Replaced the multithreaded wrapper demo with one that calls Window.perform_long_operation

This commit is contained in:
PySimpleGUI 2021-07-23 09:51:38 -04:00
parent b1081f747d
commit fca99bac39
1 changed files with 25 additions and 43 deletions

View File

@ -1,48 +1,30 @@
import threading
import time
import PySimpleGUI as sg
"""
Performing function calls that are lengthy using a "wrapper function)
This is one way to use threads and PySimpleGUI's write_event_value to enable users
to call functions that would normally cause a GUI to appear hung.
A helper function, perform_long_operation is used where a function call is normally done in your code.
Instead of directly calling the function, a lambda is used and passed to the helper function along with
the window that will receive the results and a key that is used to signal that the function
has completed running.
The overall idea here is you can directly call your function until it gets too lengthy for
the GUI to remain reasonably responsive. When you get to the point the function takes too long, then
you'll copy and paste your function call into the call to perform_long_operation. The function looks
identical to your original code, except that it's now a parameter to another function and the
return value is passed back to you later.
Demo Long Operations
How to make calls to your functions that take a very long time to complete.
One of the classic GUI problems is when a function takes a long time to complete.
Normally these functions cause a GUI to appear to the operating system to have
hung and you'll see a message asking if you want to kill your program.
PySimpleGUI has a Window method - perform_long_operation that can help in these situations
NOTE - because this method uses threads, it's important you do not make any PySimpleGUI calls
from your long function. Also, some things simply cannot be safely run as a thread. Just understand
that this function perform_long_operation utilizes threads.
window.perform_long_operation takes 2 parameters:
* A lambda expression that represents your function call
* A key that is returned when you function completes
When you function completes, you will receive an event when calling window.read() that
matches the key provided.
Copyright 2021 PySimpleGUI
"""
'''
M""""""""M dP dP
Mmmm mmmM 88 88
MMMM MMMM 88d888b. 88d888b. .d8888b. .d8888b. .d888b88
MMMM MMMM 88' `88 88' `88 88ooood8 88' `88 88' `88
MMMM MMMM 88 88 88 88. ... 88. .88 88. .88
MMMM MMMM dP dP dP `88888P' `88888P8 `88888P8
MMMMMMMMMM
'''
def perform_long_operation(func, window, end_key):
thread = threading.Thread(target=long_func_thread, args=(window, end_key, func))
thread.start()
def long_func_thread(window: sg.Window, end_key, original_func):
return_value = original_func()
window.write_event_value(end_key, return_value)
'''
M""MMMMM""M
@ -114,14 +96,14 @@ def main():
if values['-IN-'].isnumeric():
# This is where the magic happens. Add your function call as a lambda
perform_long_operation(
lambda : my_long_func(int(values['-IN-']), a=10),
window, '-END KEY-')
window.perform_long_operation(lambda :
my_long_func(int(values['-IN-']), a=10),
'-END KEY-')
else:
window['-STATUS-'].update('Try again... how about an int?')
elif event == '-END KEY-':
window['-STATUS-'].update(f'Completed. Returned: {values[event]}')
return_value = values[event]
window['-STATUS-'].update(f'Completed. Returned: {return_value}')
window.close()
@ -173,5 +155,5 @@ def old_main():
if __name__ == '__main__':
old_main()
# old_main()
main()