Remote pointcut
Muga Nishizawa, Shigeru Chiba, Michiaki Tatsubori
Abstract
Muga Nishizawa, Shigeru Chiba, Michiaki Tatsubori
Abstract
This paper presents our extension to AspectJ for distributed computing. Although AspectJ allows Java developers to modularize a crosscutting concern as an aspect, this paper shows that some crosscutting concerns in distributed computing are not modularized in AspectJ as simple aspects. Rather, aspects modularizing such a concern tend to be in code spread over multiple hosts and explicitly communicated across the network. This paper illustrates this fact with an example of testing a distributed program written in AspectJ with Java RMI. To address this complexity caused by network communication, this paper proposes an extension to AspectJ for distributed computing. The language construct that we call remote pointcut enables developers to write a simple aspect to modularize crosscutting concerns distributed on multiple hosts. This paper presents DJcutter, which is our AspectJ-like language supporting remote pointcuts.
OpenAlex reports 90 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
This paper presents our extension to AspectJ for distributed computing. Although AspectJ allows Java developers to modularize a crosscutting concern as an aspect, this paper shows that some crosscutting concerns in distributed computing are not modularized in AspectJ as simple aspects. Rather, aspects modularizing such a concern tend to be in code spread over multiple hosts and explicitly communicated across the network. This paper illustrates this fact with an example of testing a distributed program written in AspectJ with Java RMI. To address this complexity caused by network communication, this paper proposes an extension to AspectJ for distributed computing. The language construct that we call remote pointcut enables developers to write a simple aspect to modularize crosscutting concerns distributed on multiple hosts. This paper presents DJcutter, which is our AspectJ-like language supporting remote pointcuts.
Key concepts: AspectJ, Aspect-oriented programming, Computer science, Java, Programming language, Construct (python library), Simple (philosophy), Extension (predicate logic)