search

atcollab / at

Accelerator Toolbox
Apache License 2.0
48 stars 31 forks source link

bug is saving lattice #743

Open lnadolski opened 4 months ago

lnadolski commented 4 months ago

Dear @lfarv

I suspect one bug.

When saving the lattice, with at.save_lattice(ring, 'newlattice.m') Eapertures is written instead of EApertures

ring[0] is Marker('SOLEIL_II_V2366_SYM1_2023JUL24', EApertures=array([0.0075, 0.005 ]), Periodicity=1, tag='RingParam')

newlattice.m:
atringparam('SOLEIL_II_V2366_SYM1_2023JUL24', 2750000000.0, 1, 'Particle', struct('name', 'relativistic', 'rest_energy', 0.0, 'charge', -1.0), 'HarmNumber', 416, 'Beam_Current', 0.0, 'Nbunch', 1, 'M_File', 'mylattice.m', 'Eapertures', [0.0075 0.005]);...

Best regards,

Laurent.

oscarxblanco commented 4 months ago

Dear all, I believe this was already discussed here #635 due to the special behaviour of RingParam. No modification was done at that time, but #741 might force to discuss it again.

lfarv commented 4 months ago

Corrected in #744. The case of the names of all unexpected attributes (not only EApertures) is now kept unchanged.

Note that the optional Marker in ring[0] is completely ignored when saving to .m or .mat files (because of the RingParam tag). Only the Lattice attributes are used to generate the RingParam (they contain the same information).

lfarv commented 4 months ago

I add that you should not use EAperture or other attributes in a Matlab RingParam element. It's not a real element, it's there only for the storage of lattice attributes (energy, particles...).