Catching exceptions with multi-processing

Fabien fabien.maussion at gmail.com
Sat Jun 20 05:01:43 EDT 2015


On 06/19/2015 10:58 PM, Chris Angelico wrote:
> AIUI what he's doing is all the subparts of task1 in parallel, then
> all the subparts of task2:
>
> pool.map(task1, dirs, chunksize=1)
> pool.map(task2, dirs, chunksize=1)
> pool.map(task3, dirs, chunksize=1)
>
> task1 can be done on all of dirs in parallel, as no instance of task1
> depends on any other instance of task1; but task2 should be started
> only if all task1s finish successfully. OP, is this how it is? If not,
> I apologize for the noise.

That's it! Thanks for clarifying, I might have trouble explaining myself 
sometimes ;-)

Fabien



More information about the Python-list mailing list