| 摘要 |
<p>A method, system and program for testing programmatic interfaces by systematically exploring valid call sequences defined by a collection of subroutines with a collection of data. The subject invention does not explicitly write unit tests. Instead, it provides the tools to develop rules which model and verify the correct operation of software while leaving the complexities of test selection, execution, sequencing, coverage, and verification to an automated system. The number of test scenarios generated by the system depends upon cumulative execution time and is reliably independent of human effort.</p> |
