Algorithms and software tool for building equivalent representations of source codes

This article describes prototype of software tool for static analysis of software systems based on a special data set. This data set received from the source code of programs using the open source compiler. The prototype provides to obtain equivalent representation by linear and nonlinear transformations of this data set. The user interface is built by data query language, which syntax provides to set arbitrary combinations of available transformations, and additional parameters correspond to the objectives of the analysis. Conversion options are defined by configuration file being an XML-document. The prototype also provides analysis of the obtained equivalent representations. We descried “temperature” analysis of row syntactic overflow and complexity of source code blocks in the context of control flow by way of example. The prototype was tested on open source library for parsing of XML-documents.