Add long names

[aberges-SLAC]

Description

Made changes to SignalPanel class in display module to support long_names for device signals. This approach should be pretty close to compatible with bluesky/ophyd#1187, so if & when that comes through we should only need minimal changes.

Motivation and Context

Have you ever had a screen with 30+ signals and maybe multiple subcomponent devices with identical signal names? Maybe you wanted alternative signal labels for cleaner GUI that end users can play with?

Figured some of the work I did in pcdshub/pcdsdevices#1320 could be generalized in typhos and potentially trim the fat out of my screen scripts.

Long names lets you use an optional string to replace the label_text in a signal row object added to your grid of signals. When you use a long_name, it will add the signal's dotted_name to the tooltip so you can still hover and get hutch python context.

How Has This Been Tested?

I didn't have any unit tests to run, per se, but I did test this on a variety of devices for screens we've got out in prod (and some I'm cooking in a different PR).

I was able to open the entire CRIXS MODS screen without everything crashing and burning, so that's gotta amount to something, right?

test_typhos_long_name_crixs_mods.mp4

Where Has This Been Documented?

This PR! And in the code itself

Pre-merge checklist

• Code works interactively
• Code contains descriptive docstrings, including context and API
• New/changed functions and methods are covered in the test suite where possible
• Code has been checked for threading issues (no blocking tasks in GUI thread)
• Test suite passes locally
• Test suite passes on GitHub Actions
• Ran docs/pre-release-notes.sh and created a pre-release documentation page
• Pre-release docs include context, functional descriptions, and contributors as appropriate

[ZLLentz]

I like the way this is implemented. I think the docs inclusion is required before merge and the other suggestion is optional.

[ZLLentz]

If this bit is a hacky workaround, and we needed it twice, it should be pulled out into its own function. After pulling out into its own function this is a reasonable thing to write a short unit test for to make sure we've covered all the cases we expect.

I'm wondering if there's some way to revise this to make it more readable but I don't have any immediate suggestions. Some of the hasattr calls probably aren't needed I guess.

[aberges-SLAC]

Agreed, it's probably worth some:

def _get_long_name(self, device, attr, dotted_name) -> str:
"""
Check if the signal has implemented a long_name and return it. Will check both the device and the component signal for the long_name if it exists.
"""
    long_name = None
    try:
            if hasattr(getattr(device, attr), 'long_name'):
                long_name = getattr(device, attr).long_name
    except AttributeError:
            # Then maybe we have a nested component and can't touch the signal
            if hasattr(getattr(device, dotted_name), 'long_name'):
                long_name = getattr(device, dotted_name).long_name
    return long_name

I think I ended up needing both hasattr due to it getting squirrelly when you build a screen with multiple component devices where the long name is actually in device.component.attr.long_name instead of device.attr.long_name — and the fact that long_name is not in the Ophyd object by default yet. Once they implement it we can probably just replace it with some try block around if getattr(device, attr).long_name instead.

[ZLLentz]

Missing here is the pre-release notes docs step, same as how we've done it in pcdsdevices with pre-release-notes.sh

[aberges-SLAC]

Ah derp forgot to commit that, let me fix that after above changes

[ZLLentz]

Small note: the pre-commit CI job is flagging ex as an unused variable and a couple instances of trailing whitespace

[ZLLentz]

Super petty spacing nitpick to match the other docstrings

Suggested change

	device: (any)
	The Ophyd.Device with component signals
	attr: (str)
	The str name of the signal attribute
	dotted_name: (str)
	The full dotted name to the signal
	Returns
	--------
	str or None
	device: (any)
	The Ophyd.Device with component signals
	attr: (str)
	The str name of the signal attribute
	dotted_name: (str)
	The full dotted name to the signal
	Returns
	--------
	str or None

[tangkong]

I don't think this ever got committed 👀

[ZLLentz]

whoops

[aberges-SLAC]

I don't even remember getting this message on the commit o.O

[ZLLentz]

More pettiness: I normally wouldn't ask you to annotate, but this one is half annotated. Can you either finish the annotations or remove them?
device: typing.Any
attr: str
dotted_name: str
-> str | None

[aberges-SLAC]

I'll remove the annotations to be more consistent with the rest of the script, although it probably wouldn't hurt to add the annotations to all of the funcs one day

[ZLLentz]

I've only got nitpicks left
If you're happy and finished we're good to go

I still like the implementations here and the result

[aberges-SLAC]

I've only got nitpicks left If you're happy and finished we're good to go

I still like the implementations here and the result

Should pass the checks now, gave it the ole flake8 rundown. Passing black over it changed quite a few things so I'm gonna pass on that to save us a lot of frantic debugging

[ZLLentz]

Looks good to me, let's roll
Remaining CI is an unrelated weird python 3.11 race condition that I don't need to think about today