[Bug] Enable/Disable value for the Auto Bed Level allows a decimal input about octoprint-eeprom-marlin

[cp2004]

Would you mind doing a quick test to see if it does round it when sent to the printer? It's using a generic template, I guess we need more conditional work here. Problem is, in Marlin there's no consistency and I always end up with really specific cases, when I was trying to make generic definitions....

Anyway, I will see about fixing it for 3.1.0, I think there's probably something else like that around too.

[KenLucke]

Everything's printing right now, so I can't test, but a simple change to a checkbox (☑ signifying a "1" when stored) or Radio Button set ( ʘ Enabled O Disabled ), to me, would be far simpler than trying to specific-case conditionals.

[KenLucke]

OK, fired up the spare Pi and OP. Anything ≥ 1 when stored, displays as "1" when reloaded [although there's no telling what value is in the background). Anything < 1, when stored, displays as "0" - again, no telling whether it's floating point or integer in the background.

[cp2004]

Thank you - that confirms that it's not really a critical bug that breaks everything when decimals 😄.

I agree, a checkbox would be easier on the user - but it makes it harder to send to the firmware, and further away from the command itself for those used to them. I will likely just lock it to 1/0 if possible, unless I can think of a smart way to do the checkbox.

[KenLucke]

I think that the average person who uses this kind of thing doesn't necessarily care about being "further away from the command itself" - they are just looking for functionality. Also, the average non-technical/coder user doesn't really inherently know that 1==enabled/0==disabled in most things.

I'm also having a hard time figuring out why it would be so hard to use a boolean (checkbox/radio button enabled) to represent an already boolean value (bed level enabled/disabled). I admit it's been a long time since I coded, but a simple if/then test should do it.

[cp2004]

Yeah it is not difficult per se, it just requires extra steps in what is currently a very streamlined process, to convert something: true into S1 or something like that, and it requires this on all the ends where it comes in and out of the UI.

[cp2004]

OK following a nearly two month break from this plugin... I've done this one.

The latest commits in this repository have switched the entire codebase over to a single source of truth, that makes maintainability easier with definitions for all the commands. This means I have been able to support Boolean values through this quite easily, so they are now checkboxes. So it's ready for 3.1.0 🙂

[cp2004]

There's a release candidate available for testing this now: 3.1.0rc1. Feedback ticket #45 for any comments on the release. Stable release approx. 1 week depending on how the testing goes.

[cp2004]

3.1.0 has been released