ICSE '89 Proceedings of the 11th international conference on Software engineering
PROCOL: a parallel object language with protocols
OOPSLA '89 Conference proceedings on Object-oriented programming systems, languages and applications
LFP '90 Proceedings of the 1990 ACM conference on LISP and functional programming
Advances in computers
Applications of process algebra
Applications of process algebra
The essence of functional programming
POPL '92 Proceedings of the 19th ACM SIGPLAN-SIGACT symposium on Principles of programming languages
A system of constructor classes: overloading and implicit higher-order polymorphism
FPCA '93 Proceedings of the conference on Functional programming languages and computer architecture
OOPSLA '93 Proceedings of the eighth annual conference on Object-oriented programming systems, languages, and applications
Regular types for active objects
OOPSLA '93 Proceedings of the eighth annual conference on Object-oriented programming systems, languages, and applications
Extending context-free grammars with permutation phrases
ACM Letters on Programming Languages and Systems (LOPLAS)
Object-Oriented Software Construction
Object-Oriented Software Construction
An Object-Oriented Language-Database Integration Model: The Composition-Filters Approach
ECOOP '92 Proceedings of the European Conference on Object-Oriented Programming
Solving the Inheritance Anomaly in Concurrent Object-Oriented Programming
ECOOP '93 Proceedings of the 7th European Conference on Object-Oriented Programming
Behavior Protocols for Software Components
IEEE Transactions on Software Engineering
Bounding Component Behavior via Protocols
TOOLS '99 Proceedings of the Technology of Object-Oriented Languages and Systems
| Hi-index | 0.00 |
A service definition is an abstract specification of the behavior of a software component. It provides the interface between the users of an object and its (hidden) implementation. A protocol can be a part of a service definition. It captures the conditions under which interface operations can be invoked. This is of use for the designers of clients, but also for implementers of the service, at least if the protocol mechanism provides automatic (static or dynamic) acceptability checking of invocations or messages. Existing protocol formalisms are mostly based on finite state machines that describe legal orderings of messages. This is too limited, however, to model more complex services or to handle conditions that go beyond the ordering (such as access-control or time-dependencies) without referring to a service implementation.In this paper we explore a grammar-based approach to protocol definition, i.e. we define protocols as non-deterministic grammars extended with provisions for parallellism. The resulting protocols not only define legal message sequences but can also impose context-sensitive constraints on properties of messages, like its sender or a time stamp. Protocotols are truly abstract, i.e. independent of the implementation of a service. Since we want to investigate the kinds of constructs needed in specifying protocols we do not introduce a new notation, but instead express protocols as parsers written in a lazy functional language using a technique called combinator parsing. Besides giving a complete picture of the semantics of the constructs this also allows us to show how dynamic protocol checking can be provided. The approach is illustrated through several examples and its potential and limitations are discussed.