Volodymyr Agafonkin

@stepankuzmin it should work with only `updateData` in any case. Must be some kind of a race condition, need to look into it...

Also wondering: is this increase for smaller queries a fluke due or is there an overhead? > before: 1000 searches 0.01%: 11.675ms > after: 1000 searches 0.01%: 15.793ms

@muendlein might be worth increasing the number of searches between timings for a possibly more reliable measurement. Does the last commit help? I'll likely land this anyway eventually since it's...

> After observing that the gap still exists even if the code is never reached, I'm unfortunately out of ideas. One idea is to set an empyrical threshold of the...

@muendlein I've seen similar behavior before, and my guess is that it's because of v8 inlining. Over a certain threshold of complexity or size, v8 stops inlining the function, which...

@muendlein this one looks much better! - I guess you can inline `addLeafSegment` for simplicity since it shouldn't affect inlining of the main search function, right? - Now that inlining...

@muendlein so, if I do https://github.com/mourner/flatbush/commit/fb78a2e1223820996c3b4850e7bf449596f63852 and then put `if (false && minX

> the penalty for a wrong guess is pretty high It's high compared to the case when we guess right; however it's very small compared to not landing this PR....

I'm hesitant about introducing such an option, because I'd like to keep the library simple, minimal and working perfectly out of the box, and this parameter is pretty confusing and...

@muendlein all right, let this sit for a few days more, I'll try to play with it a bit... As a last resort, we could just add a duplicate method,...