Supporting the analysis of clones in software systems

Abstract

Code duplication is a well-documented problem in industrial software systems. There has been considerable research into techniques for detecting duplication in software, and there are several effective tools to perform this task. However, there have been few detailed qualitative studies into how cloning actually manifests itself within software systems. This is primarily due to the large result sets that many clonedetection tools return; these result sets are very difficult to manage without complementary tool support that can scale to the size of the problem, and this kind of support does not currently exist. In this paper we present an in-depth case study of cloning in a large software system that is in wide use, the Apache Web server; we provide insights into cloning as it exists in this system, and we demonstrate techniques to manage and make effective use of the large result sets of clone-detection tools. In our case study, we found several interesting types of cloning occurrences, such as 'cloning hotspots', where a single subsystem comprising only 17% of the system code contained 38.8% of the clones. We also found several examples of cloning behavior that were beneficial to the development of the system, in particular cloning as a way to add experimental functionality.(1) facilities to evaluate the overall cloning situation;(2) mechanisms to guide users toward clones that are most relevant to their task; and (3) methods for filtering and refining the analysis of the clones.Each of these criteria is described in more detail below. Overall system evaluationAs a first step in understanding cloning within a software system, regardless of the end goal, maintainers must have an understanding of the cloning from a high level of abstraction. This understanding will allow the user to evaluate the extent and the severity of the duplication in order to estimate the cost and/or necessity of the task.Several mechanisms can be used to evaluate cloning from a high level. Visualization methods, such as scatterplots [1,3,4,12,15], are useful for the discovery of highly related subsystems and high levels of cloning within a subsystem. They are also useful for detecting unusual types of cloning, such as cloning from system libraries to other parts of the software system. Metric-oriented reports, such as reporting the percentage of lines cloned, average length of a clone, etc., are useful for directing users to points in the system where the most cloning is occurring, or where cloning activities are unusually high in relation to subsystem size. Guide and empower the userThe possibly large sets of clones returned by the clone-detection methods make it infeasible to look at each individual clone. There are several ways to direct users toward the clones they seek. Metrics can be used to query the dataset [16]. Some examples of metrics that might be used are the size of the clone, the types of changes made to the clone, and types of external dependencies a code segment has. Such a method can direct users to promising refactoring ...

An Empirical Study of Function Clones in Open Source Software

“…A code fragment can form a clone pair with another fragment within the same file, or it can form a clone pair with another fragment of a different file located in the same directory or with a code fragment that is located in a different file in a different directory. Kapser and Godfrey [15,14] provide a categorization of function clones based on such location differences and analyze the causes, usefulness, harmfulness and possible solutions for each kind of cloning. For example, clone pairs that appear in the same file may not be harmful as they are not physically apart and might be easily maintainable.…”

“…They provide a detailed categorization of code clones in the form of a taxonomy [15], propose a new analysis framework [14] and give an in-depth study on the harmfulness / usefulness of cloning [13]. Our study differs in that we focus on the comprehensive cloning status of a wide variety of different systems in different languages, whereas they focus on the maintenance implications of cloning.…”

"Cloning Considered Harmful" Considered Harmful

“…It is suspected that approximately 10%-15% of many large systems is part of duplicated code [2,12,24,25], and it has been documented to exist at rates of over 50% of the effective lines of code (ELOC) in a particular COBOL system [12]. The literature on the topic has described many situations that can lead to the duplication of code within a software system [2,7,21,22,27,28].…”

“…Often referred to as code clones, these segments of code typically involve 10-15% of the source code [24,25]. Code clones can arise through a number of different activities.…”

“…Thus, a variety of concerns such as stability, code ownership, and design clarity need to be considered before any refactoring is attempted; a manager should try to understand the reason behind the duplication before deciding what action (if any) to take. 1 This paper introduces eight cloning patterns that we have uncovered during case studies on large software systems, some of which we reported in [23,24,25]. These patterns present both good and bad motivations for cloning, and we discuss both the advantages and disadvantages of these patterns of cloning in terms of development and maintenance.…”