Payload generation for computer software testing

摘要

A method of generating test payloads for a target system includes receiving a plurality of reference programs, each reference program modelling at least one aspect of the target system, building a specification for each received reference program, each specification defining illegal states for the respective reference program, analyzing each specification to determine one or more entry constraints that would generate an illegal state from a specific input, and synthesizing one or more payloads from the determined entry constraints.

主权项

1. A method of generating test payloads for a target system, the method comprising:
receiving a plurality of reference programs, each reference program modelling business logic of at least one aspect of the target system, building a specification for each received reference program, each specification defining illegal states and control locations in the execution of the respective reference program with respect to a functional property, performing a static analysis of each specification to determine one or more entry constraints that, when input to the business logic, would generate an illegal state from a specific input compatible with the one or more entry constraints, the static analysis performed without executing the respective reference program, and synthesizing one or more payloads from the determined entry constraints.