PyQt QThreadPool error

Phil Thompson phil at riverbankcomputing.com
Fri Jan 8 12:27:38 CET 2010


On Thu, 7 Jan 2010 15:07:10 -0800 (PST), h0uk <vardan.pogosyan at gmail.com>
wrote:

...

> Phil you right about app.exec_(). But situation is sligthly different.
> 
> I want to have more than one Job. I add these Jobs into QThreadPool
> trough cycle. And I also want these Jobs to run  sequentially.
> 
> The following code illustrate what I mean:
> 
> #!/usr/bin/env python
> # -*- coding: utf-8 -*-
> 
> import sys
> import os
> import time
> 
> from PyQt4 import QtCore, QtGui
> 
> class Job(QtCore.QRunnable):
> 	def __init__(self, name):
> 		QtCore.QRunnable.__init__(self)
> 		self._name = name
> 
> 	def run(self):
> 		time.sleep(3)
> 		print self._name
> 
> 
> if __name__ == "__main__":
> 
> 	app = QtGui.QApplication(sys.argv)
> 
> 	QtCore.QThreadPool.globalInstance().setMaxThreadCount(1)
> 
> 	for i in range(5):
> 		j = Job("Job-" + str(i))
> 		j.setAutoDelete(True)
> 		QtCore.QThreadPool.globalInstance().start(j, i)
> 	app.exec_()
> 
> After 5 cycle I get the same error:  An unhandled win32 exception
> occured in python.exe.
> 
> How I can do it?? To run my Jobs sequentially???

It's a PyQt bug. The workaround is not to use setAutoDelete() and instead
keep an explicit reference to your Job instances - for example in a list of
jobs.

Phil



More information about the Python-list mailing list