[Tutor] multiprocessing for nested function

[Emile van Sebille]

On 5/17/2011 7:29 AM tax botsis said...
> Part of my code includes the following script, which both shows a
> progress bar (for as long as 'categorize_reports()' is executed) and
> prints the fileid and the output of the 'document' function; the latter
> is the one increasing the calculation time (please see my comment). Any
> ideas on how to use all the cores of my CPU here?

A python instance lives it life on a single processor.  Accessing 
multiple cores requires you start multiple python instances.  You'd need 
to restructure to provide some form of cooperative independent 
processing.  Pyro, stackless, and twisted come to mind.  It looks like
the info at http://wiki.python.org/moin/ParallelProcessing is more up to 
date than my memory...

Emile



>
> ...
> def categorize_reports():
>      if __name__ == '__main__':
>          root2 = Tkinter.Tk(className=' Processing Progress')
>          m = Meter(root2, relief='ridge', bd=3)
>          m.pack(fill='x')
>          i=0.0
>          for fileid in fileids():
>              m.set(i, 'Processing will take a few minutes...')
>              i=i+1.000/len(fileids())
>              m.after(1000, lambda: _demo(m, i))
>              print str(fileid), document(fileid) #this slows down the
> process
>          root2.withdraw()
> ...
>
> Thanks
> Tax
>
>
>
> _______________________________________________
> Tutor maillist  -  Tutor at python.org
> To unsubscribe or change subscription options:
> http://mail.python.org/mailman/listinfo/tutor