Optimising the fit of stack overflow code snippets into existing code

Files

hdl_128857.pdf (765.08 KB)
  (Accepted version)

Date

2020

Authors

Reid, B.
Treude, C.
Wagner, M.

Editors

Advisors

Journal Title

Journal ISSN

Volume Title

Type:

Conference paper

Citation

Proceedings of the 2020 Genetic and Evolutionary Computation Conference (GECCO'20), 2020, pp.1945-1953

Statement of Responsibility

Brittany Reid, Christoph Treude, Markus Wagner

Conference Name

Genetic and Evolutionary Computation Conference (GECCO) (8 Jul 2020 - 12 Jul 2020 : Cancun, Mexico)

Abstract

Software developers often reuse code from online sources such as Stack Overflow within their projects. However, the process of searching for code snippets and integrating them within existing source code can be tedious. In order to improve efficiency and reduce time spent on code reuse, we present an automated code reuse tool for the Eclipse IDE (Integrated Developer Environment), NLP2TestableCode. NLP2TestableCode can not only search for Java code snippets using natural language tasks, but also evaluate code snippets based on a user's existing code, modify snippets to improve fit and correct errors, before presenting the user with the best snippet, all without leaving the editor. NLP2TestableCode also includes functionality to automatically generate customisable test cases and suggest argument and return types, in order to further evaluate code snippets. In evaluation, NLP2TestableCode was capable of finding compilable code snippets for 82.9% of tasks, and testable code snippets for 42.9%.

School/Discipline

Dissertation Note

Provenance

Description

Access Status

Rights

© 2020 Association for Computing Machinery.

License

Call number

Persistent link to this record