Monthly Archives: January 2010

Compile time primality test

The powerful template mechanism of C++ allows us to write pretty complex Meta Functions, which are executed by the compiler during compilation. There are two basic types of meta-functions: one whose result is a type (mainly dealt with by Boost.MPL), and the other is a compile-time computation (which can result in any compile time constant). In this post we will review an example of the latter.

We would like to achieve a compile time boolean constant of the form is_prime::res. The problem with such meta programming task, in my opinion, is that it requires a functional programming mindset. Something that we, as C++ programmers, aren’t necessarily used to. But I am sure we will be able to tackle it anyway.

Continue reading Compile time primality test

Tuples

One of the containers introduced within TR1 (which is already widely available – both in gcc and Visual Studio) is a Tuple type, which is adopted from The Boost Tuple Library. Tuple types are very convenient at times; For example, it is possible to return multiple values from a function through a tuple, or write more intuitive and expressive code by utilizing tuples.

In this post we will examine the functionality offered by the new Tuple container, and have a go at profiling its performance. Actually, the results of said profiling were a small (pleasant) surprise to me.

Continue reading Tuples

Memoization

Memoization is essentially a fancy name for caching results for future use. A generalization of dynamic programming, if you will.

While I am certain most of us use it one way or another, in many occasions, it is usually through an Ad hoc implementation.. One that is only suitable for the specific, current, use case. Why don’t we generalize it further, and supply a generic, reusable, solution for Memoization?

Continue reading Memoization