Michael Heuberger

feel free to submit a PR. doubt i will find time for this in the next months to come ...

Thanks for bringing this to my attention. Busy family father here. If you could fork this repository, display this problem in a new failing unit test case, that would be...

I think we already have some unit tests toward the end covering some deeply, nested random files. Have a look. I hope this helps.

thanks for your efforts - a couple of problems with your commits: * do not rename file otherwise git history is lost * in general i disklike promises. too magical...

great, looking much better. we're close. about the 0.0005 ... can you change it to a constant at top of test and try to increment it slowly until all test...

and yes, don't forget documentation ;)

or it is because we have different CPU speeds? what's your CPU?

okay your machine is faster than mine, although tests pass here. are you on mac os?

ok, understand. in that case change increase slightly, maybe to 1ms or more until all tests pass.

@davidlanger you still alive?