kyoob has asked for the wisdom of the Perl Monks concerning the following question:
This sort of picks up on the question asked in Test::Harness and the Debugger. I'm trying to run 4 or so tests in parallel on Win 7 but the processes only get so far as they can before the execution of the first test assertion run in the whole pool. E.g. if four of these test scripts each needed to open up a browser window before running its first test, I'd see maybe two browser windows come up before the first script in the pool actually makes an assertion and the rest of the processes stop dead, only to resume execution serially when the first test script is executed completely.
I'm not sure where the scripts are getting stopped up. Maybe in Test::More, maybe in the scripts themselves. What I thought I'd do was run them this way through a profiler, but I'm not sure how to do it with prove. I see how a debugger can be used when I'm running a .t script like any other script, but is there a way to debug using prove -j4?
|
|---|
| Replies are listed 'Best First'. | |
|---|---|
|
Re: Profiling via prove?
by tobyink (Canon) on Jan 24, 2014 at 10:00 UTC | |
by kyoob (Initiate) on Jan 24, 2014 at 17:43 UTC | |
by kyoob (Initiate) on Jan 24, 2014 at 18:20 UTC |