The Gibbs sampler discussed on Darren Wilkinson's blog and also on Dirk Eddelbuettel's blog has been implemented in several languages, the first of which was R.
The task is to create a Gibbs sampler for the unscaled density
f(x,y) = x x^2 \exp(-xy^2 - y^2 + 2y - 4x)
using the conditional distributions
x|y \sim Gamma(3, y^2 +4)
y|x \sim Normal(\frac{1}{1+x}, \frac{1}{2(1+x)})