Semantic intensity based decomposition of software systems

A method for configuring a computer-readable medium comprising:

• forming transitions in a first tangible computer-readable medium, the transitions comprising digital patterns that are interpretable by a first computer as first instructions, which first instructions, when executed by the first computer, cause the first computer to perform a method of software analysis by:
• identifying software elements in a computer program source code written in an object-oriented or procedural computer programming language, wherein the software elements comprise at least a portion of variables, classes, packages, procedures, functions, and attributes having respective scopes, the source code having scopes comprising subsets of code in which the software elements have an influence, and wherein at least a portion of the software elements have a plurality of uses in at least one of the scopes of the source code;
• establishing respective roles of the software elements in the source code, the roles each having a predefined numeric weight, wherein the roles comprise an assignment pattern, a destination role and a source role within the assignment pattern, a variable creation role, a variable read operation role, a variable write operation role, a variable attribute access role, an if-pattern, an if-condition role of the if-pattern, and an if-body role of the if-pattern;
• calculating respective semantic intensity metrics for the software elements according to the weight of the roles thereof in the source code, wherein the semantic intensity metrics (SSI) are additive according to a number of the uses of the software elements in respective scopes, and wherein the semantic intensity metric in one of the scopes is calculated as a product
• SSI(e/s)=def(e)*ev(e/s),
• wherein def(e) denotes an SSI of the source code in the one scope that specifies an element e and ev(e/s) is an SSI of all roles of the element e within the one scope, and wherein a coupling factor between a first software element A and a second software element B comprises a software semantic intensity of the first software element A in a scope of the second software element B (SSIA/B), a software semantic intensity of the second software element B in a scope of the first software element A (SSIB/A), and a correlation between the software semantic intensity of the first software element A and the software semantic intensity of the second software element B in a plurality of the scopes in the source code;
• correlating the semantic intensity metrics of the software elements to determine coupling factors between different ones of the software elements, wherein the correlation between SSIA/B and SSIB/A in a scope of a procedure p is calculated as a product
• Cor
• ⁡
• (
• A
• ,
• B
• ,
• p
• )
• =
• min
• ⁡
• (
• SSI
• A
• |
• p
• ,
• SSI
• B
• |
• p
• )
• SSI
• p
• ·
• (
• SSI
• A
• |
• p
• SSI
• A
• +
• SSI
• B
• |
• p
• SSI
• B
• )
• ;
• organizing the software elements into an arrangement of distinct software components according to the coupling factors therebetween;
• responsively to the arrangement preparing an optimized computer program source code;
• and
• writing the optimized computer program source code onto a second tangible computer-readable medium.

A computer software product for analysis of software, including a non-transitory computer storage medium in which computer program instructions are stored, which instructions are organized as distinct software modules, the distinct software modules comprising at least a semantic intensity calculation module, a coupling calculation module, and a software decomposition module;

• and when executed by a computer, cause the computer to perform the steps of:
• identifying software elements in a computer program source code written in an object-oriented or procedural computer programming language, wherein the software elements comprise at least a portion of variables, classes, packages, procedures, functions, and attributes having respective scopes, the source code having scopes comprising subsets of code in which the software elements have an influence, and wherein at least a portion of the software elements have a plurality of uses in at least one of the scopes of the source code;
• establishing respective roles of the software elements in the source code, the roles each having a predefined numeric weight, wherein the roles comprise an assignment pattern, a destination role and a source role within the assignment pattern, a variable creation role, a variable read operation role, a variable write operation role, a variable attribute access role, an if-pattern, an if-condition role of the if-pattern, and an if-body role of the if-pattern;
• with the semantic intensity calculation module calculating respective semantic intensity metrics (SSI) for the software elements according to the weight of the roles thereof in the source code, wherein the semantic intensity metrics are additive according to a number of the uses of the software elements in respective scopes, and wherein the semantic intensity metric in one of the scopes is calculated as a product
• SSI(e/s)=def(e)*ev(e/s),
• wherein def(e) denotes an SSI of the source code in the one scope that specifies an element e and ev(e/s) is an SSI of all roles of the element e within the one scope;
• with the coupling calculation module correlating the semantic intensity metrics of the software elements to determine coupling factors between different ones of the software elements, wherein a coupling factor between a first software element A and a second software element B comprises a software semantic intensity of the first software element A in a scope of the second software element B (SSIA/B), a software semantic intensity of the second software element B in a scope of the first software element A (SSIB/A), and a correlation between the software semantic intensity of the first software element A and the software semantic intensity of the second software element B in a plurality of the scopes in the source code, wherein the correlation between SSIA/B and SSIB/A in a scope of a procedure p is calculated as a product
• Cor
• ⁡
• (
• A
• ,
• B
• ,
• p
• )
• =
• min
• ⁡
• (
• SSI
• A
• |
• p
• ,
• SSI
• B
• |
• p
• )
• SSI
• p
• ·
• (
• SSI
• A
• |
• p
• SSI
• A
• +
• SSI
• B
• |
• p
• SSI
• B
• )
• ;
• with the software decomposition module organizing the software elements into an arrangement of distinct software components according to the coupling factors therebetween;
• preparing an optimized computer program source code responsively to the arrangement;
• and
• writing the optimized computer program source code onto a tangible medium.

A data processing system for analysis of software, comprising:

• a processor;
• a memory accessible to the processor storing programs and data objects therein, the programs including at least a semantic intensity calculation module, a coupling calculation module, and a software decomposition module, wherein execution of the programs cause the processor to perform the steps of:
• identifying software elements in a computer program source code written in an object-oriented or procedural computer programming language, wherein the software elements comprise at least a portion of variables, classes, packages, procedures, functions, and attributes having respective scopes, the source code having scopes comprising subsets of code in which the software elements have an influence, and wherein at least a portion of the software elements have a plurality of uses in at least one of the scopes of the source code;
• establishing respective roles of the software elements in the source code, the roles each having a predefined numeric weight, wherein the roles comprise an assignment pattern, a destination role and a source role within the assignment pattern, a variable creation role, a variable read operation role, a variable write operation role, a variable attribute access role, an if-pattern, an if-condition role of the if-pattern, and an if-body role of the if-pattern;
• with the semantic intensity calculation module calculating respective semantic intensity metrics (SSI) for the software elements according to the weight of the roles thereof in the source code, wherein the semantic intensity metrics are additive according to a number of the uses of the software elements in respective scopes, and wherein the semantic intensity metric in one of the scopes is calculated as a product
• SSI(e/s)=def(e)*ev(e/s),
• wherein def(e) denotes an SSI of the source code in the one scope that specifies an element e and ev(e/s) is an SSI of all roles of the element e within the one scope;
• with the coupling calculation module correlating the semantic intensity metrics of the software elements to determine coupling factors between different ones of the software elements, wherein a coupling factor between a first software element A and a second software element B comprises a software semantic intensity of the first software element A in a scope of the second software element B (SSIA/B), a software semantic intensity of the second software element B in a scope of the first software element A (SSIB/A), and a correlation between the software semantic intensity of the first software element A and the software semantic intensity of the second software element B in a plurality of the scopes in the source code, wherein the correlation between SSIA/B and SSIB/A in a scope of a procedure p is calculated as a product
• Cor
• ⁡
• (
• A
• ,
• B
• ,
• p
• )
• =
• min
• ⁡
• (
• SSI
• A
• |
• p
• ,
• SSI
• B
• |
• p
• )
• SSI
• p
• ·
• (
• SSI
• A
• |
• p
• SSI
• A
• +
• SSI
• B
• |
• p
• SSI
• B
• )
• ;
• with the software decomposition module organizing the software elements into an arrangement of distinct software components according to the coupling factors therebetween;
• preparing an optimized computer program source code responsively to the arrangement;
• and
• writing the optimized computer program source code onto a tangible medium.