Hello, I am using a product that has a built-in Python interpreter (ESRI ArcGIS Desktop 10.0 SP3) and have implemented multiprocessing in script that can be run by a tool within the application using the built-in interpreter.
The way the built-in interpreter works is incompatible with multiprocessing as it appears to want to start multiple instances of the host application when the subprocesses are created. I am using multiprocessing.Pool with the apply_async function. It works great in a standalone script and I'd rather not rewrite what I've done so far if I can help it, but if there is a way to simply force the standard interpreter to be used for the subprocesses that would be ideal. Thinking about it I realize that this might not be possible due to whatever messaging, pickling, and queueing that the main processes has to handle for the subprocesses, but thought I would ask anyways if only for my own enlightenment. TIA!
-- http://mail.python.org/mailman/listinfo/python-list
