Sustav za predlaganje izmjena programskog koda u jeziku Python na razini metoda

master's thesis

• Title (english): Recommendation System for Method-Level Python Code Changes
• Author: Tomislav Ante Brnčić
• Mentor: Marin Šilić (mentor)
• Committee member: Marin Šilić (predsjednik povjerenstva)
• Committee member: Siniša Srbljić (član povjerenstva)
• Committee member: Goran Delač (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2019-07-08, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom radu predložen je sustav za preporuku promjena koda na razini metoda. Predloženi sustav stvara uzorke preporuka iz skupa podataka. Sustav analizira apstraktna sintaksna stabla "loših" i "dobrih" primjera koda i stvara generalizirani uzorak umetanjem wildcard i use konstrukata u izvorni kod. Generalizirani uzorak čuva sintaktički važne dijelove koda kako bi ponudio točne preporuke. Ti uzorci se uspoređuju s dostavljenim kodom i u slučaju podudaranja se koriste za stvaranje preporuka. Preporuka podržava promjenu u semantici izvornog koda i time povećava točnost sustava. Rezultati koji su prikazani u ovom radu pokazuju sposobnost predloženog rješenja za generiranje točnih preporuka za poboljšanje koda.

Abstract (english)

In this thesis, a system for recommending method-level code changes to Python source code is proposed. The proposed system creates recommendation patterns from labeled dataset. The system analyzes the abstract syntax trees of "bad" and "good" code examples and creates a generalized pattern by inserting wildcard and use constructs in the original code. The generalized pattern keeps syntactically important parts of code in order to provide accurate recommendations. These patterns are compared with submitted code and, when matched, are used to create a recommendation that improve ssubmitted code. The recommendation supports the change in original code’s semantics and thus increases the accuracy of the system. The results presented in this thesis show the viability of the proposed solution in generating accurate code improvement recommendations.

• Keywords: sustav za preporuke; promjena koda na razini metode; refaktoriranje; uzorak; apstraktno sintaksno stablo; diferenciranje stabala
• Keywords (english): recommendation system; method-level code change; refactoring; pattern; abstract syntax tree; tree differencing
• Language: english
• URN:NBN: urn:nbn:hr:168:393876
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2020-01-07 17:03:11