Ticket2173 HRPD Sample Changer

[DominicOram]

Description of work

Added support for the HRPD sample changer, including IOC, OPI and emulator as per ISISComputingGroup/IBEX#2173.

EPICS-Top: https://github.com/ISISComputingGroup/EPICS/pull/80
Emulator: ISISComputingGroup/EPICS-DeviceEmulator#12
Support: https://github.com/ISISComputingGroup/EPICS-rotating_sample_changer
GUI: ISISComputingGroup/ibex_gui#480

To test

The basics:

• Start the IOC in devsim
• Start the emulator on the port the IOC is looking on
• Confirm that the IOC connects w/o error, all PVs appear to connect
• Create a device screen for the sample changer
• Confirm that the screen is all connected ok and that you can initialise then move the device

Testing against the VI:

• Connect the HRPD sample changer VI to the emulator (https://github.com/ISISComputingGroup/ibex_developers_manual/wiki/Emulating-Devices#connecting-an-emulator-to-labview)
• Confirm that the VI behaves in the same way as the OPI+IOC

Some slight niggles:

• The way the emulator handles errors is a bit dodgy, with no access to the device I just put in as much as I needed to get going
• The VI expects something slightly different from the documented API, I went with what the VI expects
• Some functions of the sample changer are HRPD specific, as default these aren't loaded but they are if IS_HRPD=''. I didn't expose this macro to the GUI as it's not something the user will need to change

---

Code Review

• Is the code of an acceptable quality?
• Are the PVs named according to the naming standards.
• Have suitable disable records been added?
• Have suitable simulation records been added?
• Have the finishing touches been applied?
• Is there an emulator for the device?
• Have the changes been documented in the release notes. If so, do they describe the changes appropriately?
• If an OPI has been modified, does it conform to the style guidelines? There is a script called check_opi_format.py to help with this.

Functional Tests

• Do changes function as described? Add comments below that describe the tests performed.
• Does the IOC respond correctly both in full and simulation mode, where it's possible to test both?
• If there are multiple _0n IOCs, do they run correctly?

Final steps

• Reviewer has updated the submodule in the main EPICS repo? See Reviewing work for the subModules of EPICS in the Git workflow page for details.
• Reviewer has moved the release notes entry for this ticket in the "Changes merged into master" section

[AdrianPotter]

• ROTSC needs listing in main IOC makefile
• Entry needed in release notes

[AdrianPotter]

If we get the number of the free port, we should use it

[AdrianPotter]

Use a macro instead of repeating "L0", as in other IOCs (though I expect not all of them)

[AdrianPotter]

Standard pattern is (Default: number) in brackets rather than defaults to...

[AdrianPotter]

I get an assortment of connected/disconnected states:

[DominicOram]

Did you get any associated IOC log errors? Did you try restarting the OPI a couple of times?

[AdrianPotter]

Looking at the IOC log, I think the issue might have been that devsim wasn't set correctly. I think there might be an issue in the GUI as I'm certain Devsim was set for the IOC. I can retest that. Please fix the other highlighted issues for now and I'll retest.

[AdrianPotter]

Devsim works fine, thanks. Happy with the changes. Please could you just do a couple more things:

• The (IS_HRPD=##) is ok, but following our discussion might be better suited as a macro. Otherwise it's liable to be forgotten during setup. Something generic like an integer "TYPE" would be best so it's extensible to other specific configurations (e.g. 0: Default, 1: HRPD: 2:...)
• I get a lot of "Illegal choice PV" warnings when the IOC is started up in RECSIM mode. Please could you have a look at clearing them.