test how import works with energy and polarization/angle within a single bluesky run

[EliotGann]

SST1's new acquisition scheme will allow for multiple energy scans in a single run. how these will import into pyhyper will be important to rigorously test

[pbeaucage]

do you have an example scan id?

[EliotGann]

I solved this for my Igor browser. it's slightly painful, but my logic is 1.) check if axis exists, if it exists, check if the maximum and minimum are more than 0.1 apart. if so, add that as an axis. if not, just make it a variable

[EliotGann]

48815 -> energy and polarization 48813 ->polarization 48812 -> angle and polarization (also sample x and y, but in such cases angle is the main one)

[EliotGann]

energy could be an added axis in either of those last two (it's just one value in these cases)

[cbishop4]

Working on this issue - to clarify, for scan 48815, it has been taken at several energies (typical) and also multiple (I'm assuming 2) polarizations (new)? 48813 was taken at multiple polarizations, with no other changes - i.e., was taken at a single energy? And 48812 was taken at a single energy, but with multiple angles and polarizations?

[EliotGann]

That's what I meant, yes. Energy can be an axis in most scans, then extra axes that have been added immediately are polarization and angle (which also moves X and Y). then we need to also handle Spiral scans, where X and Y are changed but there is only one energy, polarization, and angle. I can link to where I handle this in the Igor reader if that helps with the logic.

[EliotGann]

https://github.com/NSLS-II-SST/TiledRSoXS/blob/33f1d34056bbb849c0db6f0482367158585f95ce/TiledRSoXS.ipf#L849-L894 is where I do the logic of naming the images based on if each of the axes exist, and if they deviate more than 0.1.

[EliotGann]

https://github.com/NSLS-II-SST/TiledRSoXS/blob/33f1d34056bbb849c0db6f0482367158585f95ce/TiledRSoXS.ipf#L878-L893 is the specific part where I do the logic of choosing which to make axis labels - using x and y only if angle hasn't moved etc

[cbishop4]

live_analysis branch has been updated to load data from current runs. It is backwards-compatible with old scans with different metadata, at least back to 2022 beamtimes. Example notebook added in tutorial of the live_analysis branch.

[pbeaucage]

I'm going to leave this open for a critical issue: we should really have unit tests in the code of all these odd loading scenarios, even if we have to mark them skipped until public Tiled server is available.

[pbeaucage]

@cbishop4 and @EliotGann - scan 48815 is a NEXAFS scan. Is there an equivalent RSoXS scan for testing?

[pbeaucage]

This is now completed in PR #65. Seems to work well.