Identification and extraction of 'domain independent' components in large programs

The problem of the identification and extraction of domain independent components from a large working program lacking complete documentation is addressed. A modified version of Weiser's slice (Direct Slice) is defined, which makes it possible to individuate and extract particular kinds of code segments distributed among many of the modules composing the structure of the working program. Program components may be individuated and aggregated coherently according to the principles of information hiding. The techniques described are characterized by a certain simplicity and can thus be automized. The study is based on the results of an experimental project performed on applicative programs used in the banking sector, characterized by a high grade of difficulty as regards maintenance.<
>