In today’s world, our day-to-day lives depend on many software programs. The outage of the services of some crucial software like banking and airline will affect our lives badly. It is the necessity of this era to locate and repair the bug in software in less time. This paper presents a novel method to locate the bug at statement-level with the help of a code graph. The project code base is converted to a graph and stored in the graph database. The sub-graph related to the bug report is pulled from the graph database by using the bug report as the query. The nodes in the sub-graph are then listed in the decreasing order of their similarity score. Each node in the graph represents the statements in the code. Hence, the proposed method ensures fine-grained bug localization with the help of a code graph. Moreover, it gives the dependencies of the buggy code, which helps in test case generation. The proposed method reduces the effort for software maintenance and increases the productivity of the maintenance team.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Fine-Grained Bug Localization in Software Projects Using Code Graph

  • E. V. Sunitha

摘要

In today’s world, our day-to-day lives depend on many software programs. The outage of the services of some crucial software like banking and airline will affect our lives badly. It is the necessity of this era to locate and repair the bug in software in less time. This paper presents a novel method to locate the bug at statement-level with the help of a code graph. The project code base is converted to a graph and stored in the graph database. The sub-graph related to the bug report is pulled from the graph database by using the bug report as the query. The nodes in the sub-graph are then listed in the decreasing order of their similarity score. Each node in the graph represents the statements in the code. Hence, the proposed method ensures fine-grained bug localization with the help of a code graph. Moreover, it gives the dependencies of the buggy code, which helps in test case generation. The proposed method reduces the effort for software maintenance and increases the productivity of the maintenance team.