"Latching" variables in function

[William Ray Wing]

On Apr 9, 2014, at 12:35 AM, Terry Reedy <tjreedy at udel.edu> wrote:

> On 4/8/2014 4:09 PM, Grawburg wrote:
>> 
>> I've probably used the wrong term - I'm thinking of what I do when writing PLC code - so I can't find how to do this in my reference books.
>> This is part of a project I'm working on with a Raspberry Pi and an MCP23017 port expander.
>> I have a N/O pushbutton that I want to "latch" a value to a variable when it's been pressed.  I have this function that gets called periodically in
>> a 'while True' statement:
>> 
>> def button():
>>    pushbutton = 0
>>   button_value = 0
>>    pushbutton=bus.read_byte_data(address,GPIOB)
>>    if pushbutton > 0:
>>         button_value = 1
>>    return button_value
>> 
>> I need button_value to become '1' when the button is pressed and to remain '1' until the entire program (only about 25 lines) ends with a sys.exit()
>> 
>> What do I use to 'latch' button_value?
> 
> It depends on whether you can set up your system so that pushing the button generates an interrupt. But I know little about R.Pi and less about the 'port expander'. If there were an interrupt, you would just have to write an interrupt handler. When possible, this is much better than polling.
> 
> -- 
> Terry Jan Reedy
> 

I think what the OP was asking for was a way in standard Python to “lock" the value of an instance variable, which of course, you really can’t do. However, what he (I assume it’s a he) could do is arrange his calling program so that after the button method returns a “1”, the method isn’t called again.  That is, test for truth of button = 1 in the calling program, and if true, skip the call.  The next time the program runs, the button value will be re-initialized to zero and everything is back to square one.

-Bill