hi @sponce - That looks good for the content of the cell. So maybe something like this?

PlanarCaloCells : [
   "HCalo": {
      "plane" : [x,y,z,d].
      "cells": [ {"pos" :[x,y], "energy":e, "otherdecoration":"whatever"}, {}, {} ]
     },
    "ECalo": {
      "plane" : [x,y,z,d].
      "cells": [ {"pos" :[x,y], "energy":e, "otherdecoration":"whatever"}, {}, {} ]
     },
]

i.e. you have the object type, then the names of the containers, with the plane coords and an attribute cells, which contains the cell objects

If they're all the same size, we could make that a property of the collection i.e.

   "HCalo": {
      "plane" : [x,y,z,d].
      "size": [0.1,0.1],
      "cells": [ {"pos" :[x,y], "energy":e, "otherdecoration":"whatever"}, {}, {} ]
     },

If they vary, then it could be a property of the cell itself.

[sponce]

Looks good to me. The size attribute at the top level could be seen as a default size which can be overwritten by a specific one in a given cell. @andrewpap22 do you feel like trying to implement that ? I obviously will help !

[9inpachi]

If LHCb is not the only experiment with PlanarCaloCell then we should definitely have it in the PhoenixLoader.

For a quick direction, the code to draw the 3D object will go to PhoenixObjects and then we can load the new object by extending loadObjectTypes to include PlanarCaloCell like we do for tracks.

[andrewpap22]

Alright then I updated my notes as well and I'll be trying to implement what was discussed here! I will keep you updated! 😃