{"title":"A Tool for Describing and Checking Natural Semantics Definitions of Programming Languages","authors":"Georgian-Vlad Saioc, Hans Hüttel","doi":"10.4204/EPTCS.369.4","DOIUrl":null,"url":null,"abstract":"Many universities have courses and projects revolving around compiler or interpreter implementation as part of their degree programmes in computer science. In such teaching activities, tool support can be highly beneficial. While there are already several tools for assisting with development of the front end of compilers, tool support tapers off towards the back end, or requires more background experience than is expected of undergraduate students. Structural operational semantics is a useful and mathematically simple formalism for specifying the behaviour of programs and a specification lends itself well to implementation; in particular big-step or natural semantics is often a useful and simple approach. However, many students struggle with learning the notation and often come up with ill-defined and meaningless attempts at defining a structural operational semantics. A survey shows that students working on programming language projects feel that tool support is lacking and would be useful. Many of these problems encountered when developing a semantic definition are similar to problems encountered in programming, in particular ones that are essentially the result of type errors. We present a pedagogical metalanguage based on natural semantics, and its implementation, as an attempt to marry two notions: a syntax similar to textbook notation for natural semantics on the one hand, and automatic verification of some correctness properties on the other by means of a strong type discipline. The metalanguage and the tool provide the facilities for writing and executing specifications as a form of programming. The user can check that the specification is not meaningless as well as execute programs, if the specification makes sense.","PeriodicalId":9644,"journal":{"name":"Catalysis Surveys from Japan","volume":"87 10 1","pages":"51-66"},"PeriodicalIF":0.0000,"publicationDate":"2022-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Catalysis Surveys from Japan","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.4204/EPTCS.369.4","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Many universities have courses and projects revolving around compiler or interpreter implementation as part of their degree programmes in computer science. In such teaching activities, tool support can be highly beneficial. While there are already several tools for assisting with development of the front end of compilers, tool support tapers off towards the back end, or requires more background experience than is expected of undergraduate students. Structural operational semantics is a useful and mathematically simple formalism for specifying the behaviour of programs and a specification lends itself well to implementation; in particular big-step or natural semantics is often a useful and simple approach. However, many students struggle with learning the notation and often come up with ill-defined and meaningless attempts at defining a structural operational semantics. A survey shows that students working on programming language projects feel that tool support is lacking and would be useful. Many of these problems encountered when developing a semantic definition are similar to problems encountered in programming, in particular ones that are essentially the result of type errors. We present a pedagogical metalanguage based on natural semantics, and its implementation, as an attempt to marry two notions: a syntax similar to textbook notation for natural semantics on the one hand, and automatic verification of some correctness properties on the other by means of a strong type discipline. The metalanguage and the tool provide the facilities for writing and executing specifications as a form of programming. The user can check that the specification is not meaningless as well as execute programs, if the specification makes sense.