Improve time settings

[yumouwei]

Implemented changes

• GLT: introduced _postprocess for deduplication and unit conversion,
• EfitTimeSetting
  • Add d3d_times().
  • Create a generic get_efit_times method.
• IpTimeSetting
  • Add cmod_times and east_times
  • Fix unit conversion in d3d_times
  • Add use_efit_time_range (default to False) option to remove time points outside the range of the EFIT timebase.
• SignalTimeSetting
  • Add d3d wrapper function to allow using e.g. tree_name="ptdata", signal_path="ip"
• Other changes
  • Fix all np.where output syntax Fix get_end_of_shot method in time_setting #443
  • Figure out how to get the dimension from MDSplus to do unit conversion without hardcoding it based on the device.
  • Set up test cases in test_retrieval_settings.py.

Figures

[yumouwei]

Misclicked Merge instead of Rebase

[yumouwei]

params.mds_conn.get(f"units_of(dim_of({self.signal_path})), self.tree_name).data()

returns the unit of the signal's timebase. However, this does not work for any PTDATA signal; e.g.

params.mds_conn.get("units_of(dim_of(ptdata("ip", 161228))).data()

returns a whitespace character (' ').

There is a PTDATA expression (ptunits("ip", 161228)) for getting the unit of a signal, but I can't figure out how to wrap it to get the unit of that signal's timebase.

Update

Upon checking on CMOD, I found that params.mds_conn.get(f"units_of(dim_of(\efit_aeqdsk:ali)), "_efit_tree").data() also returns a whitespace, even though both "units_of(\efit_aeqdsk:atime)" and "units_of(\efit_aeqdsk:time)" give the correct unit of second. I need to take a look at this again, but if this doesn't work then we should give up on the idea of creating a machine independent EFIT timebase method.

[gtrevisan]

cool! we should be able to talk in person and iron out the last details.

[gtrevisan]

okay, SharedTimeSetting should be much better! LMK.

[yumouwei]

test_time_setting.py passed on EAST

[gtrevisan]

to do:

• refactor all time parameters into machine configs,
• refactor common lines into new common functions,
• ...