Inferring Semantic Interfaces of Data Structures

Abstract

Abstract. In this paper, we show how to fully automatically infer semantic interfaces of data structures on the basis of systematic testing. Our semantic interfaces are a generalized form of Register Automata (RA), comprising parameterized input and output, allowing to model control-and data-flow in component interfaces concisely. Algorithmic key to the automated synthesis of these semantic interfaces is the extension of an active learning algorithm for Register Automata to explicitly deal with output. We evaluated our algorithm on a complex data structure, a "stack of stacks", the largest of which we could learn in merely 20 seconds with less than 4000 membership queries, resulting in a model with rougly 800 nodes. In contrast, even when restricting the data domain to just four values, the corresponding plain Mealy machine would have more than 10 9 states and presumably require billions of membership queries.

Combining Black-Box and White-Box Techniques for Learning Register Automata

“…Registers and data parameters can be compared for equality. The authors demonstrate the effectiveness of their approach by inferring models of data structures [42] and extend the expressivity to allow for arbitrary data relations that meet certain learnability criteria [14,15]. Aarts et al develop a slightly different approach for inferring register automata models that can compare registers and data parameter for equality [1,5].…”

“…Model learning has been successfully used in several different application domains, including -generating conformance test suites of software components, a.k.a. learningbased testing (e.g., [38,39]), -finding mistakes in implementations of security-critical protocols (e.g., [3,16,[28][29][30]62]), -learning interfaces of classes in software libraries (e.g., [42]), -checking that a legacy component and a refactored implementation have the same behavior (e.g., [63]).…”

Grey-Box Learning of Register Automata

“…Model learning, also known as active automata learning, is a black-box technique for constructing state machine models of software and hardware components from information obtained through testing (i.e., providing inputs and observing the resulting outputs). Model learning has been successfully used in numerous applications, for instance for generating conformance test suites of software components [13], finding mistakes in implementations of security-critical protocols [8][9][10], learning interfaces of classes in software libraries [14], and checking that a legacy component and a refactored implementation have the same behaviour [19]. We refer to [17,20] for surveys and further references.…”

A Myhill-Nerode Theorem for Register Automata and Symbolic Trace Languages

“…active automata learning) is a black-box technique which constructs state machine models of software and hardware components from information obtained by providing inputs and observing the resulting outputs. Model learning has been successfully used in numerous applications, for instance for generating conformance test suites of software components [20], finding mistakes in implementations of security-critical protocols [13][14][15], learning interfaces of classes in software libraries [23], and checking that a legacy component and a refactored implementation have the same behavior [35]. We refer to [26,38] for surveys and further references.…”