acpi should take precedence in battery

[Snuggle]

Description

Shuffled battery section around, should fix issue #526. acpi seems to give a nicer output than upower and doesn't give a bunch of extra battery/peripheral devices that are hard to filter. If acpi returns 0% charge, the device is likely a desktop without a battery.

Changes haven't been tested yet, testing yourself would be appreciated!

Fix #526

[salmanulfarzy]

Thanks @Snuggle for PR.

It might be better if we swap order between acpi and upower, and leave pmset as it is. Like pmset → acpi → upower. Looks like you missed the variable used in conditional.

This needs testing 🛠️

[salmanulfarzy]

I haven't locally tested this, But code look good 👍

[Snuggle]

In the meantime, could I please ask for some documentation on how to actually locally run your tests? I just get some npm error about some shintango2 package or something not being installed from ./scripts/tests.sh. I can't seem to install it with npm either.

[salmanulfarzy]

about some shintango2 package or something not being installed

😆 We depend on kward/shunit2 for testing as submodule.

3. Update submodules in your fork (via git submodule update --init --recursive)

CONTRIBUTING.md#setup

After that npm test should work fine. Let us know if you've any trouble after that.

[denysdovhan]

Any progress on this? I'd like to merge this.

[Snuggle]

@denysdovhan — Just waiting for a second person to test & review, I believe.
There's nothing else to progress on, this pull request is finished & waiting.

[denysdovhan]

@Snuggle good. I'm merging this.

P.S: I'll be grateful for a PR with some tests for battery section.