3 ms·
Go-fuzz: Randomized testing for Go
- rmcpherson 11y agoIt'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/ 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/slides/go-fuzz.slide#1 http://go-talks.appspot.com/github.com/dvyukov/go-fuzz/slide...
- wtbob 11y agoHighly recommended: every time I use this library I find a bug. Sadly, some people refuse to handle its results, which I simply cannot comprehend.
- Kototama 11y agoReally? That's surprising. Maybe they think the case where it fails is not susceptible to occur during a normal execution of the program. This reasoning could work if you don't define "a normal execution" :-)
- Taek 11y agoI'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 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.
- vessenes 11y agoMr. Vyukov is a beast; you should just post an issue on github and politely ask; I bet he'd enjoy implementing.