This package is my entry in @colmmacc's contest for a readable implementation of Daniel Lemire's algorithm for picking a random number between 0 and a given N.
pickrandcontains the documentation, test and implementation for the algorithm. Godoc is the best option to read the docs; I hope I did a good job in distilling the original paper and why it works.main.gohas a sample usage forpickrand.Uint64: generating a random network using Barabási-Albert's model, where most nodes have few connections and a few nodes have MOST connections. I also compute a vertex cover for the generated network, so it's clear that from a few nodes one can reach all others in one hop.statshas an attempt on creating statistical tests to validate that the produced numbers are indeed random and not biased. I can't say that I was successful, because even a clearly biased algorithm (simply$RANDOM % $N) didn't produce enough of a signal for Kolmogorov-Smirnov or Cramer-Von Mises tests, in the ranges that I tried, to clearly indicate a biased distribution. Most likely I did something wrong, don't use this code.
Enjoy!