Metrics are one of the most advocated and less implemented quality-oriented practice. There are basically two reasons for this situation:

• Metrics are hard to collect, up to a point where it is unreasonable to consider gathering them otherwise than by automated means. Hence, one must rely on products for the collection of data.
  
  
• The results delivered by existing products are not flexible enough. For instance, base bone cyclomatic complexity produces biased results if the code it applies to uses long sequences of IF ... ELSE IF ... ELSE IF ... instead of the SELECT statement. Such coding guidelines are followed by organisations that aim at portability onto platforms where the available compilers do not support the SELECT statement.

RainCode takes your metrics strategy far beyond the initial (and often, lethal) wishful thinking stage. RainCode provides a convenient access to the entire source code in an entirely structured way. Hence, all possible metrics can be gathered easily, and the results it delivers can be tuned to meet one's need exactly.

For instance, even the most complex metrics, based on sophisticated semantic analysis of the code (such as fan-in, fan-out, for instance) can be produced easily by RainCode; while no simplistic text-based approach based on Perl or any other text processing tool can achieve the same result.