Feature Location in a Collection of Product Variants

Abstract

Companies often develop and maintain a collection of product variants that share some common features but also support different, customer-specific features. To reengineering such legacy product variants for systematic reuse, one must identify features and their implementing code units (e.g. functions, files) in different product variants. Information retrieval (IR) techniques may be applied for that purpose. In this paper, we discuss problems that hinder direct application of IR techniques to a collection of product variants. To counter these problems, we present an approach to support effective feature location in product variants. The novelty of our approach is that we exploit commonalities and differences of product variants by software differencing and FCA techniques so that IR technique can achieve satisfactory results for feature location in product variants. We have implemented our approach and conducted evaluation with a collection of nine Linux kernel product variants. Our evaluation shows that our approach always significantly outperforms a direct application of IR technique in the subject product variants.

Variability extraction and modeling for product variants

“…Xue et al use diffing algorithms to identify the common and variable parts of product variants, which are subsequently partitioned using Formal Concept Analysis [38]. To these partitions, Information Retrieval algorithms are applied to identify the code units specific to a feature.…”

Towards a language-independent approach for reverse-engineering of software product lines

“…However, this work only focus on the extraction of variability from the source code with conditional compilation directives. Xi et al [23] propose an approach based on Formal Analysis Concept (FCA) for the identification of code units that are associated with a set of existing features. Indeed, in addition to the source code of product variants, this approach also takes as input data the list of features associated to each product variant.…”

A large scale Linux-Kernel based benchmark for feature location research

“…We then propose an approach to identify common features and their implementing code units across product variants by effectively incorporates software differencing, formal concept analysis, and IR techniques [11]. We evaluate the proposed approach with the Linux kernel benchmark.…”

“…To date, we have conducted a few studies using this benchmark, including a large-scale comparative study of the effectiveness of 10 IR techniques for feature location [10], and two new approaches to feature location [5] [11]. In addition to our own research, we believe this benchmark will also have a strong positive effect on the research community, because it will bring opportunities of communication and collaboration among different researchers, leading to clear research goal and development of feature location methods.…”