pprofile style context profiling

I’m usually not interested in profiling an entire program, I’m more interested in profiling some hotspot of code, or just some new piece of code.

pprofile allows for just profiling a specific region of code (from pprofile’s main page):

def someOtherHotSpotCallable():
    # Statistic profiler
    prof = pprofile.StatisticalProfile()
    with prof(
        period=0.001, # Sample every 1ms
        single=True, # Only sample current thread
    ):
        # Code to profile
    prof.print_stats()

It would be nice if scalene allowed for such granularity.

[ClemensA4t1qbit]

My bad; I used from scalene import profile, but the trick is to, as you say, just use @profile. I was just a bit confused because my IDE complained that in that case profile isn’t defined anywhere.

I changed it (removed my import) and it does seem to work.

PS: I loved your ACM Tech Talk. So many spot-on Simpsons references.

[emeryberger]

UPDATE: As of version 1.1.5, Scalene now supports @profile decorators.