SOSRepair: Expressive Semantic Search for Real-World Program Repair

Afzal, Afsoon; Motwani, Manish; Stolee, Kathryn T.; Brun, Yuriy; Goues, Claire Le

doi:10.1109/tse.2019.2944914

Cited by 34 publications

(15 citation statements)

References 121 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ManyBugs [16] benchmark consists of 185 defects taken from nine large, open source C projects. This benchmark is commonly used in evaluating automatic program repair tools [9], [26], [27]. Prior work [28] argues for explicitly defining the defect classes while evaluating various program repair tools, to ensure a fair comparison of tools on comparable classes.…”

Section: Discussionmentioning

confidence: 99%

“…TRIDENT supports the defect classes of Angelix 3 while additionally supporting those defect classes with side effects that assignment can model. Following previous work [27], we eliminated the samples that do not belong to TRIDENT's defect classes or that TRIDENT could not compiled due to version incompatibilities; this led to MB37, which is a dataset of 37 samples.…”

Section: Discussionmentioning

confidence: 99%

“…TRIDENT with disabled patch prioritisation; TP TRIDENT with enabled patch prioritisation; PR Prophet [26] with default configuration; SOS SOSRepair [27] with default configuration; AKN Angelix-like assignment synthesis with disabled KLEE merging;…”

Section: Tnmentioning

confidence: 99%

“…TRIDENT addresses this important limitation. SOSRepair [27] can synthesize patches with side effects, but, to do so, relies on a database of patterns extracted from the program.…”

Section: Automated Program Repairmentioning

confidence: 99%

See 3 more Smart Citations

Trident: Controlling Side Effects in Automated Program Repair

Parasaram¹,

Barr²,

Mechtaev³

2021

IIEEE Trans. Software Eng.

View full text Add to dashboard Cite

The goal of program repair is to eliminate a bug in a given program by automatically modifying its source code. The majority of real-world software is written in imperative programming languages. Each function or expression in imperative code may have side effects, observable effects beyond returning a value. Existing program repair approaches have a limited ability to handle side effects. Previous test-driven semantic repair approaches only synthesise patches without side effects. Heuristic repair approaches generate patches with side effects only if suitable code fragments exist in the program or a database of repair patterns, or can be derived from training data. This work introduces Trident, the first test-driven program repair approach that synthesizes patches with side effects without relying on the plastic surgery hypothesis, a database of patterns, or training data. Trident relies on an interplay of several parts. First, it infers a specification for synthesising side-effected patches using symbolic execution with a custom state merging strategy that alleviates path explosion due to side effects. Second, it uses a novel component-based patch synthesis approach that supports lvalues, values that appear on the left-hand sides of assignments. In an evaluation on open-source projects, Trident successfully repaired 6 out of 10 real bugs that require insertion of new code with side effects, which previous techniques do not therefore repair. Evaluated on the ManyBugs benchmark, Trident successfully repaired two new bugs that previous approaches could not. Adding patches with side effects to the search space can exacerbate test-overfitting. We experimentally demonstrate that the simple heuristic of preferring patches with the fewest side effects alleviates the problem. An evaluation on a large number of smaller programs shows that this strategy reduces test-overfitting caused by side-effects, increasing the rate of correct patches from 33.3% to 58.3%.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Tnmentioning

confidence: 99%

“…TRIDENT addresses this important limitation. SOSRepair [27] can synthesize patches with side effects, but, to do so, relies on a database of patterns extracted from the program.…”

Section: Automated Program Repairmentioning

confidence: 99%

See 2 more Smart Citations

Trident: Controlling Side Effects in Automated Program Repair

Parasaram¹,

Barr²,

Mechtaev³

2021

IIEEE Trans. Software Eng.

View full text Add to dashboard Cite

show abstract

“…The method used for each of these steps can significantly affect the tool's success. Existing research in APR has mostly focused on devising novel patch generation algorithms (e.g., heuristic-based [12]- [16], constraint-based [17]- [20], and learning-based [21]- [23]) aimed to produce more correct patches. Recently, researchers have started investigating the effect of using different technologies, assumptions, and adaptations of fault localization techniques [24]- [29], and patch validation methodologies [30]- [35] on the performance of APR tools.…”

Section: Research Problem and Hypothesismentioning

confidence: 99%

High-Quality Automated Program Repair

Motwani¹

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Standard Components Query System Based on Logical Filtering and Semantic Retrieval

Huang,

Bian,

Yang

2024

Lecture Notes in Mechanical Engineering

View full text Add to dashboard Cite

The establishment of a robust standard components database is essential in various industries to streamline product development and ensure quality. This paper presents a system for querying standard components data, leveraging the power of logical filtering and semantic retrieval. The structured approach of this system includes a well-defined database structure, logical filtering capabilities at different data levels, and advanced semantic retrieval techniques. The outputs of the system demonstrate its effectiveness in handling user queries, analysing unstructured data, and providing meaningful feedback based on logical filtering outcomes. This research contributes to the efficient utilization of standard components data through an innovative and powerful digital query system.

show abstract

SOSRepair: Expressive Semantic Search for Real-World Program Repair

Cited by 34 publications

References 121 publications

Trident: Controlling Side Effects in Automated Program Repair

Trident: Controlling Side Effects in Automated Program Repair

High-Quality Automated Program Repair

Standard Components Query System Based on Logical Filtering and Semantic Retrieval

Contact Info

Product

Resources

About