Correct way to call Pythonista script from within a Shortcuts workflow?

cvp

@bosco Thanks for him, and for me, so I don't have to test anymore.
Do you know why the handler does not have a 2nd parameter (error) like described in Apple doc?

JonB

@cvp, I think this one was correct:

@Block
def handler(altitudeData:ObjCInstance, err:ObjCInstance) -> None:
   print(altitudeData)

But then pass handler directly to ObjC -- don't call ObjCBlock on it. ObjCBlock makes an Objc block callable by python, which isn't needed here.

Or, to make no other changes:

@Block
def handler_block(altitudeData:ObjCInstance, err:ObjCInstance) -> None:
   print(altitudeData)

cvp

@mikeno With the barometer module of @bosco, where print lines are commented, this script works and continues to log the pressure even if I close the iPad cover. Not tested during a day.

import background as bg
import barometer

with bg.BackgroundTask() as b:
  while True:
    result = barometer.get_pressure()
    l= f"{b.execution_time()}:{result}\n"
    with open("/private/var/mobile/Library/Mobile Documents/iCloud~is~workflow~my~workflows/Documents/bg.txt", mode='at') as fil:
      fil.write(l)
    #print(b.execution_time(), result)
    b.wait(5)

mikeno

Hi everybody, thx for helping, I’ll buy Pyto and try.

cvp

@JonB said

But then pass handler directly to ObjC

how do I do that?

In the bosco solution, there is no @Block line

Édit: ok, understood, use the handler_block directly in

    altimeter.startRelativeAltitudeUpdatesToQueue_withHandler_(main_q, handler)

Thanks for your explanations

Edit2: but I need

@Block
def handler(altitudeData:ObjCInstance, err:ObjCInstance) -> None:
   print(ObjCInstance(altitudeData).pressure)

bosco

I dropped the 2nd parameter (error) because it caused an exception: "item 2 in argtypes has no from_param method", so I tried running without the error parameter.

After reading the last comment by @JonB I now understand the proper use of @Block witch can be called directly.

This works for me.

# coding: utf-8
from rubicon.objc import Block, ObjCClass, ObjCInstance, py_from_ns
from rubicon.objc.runtime import objc_id

pressure = None

@Block
def handler(altitudeData:ObjCInstance, err:ObjCInstance) -> None:
    nspressure = ObjCInstance(altitudeData).pressure
    global pressure
    pressure = py_from_ns(nspressure)

"""
def bhandler(_data) -> None:
    nspressure = ObjCInstance(_data).pressure
    global pressure
    pressure = py_from_ns(nspressure)

handler_block = Block(bhandler, None, (objc_id))
"""

def get_pressure():
    CMAltimeter = ObjCClass('CMAltimeter')
    NSOperationQueue = ObjCClass('NSOperationQueue')
    if not CMAltimeter.isRelativeAltitudeAvailable():
        print('This device has no barometer.')
        return
    altimeter = CMAltimeter.new()
    main_q = NSOperationQueue.mainQueue
    #altimeter.startRelativeAltitudeUpdatesToQueue_withHandler_(main_q, handler_block)
    altimeter.startRelativeAltitudeUpdatesToQueue_withHandler_(main_q, handler)
    print('Started altitude updates.')
    try:
        while pressure is None:
            pass
    finally:
        altimeter.stopRelativeAltitudeUpdates()
        print('Updates stopped.')
        return pressure

if __name__ == '__main__':
    result = get_pressure()
    print(result)
    del pressure

cvp

@bosco said

After reading the last comment by @JonB I now understand the proper use of @Block witch can be called directly.

Yes, this is what I had also found and explained in my last post

bosco

@cvp Correct. After my morning brain fog cleared, that is what you said in your last edit. :-)

cvp

@bosco this, you bought Pyto and already tried...

bosco

@cvp I bought Pyto in 2019. I am currently running Pyto 17.1.1 I have tested barometer.py on iPhone 12 mini and iPad mini 4th gen.

cvp

@bosco I'm sincerely sorry, my "thus, you bought Pyto and already tried..." was erroneously for @mikeno. I guess that you, @bosco, you know Pyto because you have been part of the solution with @jonB.

bosco

@cvp No problem. I thought maybe your question was intended for @mikeno.

emildepp

@cvp said in Correct way to call Pythonista script from within a Shortcuts workflow?:

TypeError: unhashable type: 'list'

The "TypeError: unhashable type: 'list'" error message in Python occurs when you try to use a mutable object (such as a list) as a key in a dictionary. Since dictionaries use keys to index values, keys must be hashable objects, meaning they must be immutable (i.e. their value cannot be changed). Lists are mutable, so they cannot be used as dictionary keys. To fix the error, use an immutable object like a tuple, string, or number as the dictionary key instead.

To resolve the TypeError: unhashable type: 'list' error, you need to use an immutable object as a key in a dictionary instead of a mutable one like a list. For example, you can use tuples, strings, or numbers as keys, which are all hashable objects.

d = {[1, 2, 3]: "list_key"} //Using a list as a key in a dictionary

//This will raise the TypeError: unhashable type: 'list' error

d = {(1, 2, 3): "tuple_key"} //Using a tuple as a key in a dictionary

//This will work fine as tuples are hashable