🌐 Mailman 3 January 2009 - Python-list - python.org

Re: greenlets and how they can be used
by Aaron Brady Jan. 1, 2009

Jan. 1, 2009

On Dec 30, 9:40 pm, "James Mills" <prolo...(a)shortcircuit.net.au> wrote: > Hey all, > > The "greenlet" fromhttp://codespeak.net/py/dist/greenlet.html > is a rather interesting way of handling flow of control. > > I can't seem to find anything else on the subject > except for the above link and the most recent version > 0.2 and it's tests. > > What can "greenlet"'s be used for ? What use-cases > have you guys used them for (if any) ? > > Can they be used in place of threads with much > the same effect - but more lightweight ? I don't know Erlang, but it seems to be useful for a construction of a unit of work that is more consistent with the underlying goal. I had a dream for a while that in a GUI framework, every event would spawn a unique thread. The GUI would remain responsive even while executing minor tasks. Of course, shaving a second off running time isn't exactly mission-critical to GUI users.

2 1

multiprocessing BaseManager doesn't clean up net connections?
by Chris Brooks Jan. 1, 2009

Jan. 1, 2009

Hi, I'm trying to use remote managers in the multiprocessing module to listen for some events synchronously while my program goes off and does other things. I use the .start() method which forks a new process to handle communication. When I catch the sigint and call sys.exit() though, the network port is still bound even after the python interpretor finishes. Here is my short code and the output: from multiprocessing.managers import BaseManager import threading import sys import signal import time class WorkManager(BaseManager): def __init__(self): BaseManager.__init__(self,address=('', 51114), authkey='chris') self.register('get_string', callable=self.getString) def getString(self): return "hi" manager = WorkManager() manager.start() def quit( arg1, arg2 ): sys.exit() signal.signal(signal.SIGINT, quit) #busy wait while 1: time.sleep(1) cab938@ubuntu:~$ python2.6 server.py ^Ccab938@ubuntu:~$ python2.6 server.py Process WorkManager-1: Traceback (most recent call last): File "/usr/local/lib/python2.6/multiprocessing/process.py", line 231, in _bootstrap self.run() File "/usr/local/lib/python2.6/multiprocessing/process.py", line 88, in run self._target(*self._args, **self._kwargs) File "/usr/local/lib/python2.6/multiprocessing/managers.py", line 517, in _run_server server = cls._Server(registry, address, authkey, serializer) File "/usr/local/lib/python2.6/multiprocessing/managers.py", line 136, in __init__ self.listener = Listener(address=address, backlog=5) File "/usr/local/lib/python2.6/multiprocessing/connection.py", line 97, in __init__ self._listener = SocketListener(address, family, backlog) File "/usr/local/lib/python2.6/multiprocessing/connection.py", line 217, in __init__ self._socket.bind(address) File "<string>", line 1, in bind error: [Errno 98] Address already in use Traceback (most recent call last): File "server.py", line 16, in <module> manager.start() File "/usr/local/lib/python2.6/multiprocessing/managers.py", line 499, in start self._address = reader.recv() EOFError cab938@ubuntu:~$ Comments? Is this a bug, or is there a better way to clean up the manager myself before shutting down? Chris -- View this message in context: http://www.nabble.com/multiprocessing-BaseManager-doesn%27t-clean-up-net-co… Sent from the Python - python-list mailing list archive at Nabble.com.

1 0

How to debug embeding Python?
by Hongtian Jan. 1, 2009

Jan. 1, 2009

Hi Friends, My application is written in C/C++ and Python is embed to extend some functions (several .py files are invoked). But I am confused how to debug these embed Python? Can I use 'print-debuging'? and where can I capture the output string? Or Python can support 'break' debug for such scenario? Thanks.

4 4

Re: multiprocessing vs thread performance
by Aaron Brady Jan. 1, 2009

Jan. 1, 2009

On Dec 29, 9:29 am, mk <mrk...(a)gmail.com> wrote: > Christian Heimes wrote: > > mk wrote: > >> Am I doing smth wrong in code below? Or do I have to use > >> multiprocessing.Pool to get any decent results? > > > You have missed an important point. A well designed application does > > neither create so many threads nor processes. > > Except I was not developing "well designed application" but writing the > test the goal of which was measuring the thread / process creation cost. > > > The creation of a thread > > or forking of a process is an expensive operation. > > Sure. The point is, how expensive? While still being relatively > expensive, it turns out that in Python creating a thread is much, much > cheaper than creating a process via multiprocessing on Linux, while this > seems to be not necessarily true on Mac OS X. > > > You should use a pool > > of threads or processes. > > Probably true, except, again, that was not quite the point of this > exercise.. > > > The limiting factor is not the creation time but the communication and > > synchronization overhead between multiple threads or processes. > > Which I am probably going to test as well. I had an idea. You could use 'multiprocessing' for synchronization, and just use an mmap for the actual communication. (I think it's a good idea.)

3 3