SIRIUS Models in different simulation tools

[VitorSouzaLNLS]

No description provided.

[fernandohds564]

no parenthesis in the file names. Add only the files that will be on the on the final PR, try to avoid pushing intermediate files to the main repository.

[murilobalves]

Maybe put the information that it is the version SI_V25_04 in the filename? something like sirius_si_v25_04.seq

[fernandohds564]

Maybe put the information that it is the version SI_V25_04 in the filename? something like sirius_si_v25_04.seq

@murilobalves, @VitorSouzaLNLS, I don't think this is a good way of keeping this information. It is not easy to visualize the changes from one version to another. Maybe we could keep this information in comments inside the file.

[VitorSouzaLNLS]

Maybe put the information that it is the version SI_V25_04 in the filename? something like sirius_si_v25_04.seq

@murilobalves, @VitorSouzaLNLS, I don't think this is a good way of keeping this information. It is not easy to visualize the changes from one version to another. Maybe we could keep this information in comments inside the file.

@fernandohds564 What names do you suggest for the MAD-X sequence and the (python) constructor files?

[fernandohds564]

I don't know, I don't have a strong opinion about it...
Maybe sirius_madx.seq?

[fernandohds564]

@VitorSouzaLNLS, what is missing here for this PR be ready for review?

[VitorSouzaLNLS]

@fernandohds564 I'll open the PR-review. But, as we discussed before the SI-MAD-X model is not "automatically" updated when the SI-pymodels model is updated.

[fernandohds564]

@VitorSouzaLNLS , I don't understand why the file madx_si_constructor.py is needed if we also have the file sirius.seq. It seems as difficult to change one as it is to change another. Is it there, as a process of a more automated script in the future?

[fernandohds564]

why not taking this segmented model from the segmented model file?

[fernandohds564]

same here.

[VitorSouzaLNLS]

@fernandohds564

The .seq file is the one actually used in MAD-X, while the .py file serves as a constructor. I wrote the constructor to be able to track the changes in pymodels and apply mods more easily.

It's certainly possible to write a more automated script to convert the pymodels SI model into a MAD-X model. However, I opted for a quicker approach since I was asked by Murilo and people from ESRF for studies in September (if I'm not mistaken). As result, I didn't focus on deep optimizations.

About the usage (non-usage) of the pymodels.SI_V5_04.segmented_models: at first, it was easier to understand the conversion of BC, B1, B2 using the segmented models in-file. However, I believe it would now be "simple" to rewrite the conversions pulling the dipole pieces directly from pymodels.

[fernandohds564]

I think we can delay this discussion until I come back, in April. Is it ok for you @VitorSouzaLNLS ?