Vetting Anti-patterns in Java to Kotlin Translation

Abstract

With Kotlin becoming a viable language replacement for Java, there is a need for translators and data flow analysis libraries to create maintainable and readable source code. Instagram, Uber, and Gradle are only a few of the large corporations that have either switched from Java to Kotlin completely or started to use it in internal tools in order to reduce code base size. Developers have claimed that Kotlin is fun to use in comparison to Java and much of the boilerplate code is reduced. With Java being the main language for the open source organization, PhenoApps, there is a need to support both Java and Kotlin to increase the maintainability of the code. Fortunately, JetBrains has an open-source IDE plugin for translating Java to Kotlin; however, the translation has some fundamental issues which shall be discussed further in this paper. Introducing, j2k, a CLI translation tool which includes various anti-pattern detection for syntactical formatting, performance, and other Android requirements. The new tool introduced within this paper, j2kCLI allows users to directly translate strings of Java code to Kotlin, or entire directories. This facilitates the maintainability of a large open source code base.

An experience-based recommendation system to support migrations of Android applications from Java to Kotlin

“…Other works have focused on helping developers to develop Kotlin apps. For example, Courtney and Neilsen present a tool, named j2kCLI [55], that allows users to translate Java code to Kotlin faster than the same functionality provided by Android Studio. From the JetBrains research group, Bryksin et al [56] investigated code anomalies in Kotlin and whether these anomalies could improve the Kotlin compiler.…”