Scripting Vertical Slice project - i05 #429

Description

Our vertical slice project has asked us to priorities beamlines i10 and i05 for the new Diamond II software. I10 devices are more or less completed already. However, the i05 side we still need to make progress with porting device logic over.

Scripts

We have been supplied some GDA scripts from the i05 beamline. We need to be able to replicate the current behaviour of scripting. These are listed below:
co-alligment_M1_vs_M3.py
M1_roll_vs_PGM_cff_B800.py
gas_cell_cff_scans_1600.py
gas_cell_cff_scans_800.py
M1_roll_vs_PGM_cff_1600.py
exit_slit_He_res_scan2.py

Devices

List of devices that are required for these scripts (left side is GDA, right side is bluesky equivalent if already done):

• m1_pitch -> m1.pitch
• m1_roll -> m1.roll
• m3mj6_x -> m3mj6.x
• m3mj6_pitch -> m3mj6.pitch
• pgm_energy -> pgm.energy
• pgm_cff -> pgm.cff
• ring_current -> synchrotron.current
• energy -> energy
• d3_cy

The scripts all use standard move and scan plans. The only blocker I can foresee is that the GDA scan syntax allows you to specify to move a device to position x in the same scan start stop step. e.g

GDA scan a 1 5 1 b c 1 -> scan([b], a, 1, 5, 1, ...) # No equivalent for c 1 move in same scan.

We will need to either fix this directly in bluesky which will require collaboration and will be slow. Or we wrap the scan plans and allow for additional move in same plan. We should also consult other teams in DLS to know if we want to continue to support this behaviour and if so, what the best solution is.

Other blockers including getting the scan plan syntax in a usable state. This requires DiamondLightSource/blueapi#1662 for simple scan syntax and DiamondLightSource/blueapi#1671 for users to know what child devices exist.