Synthesis of interface specifications for Java classes

Abstract

While a typical software component has a clearly specified (static) interface in terms of the methods and the input/output types they support, information about the correct sequencing of method calls the client must invoke is usually undocumented. In this paper, we propose a novel solution for automatically extracting such temporal specifications for Java classes. Given a Java class, and a safety property such as "the exception E should not be raised", the corresponding (dynamic) interface is the most general way of invoking the methods in the class so that the safety property is not violated. Our synthesis method first constructs a symbolic representation of the finite state-transition system obtained from the class using predicate abstraction. Constructing the interface then corresponds to solving a partial-information two-player game on this symbolic graph. We present a sound approach to solve this computationally-hard problem approximately using algorithms for learning finite automata and symbolic model checking for branching-time logics. We describe an implementation of the proposed techniques in the tool JIST-Java Interface Synthesis Tool-and demonstrate that the tool can construct interfaces accurately and efficiently for sample Java2SDK library classes. ABSTRACTWhile a typical software component has a clearly specified (static) interface in terms of the methods and the input/output types they support, information about the correct sequencing of method calls the client must invoke is usually undocumented. In this paper, we propose a novel solution for automatically extracting such temporal specifications for Java classes. Given a Java class, and a safety property such as "the exception E should not be raised", the corresponding (dynamic) interface is the most general way of invoking the methods in the class so that the safety property is not violated. Our synthesis method first constructs a symbolic representation of the finite state-transition system obtained from the class using predicate abstraction. Constructing the interface then corresponds to solving a partial-information two-player game on this symbolic graph. We present a sound approach to solve this computationally-hard problem approximately using algorithms for learning finite automata and symbolic model checking for branching-time logics. We describe an implementation of the proposed techniques in the tool JIST-Java Interface Synthesis Tool-and demonstrate that the tool can construct interfaces accurately and efficiently for sample Java2SDK library classes.

Automatic Synthesis of Assumptions for Compositional Model Checking

“…The result of model checking is the same if we use the assumption or the original process. Second, while using the assumption may significantly accelerate the model checking, there is no penalty in the form of increased complexity as introduced by the intermediate state explosion problem [10,8] or by using deterministic automata [6,2,1,8]. In the worst case, the generated assumption is as large as the process itself.…”

“…To obtain a coarser forward equivalence relation F , we refine ≈ by distinguishing any two states m 1 …”

“…Several researchers have therefore investigated a different construction based on the L * algorithm, a learning technique for deterministic automata [6,2,1]. In this setting, a candidate assumption A , represented as a deterministic automaton, is evaluated against both N and P by model checking.…”

Automated Integration of Service-Oriented Software Systems

“…We focus on black-box/grey-box techniques able to elicit behavioural models of the software. The reader interested on white-box techniques can refer to [3,37,38] and references therein.…”

Automatic Extraction of Behavioral Models from Distributed Systems and Services

“…(We will detail the concepts of relevant and essential CFG nodes in Section 4.2). This is different from the classical approach of defining the states as corresponding to predicates over the state variables, as done in the related approaches in the context of specification mining by static analysis for classes [13], [2]. We have chosen this approach because in real applications all the state variables can be complex data structures and it may be a complex task to determine predicate abstractions in this case.…”