Identifying classes in legacy javaScript code

• Ano de defesa: 2017
• Autor(a) principal: Leonardo Humberto Guimar?es Silva
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Federal de Minas Gerais
• Palavras-chave em Português: Reengenharia; Computação; Engenharia reversa; Compreensão de programas; JavaScript (Linguagem de programação de computador); Javascript
• Link de acesso: https://hdl.handle.net/1843/JCES-ATKN9X

JavaScript is the most popular programming language for the Web. Although the language is prototype-based, developers often emulate class-based abstractions in JavaScript to master the increasing complexity of their applications. Identifying structures similar to classes in JavaScript code can support these developers in the following activities: (i) program comprehension; (ii) migration to the new JavaScript syntax that supports classes, introduced by ECMAScript 6 (ES6); and (iii) implementation of supporting tools, including IDEs with class-based views and reverse engineering tools. In this thesis, we dene, implement, and evaluate a set of heuristics to identify class-like structures, and their dependencies, in legacy JavaScript code, i.e., code implemented in versions prior to ES6. We report on a large and in-depth study to understand how class emulation is employed, using a dataset of 918 JavaScript applications available on GitHub. We found evidence that structures emulating classes are present in almost 70% of the studied systems. We perform a eld study with 60 developers to evaluate the accuracy of our strategy and tool. The results indicate that our tool is able to identify class-like structures in legacy JavaScript systems. We also demonstrate how to use a static type-checker to infer types that correspond to class references. We perform a study with two open-source applications aiming to measure the accuracy of the proposed approach to identify class-to-class dependencies. We achieve precision of 100% in both systems, and the values of recall ranges from 80% to 86% for dependencies in general and from 85% to 96% for associations. Moreover, we present a set of rules to migrate class-like structures to use the new ES6 class syntax. In our study, we detail cases that are straightforward to migrate (the good parts), cases that require manual and ad-hoc migration (the bad parts), and cases that cannot be migrated due to limitations and restrictions of ES6 (the ugly parts). Finally, we present a set of reasons that can lead developers to postpone or reject the adoption of ES6 classes, based on the feedback received after submitting pull requests suggesting the migration