Using multiprocessing Pool with PyQt5 signals

Asked Nov 19 '21 at 15:47

Active Nov 19 '21 at 16:06

Viewed 30 times

Using multiprocessing Pool with PyQt signal gives pickle error, any way I can send my progress in the loop back to the GUI?

class Analysis(QObject):

    progress = pyqtSignal(str)
    
    def __init__(self, data):
        self.data = data
        self.result = []
        
    def worker_process(self, devices):
        for n1, n2 in itertools.product(self.data[devices[0]], self.data[devices[1]])
            if n1 == n2:
                self.result.append((n1, n2))
            self.progress.emit(f'Running on {device[0]} of network {n1} compare with {device[1]} of network {n2}')
            
    def run(self):
        devices = ['device_a', 'device_b', 'device_c', ...]
        pool = Pool(os.cpu_count() - 1)
        return_value = pool.map(self.worker_process, itertools.combinations(devices, 2)) <-- can't pickle Analysis object

edited Nov 19 '21 at 16:06

eyllanesc

235,170
19
170
241

asked Nov 19 '21 at 15:47

ReverseEngineer

Use a multiprocessing.Queue object to communicate back with the parent process. – AKX Nov 19 '21 at 15:50
I thought about it before. But can't seem to make it work. Signal can triggers slots from parent process to update the GUI widgets, but queue can't. – ReverseEngineer Nov 19 '21 at 15:57
You would probably need to poll the queue. – AKX Nov 19 '21 at 16:40

Using multiprocessing Pool with PyQt5 signals

0 Answers0