RFE: RP2 port - machine.Pin.mode()

[DeeJay]

(I know this is not the place to raise an Enhancement Request. This is simply to explore whether such a request via the GH repo would have any chance of success. )

I have a potential monitoring application for an RPi Pico board that needs to determine the mode in which a GPIO Pin is running.

machine.Pin.mode() is not implemented in the RP2 port.

I can use a workaround by parsing the output from machine.Pin(pin_id). But a simple call to a Pin.mode() method would be cleaner.

My practical knowledge of C and its integration with MicroPython is very limited.

I think that any changes would be needed only to ports/rp2/machine_pin,c

The API would be similar to that for pin.value() at line 457

The machine_pin_print() function seems to contain model code for determining the current mode ready for printing. Lines 213-250

There is a table of instance methods at line 586 that would need updating.

I could try to work up a PR based on these ideas. But it would probably be quicker, more efficient, and less painful (time-consuming) if someone already knowledgeable did it?

Any thoughts on whether this is worth pursuing and whether I should raise it as a formal GH Issue?

Thanks for listening - DJ

[dpgeorge]

In your project are you able to store in a variable somewhere the current mode of a pin, and then query that variable when you need to know the mode?

I know that's not the best solution, but I'm wondering if it's possible to do it that way?

[DeeJay]

Thank you for your prompt response. That's essentially what I am doing, but I thought that knowing the current mode of a pin when the code started running was necessary, and in the absence of Pin.mode() I was using this -

def gpio_get_mode_name(pin):
    pReport = Pin(pin)
    sReport = str(pReport).replace(')','')
    modearg = sReport.split(",")[1]
    modename = modearg.split("=")[1]
    return modename

(That's fragile and vulnerable to unannounced changes to the format of the print method for a Pin object.)

But your sensible 'rubber duck' comment has made me wonder whether it is a great loss to assume that all Pin modes are initially undefined.

I'll pursue that first.