Go-fuzz: Randomized testing for Go

It's amazing that go-fuzz found 137 bugs in the standard library. The design is based on American Fuzzy Lop (http://lcamtuf.coredump.cx/afl/)

Dmitry Vyukov gave a great talk at Gophercon on go-fuzz and other dynamic analysis tools for go. Slides: http://go-talks.appspot.com/github.com/dvyukov/go-fuzz/slide...

I'm really glad to see this, it's one of the major remaining testing tools that I've wanted for golang. The fact that it has so many entries in its 'trophy case' is encouraging.

The last thing I'm waiting for is mutation testing. There is this library: https://github.com/StefanSchroeder/Golang-Mutation-testing, but it hasn't received a lot of attention yet.

Mutation testing alters your code to verify that your test suite covers all potential simple programming mistakes. For example:

  if val < otherVal {

could be mutated to

  if val <= otherVal {

If the test suite doesn't fail after this change, it means that the test coverage is missing a potential off-by-one error. Fuzzing is good for directly finding problems using randomized processes, and mutation testing is good for finding shortcomings in your test suite.

Highly recommended: every time I use this library I find a bug.

Sadly, some people refuse to handle its results, which I simply cannot comprehend.