Systematic Mapping Study: The Use of Transpilers

Abstract

A systematic mapping of the literature on the topic of transpiler use was conducted. Research questions addressing the research topic have been compiled. A search for publications narrating the topic of transpilers in various digital sources was conducted. The selection of unique and relevant publications was performed. A quantitative distribution of works by type of publication channel, geography, and quantitative composition of authors was performed. Based on the obtained sample of publications the answers to the research questions were obtained. The terms used in place of the term "Transpiler" were researched. The most common synonym turned out to be the term "Source-to-source compiler". The fields of application of transpiling were investigated. Most often transpilers were used for code generation, computation, and optimization purposes. The programming languages most commonly encountered in research were also examined. The most common programming languages were C/C++ and Python. It was also noted that JavaScript is never the source language of transpilation. Correlations between research questions were examined. A correlation was found between research questions of synonyms and fields of application, where the names transpiler and transcompiler are more often used in the field of code generation. A connection was also found between research questions of synonyms and programming languages, where transpilers used within the same programming language are more often referred to as source-to-source compiler. A growing trend in the number of studies on transpiler topics was demonstrated. The prospects of the work were also identified – the classification of programming languages by their level (low, high) and the study of correlation with other classifications.