Deriving efficient program transformations from rewrite rules

An efficient optimizing compiler can perform many cascading rewrites in a single pass, using auxiliary data structures such as variable binding maps, delayed substitutions, and occurrence counts. Such optimizers often perform transformations according to relatively simple rewrite rules, but the subtle interactions between the data structures needed for efficiency make them tricky to write and trickier to prove correct. We present a system for semi-automatically deriving both an efficient program transformation and its correctness proof from a list of rewrite rules and specifications of the auxiliary data structures it requires. Dependent types ensure that the holes left behind by our system (for the user to fill in) are filled in correctly, allowing the user low-level control over the implementation without having to worry about getting it wrong. We implemented our system in Coq (though it could be implemented in other logics as well), and used it to write optimization passes that perform uncurrying, inlining, dead code elimination, and static evaluation of case expressions and record projections. The generated implementations are sometimes faster, and at most 40% slower, than hand-written counterparts on a small set of benchmarks; in some cases, they require significantly less code to write and prove correct.

Evaluasi Deteksi Smell Code dan Anti Pattern pada Aplikasi Berbasis Java

This paper presents an evaluation result of smell code and anti pattern detection in java based application development. The main objective to be achieved in this research is to determine the proper way in the detection of smell code and anti pattern in the development of java based software, and to evaluate the impact of using code inspection tools and software metrics to refactoring code in java based software development. Smell code to be detected in this research is Long Parameter List, Large Class, Lazy Class, Feature Envy, Long Method, and Dead Code. Anti pattern that will be detected is The Blob / God Class and Lava Flow. The selection of smell code and anti pattern is based on the definition, characteristics, detection factor, and software metrics. To support the research process is done through the evaluation stage of a case study java based application as a sample for inspection of code for the detection of smell code and anti pattern and calculation software metrics. Case studies of selected applications as sample applications are E-Commerce applications with functional master data management of goods and customers as well as management of sales and payment transactions. The detection of the smell code and anti-pattern on the case study is done in stages so it can be determined whether or not to refact. As well as ensuring the technique of making the program better fit the characteristics and rules of object-oriented programming.