Add getValueQuantile and unit-test.

[reconditesea]

Add getValueQuantile and unit-test.
Fix a bug (at least I think it is) in trueQuantile.

[johnynek]

can you add a few comments:

1. how do you know there is only one item in this list?
2. I am confused about the algorithm have >= and 1.0 in both cases. Can you comment a bit?

It seems to me that you want something like:
quantilePoints.filter { case @b(l, u) => (l <= value && value <= u) }
Why is that not what we want?

[reconditesea]

I'm thinking situations like this:
Given an ordered and unique list of quantiles, e.g. [0, 0.1, 0.2, 0.3,... 0.9, 1.0] and specific value x. This method will return the estimated quantile for x in this QT. Wondering if this is a valid and common use case?

Since the list is ordered and unique, there will be only one item in the list. Also It's possible that x's quantile is somewhere b/w the two items in the list. For example, x's quantile is 0.25 which is in the middle of [0.2, 0.3]. In this case, I think it should return the the interval as lower and upper bound. So that's why the algorithm uses >= to do the filtering. The 1.0 is because if no bounds in the QT is larger than value x, then I think its quantile should be 1.0 since it's larger than all values in this QT.
Does this make sense?

[reconditesea]

@johnynek Does the above make sense to you?

[ianoc]

Sorry my bad, git foo on cmd line broke stuff and closed all of these

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}