M-A
Ignore the test failure, I need to redo the way I do forward-testing.
Frankly I'd simply use Mode1 and see if anyone complains on the next release. Users can always revert back to the previous release if it breaks them and we will...
Sorry for the delay, lots of things happening here. Thanks for the data point about some LEDs not working with mode3. I'd still prefer to use a bool defaulting to...
Revisiting this PR: I'd really prefer to use a bool instead of a spi.Mode; this way the default bool value in Go (false) is the right one.
Apologies for the delay. Thanks I'll try to look at it ASAP.
What I would recommend is to start with only one subsystem, let's say i2c, and go from there. The PR is fairly large and will make the back and forth...
I generally favor composable components. I see that the current hd44780 New() function takes 6 pins as input; https://pkg.go.dev/periph.io/x/devices/v3/hd44780#New If you were to use a IO expander, I would expect...
What change in conn/i2c did you have in mind specifically? I don't see such need.
Ah you mean parallel gpios. That's something I wanted to do for a long time but I think it belongs to conn/gpio. This is not I²C specific. For example one...
I'm fine with a specific new method on the Dev struct. I wished I had to time to make a pure mechanism but still hasn't done that.