Identifying similar code fragments in Haskell

Identifying syntactically or functionally similar code fragments in source code is an important research question in software engineering. Many researchers pointed out that detecting and refactoring code clones can be beneficial for software maintenance purposes. This work presents a clone detector algorithm. The proposed approach uses a source code AST representation and identifies similar fragments by generating and filtering pattern trees created from subtree pairs. The algorithm extends the existing technique of tree comparison by including Munkers assignment algorithm (The Hungarian method) and the longest common subsequence algorithm. Munkres assignment algorithm helps to identify rearranged code fragments. The LCS algorithm is used to compare subtrees of different syntactic variants. Experiments showed that these algorithms can help better evaluate code fragments to identify duplicates.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Computer Science

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2020

Author dc:creator

• Vardishvili, Mariam

Contributors dc:contributor

• Gunter, Elsa L

• Cloning
• Software maintenance
• duplicated code
• Haskell

Statement dc:rights

• Copyright 2019 Mariam Vardishvili

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/106294

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/106294