On Tue, Jul 9, 2013 at 10:30 AM, George Georgiev <george.georgiev.sf@gmail.com> wrote:

Hi,

I find outputting the action after it is done instead before it starts a little bit confusing and in some cases dysfunctional. For example if a unittest hang you do not know what is going on.

Is there option or a simple way to change this behavior?

I don't remember an option to control that.

Another related question: can b2 start flushing the output from subprocesses immediately - instead of waiting they to completely finish.

Unfortunately both of those behaviors are intentional and required to get coherent output in the face of the parallel action execution. Without this behavior the output from multiple actions would be interleaved and be painfully hard for humans and impossible for computers to interpret the output.

About the only change I can think of is to switch to immediate flushing when we are using "-j1".

--
--
-- Grafik - Don't Assume Anything
-- Redshift Software, Inc. - http://redshift-software.com
-- rrivera/acm.org - grafik/redshift-software.com
-- 102708583/icq - grafikrobot/aim - grafikrobot/yahoo