Disable bad performers

[joesilber]

main features

• option for additional output column to control batch classification of good vs bad positioners
• additional note column to explain rationale and tolerance used (these will get posted to posmovedb as well)

side stuff

• some code cleanup
• use of nominal offset xy positions rather than complicated calcs on metrology
• fit_posparams now keeps hold of DEVICE_LOC, PETAL_ID, PETAL_LOC in its output file

example sequence

get_posmoves --host beyonce.lbl.gov --port 5432 --password reader --petal-ids 0 --exposure-ids 3506 --recent-rehome-exposure-ids 3506 -o . -c --pos-ids movers
fit_posparams -i M?????.csv -o .
merge_posparams -i *_fitparams.csv -o merged.csv
prepare_posparams_for_instr -i merged.csv -e

[julienguy]

It's not very convenient to use glob to get the list of input files instead of the file system. At nersc, the current implementation requires we run

merge_posparams -i '*_paramfits.csv' -o merged.csv

instead of

merge_posparams -i *_paramfits.csv -o merged.csv

It could be fixed by removing the call to glob and simply directly getting a list of filenames as an argument if one uses the extra nargs="*" in the descriptor of the argument as it was in the previous version. But I guess you changed that to run on windows, so I let you see if there is another solution to find. I'm leaving the PR open for now.

Except for this detail it works fine. Maybe add a sentence in prepare_posparams_for_instr --help explaining what check_static_fit_bounds , check_xy_offsets, check_arm_lengths do. Are these tests against default values and metrology?

[joesilber]

Suggestion to modify the handling of multiple files in the arguments to the get_posmoves script (see comment).

I put in a check for Windows, in which case use glob. Otherwsie revert to nargs='*' method.