Sümer Cip

IIUC, this issue is same as #3. Please do not hesitate to open a PR against yappi as well. I will happily merge. Thank you!

You are absolutely right about that. 1/ You can look at a simple one that I have written recently here: https://github.com/blackfireio/python-sdk/blob/master/setup.py 2/ You should definitely have some kind of unit...

> Yeah, I've been telling myself the same thing since I created this repo, but I'm still trying to figure out how one is supposed to unit-test a timing/profiling tool,...

Hello everyone, I am the author of yappi and as I said in the issue opened by @dmontagu https://github.com/sumerc/yappi/issues/21 , I would be more than happy to try my best...

And regarding the other problem(wall time is not being correct). IIRC, this was something we have solved for gevent library. See this code here: https://github.com/ajdavis/GreenletProfiler/blob/700349864a4f368a8a73a2a60f048c2e818d7cea/_vendorized_yappi/_yappi.c#L577 _pause_ctx() _resume_ctx() So, every time...

I am also writing here for reference: (along with https://github.com/sumerc/yappi/issues/21) I have done few experiments on python 3.7 for `coroutine wall-time profiling`. It is in very alpha stage(far from production...

I would like to give a small update: This problem is *wayyy* harder than I originally thought as multiple threads are involved and coroutines are not profile friendly as various...

Latest update: Implementation is mostly finished, I am currently extensively writing more and more tests to validate corner cases, but initial tests seems pretty promising and somewhat stable. A new...

> The result is weird : if you increase tasks number the endpoint is slower and slower. Just my two cents on above: With profiling there is always the risk...

Hmm. Interesting, that is actually too much overhead. 2 questions: Are you using the latest code? The last commit I have commited this morning might decrease the overhead a lot....