How to parse the stacked profile data?

[jmon12]

filprofiler offers a great overview via flamegraphs of the memory consumptions. I'm profiling using your great tool in various situations, and I'd like to "extract back" the actual numbers of the stacked profile, e.g. absolute and relative memory consumption of some functions. What would be the way to have some tree representation of the flamegraph and get back actual numbers?

E.g. something like

profile = parse(profile.prof)
RAM_func1 = profile.get('/main/func1')

I have the feeling to maybe take all of that upside down, disaggregate the friendly aggregated data... Should I go back to some basic tools like resource/psutil like in your article instead?

This is somehow related to #236 and #512