use multiprocessing in flexx after freeze, The window keeps popping up about flexx

[almarklein]

I'm sorry but this is not enough information to provide any help. Is it possible to create example code, that is as small as possible, but that I can run and that demonstrates the problem?

[1751660300]

ok, this is a complete example, you need
You need to package an application through the freeze method, then run, click this button.

import time
from multiprocessing import Pool
from flexx import flx

def down_single(index):
    print(f"{index}")
    time.sleep(index)

def down_multi():
    executor = Pool(2)
    for i in range(4):
        executor.apply_async(down_single, args=(i, ))
    executor.close()
    executor.join()


class Example(flx.PyWidget):

    def init(self):
        with flx.VBox():
            self.lable = flx.MultiLineEdit()
            self.button = flx.Button(text="test")
            flx.HBox(flex=1)

    @flx.reaction("button.pointer_click")
    def down_btn(self, *events):
        self.down_ac()

    @flx.action
    def down_ac(self):
        down_multi()
        self.lable.set_text("success!")

[1751660300]

After running, multiple windows will be launched and cannot be closed, as if in a computer virus, hahaha

[almarklein]

Did you put the code to launch the app in an if __name__ == "__main__"?

if __name__ == '__main__':
    m = flx.App(Example).launch('firefox')
    flx.run()

[1751660300]

Here's the code I packed

def test_make_package_3(self):
flx.App(Example).freeze("./", excludes=[], launch="app")

[almarklein]

Oh, I have no idea how multiprocessing works together with frozen apps. But I do know that when you "launch the pool", it spawns new processes using the same __main__ module, so if you don't guard with if __name__ == "__main__" it will re-launch the app again, and again ...

[1751660300]

yeah，i run with if name == "main"， the program running ok，hope this problem can be solved later. thank you

[1751660300]

hello,
Pyinstaller packages programs that use multiprocessing must call the multiprocessing.freeze_support () method to freeze the process. I try to write the generated examp.py in the following format to solve this problem.

pyinstaller.exe "E:\code\flexx\learning\example.spec"