Goanna Static Analysis by Red Lizard Software

I used to be a keen windsurfer and once attended a training camp/holiday in Greece. We covered advanced maneuvers such as 360’s, forward rolls and duck gybes etc. The instructor always said, “Doesn’t matter if you pull it off, just so long as it looks good!”

The reason for this story is that at Red Lizards we have a mantra of “Bringing higher quality software to market faster”, and so the flip side of this coin is that we can’t ship a release until it is of sufficiently high quality. Double edged sword perhaps, yet this is the business we chose.

Read the rest of this entry »

We’ve been mulling over how to add new unit tests to our OCaml codebase. Our first thought was to write a test suite apart from the code itself; we’ve already some end-to-end tests written in that way. The problem with that approach is that the tests are separated from the code they’re testing, which discourages writing tests.

A few weeks ago, we wrote some time- and heap-profiling CamlP4 macros, which add entries to a centralized table as our code runs. When the program finishes, it spits out statistics that we can examine for time and space sinks. The profiling hints — macros, that is — wrap the code to be profiled. So why not use the same approach for our tests?

Read the rest of this entry »

The visual studio version of Goanna is almost ready:

This is the plugin being run over one of the PLT Scheme projects, mzscheme. A major part of our testing routine is using Goanna on open source projects.

We will have a trial version of our Visual Studio plugin for you soon!

So what’s different about Goanna and why did we approach the static analysis challenge with model checking?

In case you weren’t aware; Goanna is the first static analysis tool to implement model checking after 4 years research at NICTA (National ICT Australia), which is Australia’s leading ICT government research organisation. 

The result  is that the key attributes for Goanna’s differences are as follows;
Read the rest of this entry »

Internally we are encouraging new team members to watch the so called Sean Tapes. There is nothing dubious about these tapes, unless you call Static Analysis and OCaml “dubious”. They are actually just screen captures, in which Sean explains how the Goanna tool performs static analysis for C/C++ code.

We are happy to share Sean’s prized voice-over with you all. Here’s the first video on the use of Goanna.

OCaml is a wonderful language. But as with any language, there are certain features you would like to see.

The obvious one in OCaml is the lack of type-classes, specifically the lack of Haskell’s Show. There is no more useful debugging tool than printf, and it is a pain trying to debug complex types when you can’t see them.

This is where s-expressions come in. Specifically, Jane Street’s s-expression library.

Read the rest of this entry »

The other day we met with a new prospective alpha partner for our upcoming MSVS release (Visual Studio support is scheduled for July 2009). This customer has a lot of smarter people creating impressive code bases to challenging requirements.

Having presented how Goanna is different in that we fit within the IDE, as opposed to sitting with/behind the central build, I asked the question as to when and how they run peer reviews.

Read the rest of this entry »

A static analysis tool for Java is FindBugs. I found this interesting Google Video, in which Bill Pugh talks about static analysis. While he talks mostly about bugs in Java it is still quite interesting in general. His most interesting lesson is “Smart people make dumb mistakes”. I’d like to add, “Really smart people are glad to find their dumb mistakes, and are not afraid to get help finding them“. Read the rest of this entry »

Pascal Cuoq made an interesting comment on my last post about C developers accidentally writing “==” in OCaml when they meant to use “=”. It reminds me of a similar issue I run into, when I am writing in OCaml but thinking in Haskell, and I am confronted with a value of type:

   'a option list

Read the rest of this entry »

We have a set of end-to-end tests that run on Goanna every night. This ensures that our commits during the day don’t break our development tree too badly.

Here is one of those tests. It was producing a strange result.

void example(void) {
  int *x;
  x = malloc(sizeof(int));
  free(x);
  if (rand()) {
    x = malloc(sizeof(int));
  }

  *x++;
}

Read the rest of this entry »