System and method for adaptive software testing

A computer readable medium storing instructions that, when executed by a computer, cause the computer to perform a method of mapping software components to test cases that test the software components, the method comprising:

• for each component in a set of software components, generating a corresponding component node in a bipartite graph;
• for each test case in a set of test cases configured to test the software components, generating a corresponding test case node in the graph;
• for each test case:
• applying said test case to test one or more software components;
• collecting data reflecting application of said test case;
• and
• updating the graph with edges coupling said corresponding test case node and one or more component nodes corresponding to software components tested by the test case;
• and
• for each software component tested by a test case, calculating a coverage rating configured to indicate the coverage achieved from executing said test case on said software component wherein the coverage rating is normalized to exclude dead code.

A computer readable medium containing a data structure configured for mapping software components to test cases configured to test the software components, the data structure comprising:

• a set of component nodes, each said component node corresponding to a software component;
• a set of test case nodes, each said test case node corresponding to a test case configured to test one or more software components;
• and
• edges coupling each said test case node to the component nodes corresponding to the software components tested by said corresponding test case, wherein each said edge coupling a test case node and a component node has an associated rating configured to reflect the coverage achieved from executing the corresponding test case on the corresponding software component, and wherein the coverage rating is normalized to exclude dead code.

A computer system for testing software, the computer system comprising:

• a set of software components;
• a set of test cases configured to test the software components;
• a testing tool configured to apply each test case in the set of test cases against a subset of the set of software components;
• a graph engine configured to generate a graph comprising:
• component nodes corresponding to the software component;
• test case nodes corresponding to the test cases;
• and
• edges coupling each test case node to component nodes corresponding to software components tested by the corresponding test case;
• and
• a processor configured to generate, for each software component tested by a test case, a rating indicating the coverage achieved from executing the test case on the software component, wherein the coverage rating is normalized to exclude dead code.