Remove separate longpress timeout for shift (caps lock)

[devycarol]

This patch removes the separate longpress duration attribute for the shift key altogether. I believe it's not useful, as I opine in the issue below. The only thing I'm side-eyeing is that "config_accessibility_long_press_key_timeout". I don't know the circumstances in which that value (3000) is used, but it's the only possible case where that hard-coded shift-lock duration is less than the general one.

Maybe there should be a special accessibility case where the shift-lock duration is set to 1200? I don't know.

Resolves #923

[devycarol]

Upon further inspection, that 3000 might not have even been overruled by the 1200. But I haven't tested this.

[Helium314]

Making the time related to longpressTimeout definitely makes sense. I'm not sure about the no-multiplier-needed though.
Did you test / compare in normal keyboard use? I can imagine a multiplier of 1.5 or 2 might be a good idea (at least on a very short test my 180 ms longpressTimeout felt a little too short)

Upon further inspection, that 3000 might not have even been overruled by the 1200. But I haven't tested this.

You can just add a log entry in an appropriate plance, or use debugging. It's currently using 1200.

[devycarol]

Do you know how to enable that "accessibility mode?" I don't.

[Helium314]

Ah, that weird thing. I had tried, assuming it's somehow related to Android accessibility settings (because the AccessiblityManager system service is used), and failed.

[devycarol]

Yeah, odd that you'd have a 3-second touch-hold delay for anything. That could make accidental dismissals pretty frustrating (but again, I don't know how it works.)

Would you like me to put a 1.5x multiplier for the shift key timeout?

[Helium314]

Would you like me to put a 1.5x multiplier for the shift key timeout?

I would somewhat prefer it, but I'll leave it to you and will merge if you prefer to keep it without multiplier.
If the new timeout is too short, I guess someone will open an issue anyway.

[devycarol]

Mind if I also apply that multiplier to the symbols long-press for the numpad? I find myself agreeing that it's warranted for the shift key, and I think it'd be good for that one as well.

[devycarol]

If yes, ready for merge.

[Helium314]

Mind if I also apply that multiplier to the symbols long-press for the numpad? I find myself agreeing that it's warranted for the shift key, and I think it'd be good for that one as well.

Agree, this makes sense.