The shape of feature code: an analysis of twenty C-preprocessor-based systems

Abstract

Feature annotations (e.g., code fragments guarded by #ifdef C-preprocessor directives) control code extensions related to features. Feature annotations have long been said to be undesirable. When maintaining features that control many annotations, there is a high risk of ripple effects. Also, excessive use of feature annotations leads to code clutter, hinder program comprehension and harden maintenance.To prevent such problems, developers should monitor the use of feature annotations, for example, by setting acceptable thresholds. Interestingly, little is known about how to extract thresholds in practice, and which values are representative for feature-related metrics. To address this issue, we analyze the statistical distribution of three feature-related metrics collected from a corpus of 20 well-known and longlived C-preprocessor-based systems from different domains. We consider three metrics: scattering degree of feature conCommunicated by Prof. stants, tangling degree of feature expressions, and nesting depth of preprocessor annotations. Our findings show that feature scattering is highly skewed; in 14 systems (70 %), the scattering distributions match a power law, making averages and standard deviations unreliable limits. Regarding tangling and nesting, the values tend to follow a uniform distribution; although outliers exist, they have little impact on the mean, suggesting that central statistics measures are reliable thresholds for tangling and nesting. Following our findings, we then propose thresholds from our benchmark data, as a basis for further investigations.

A Study of Feature Scattering in the Linux Kernel

“…Scattered features may significantly increase system maintenance efforts [5,6]. Yet, feature scattering is common in practice [7][8][9].…”

“…Many long-lived and large-scale software systems have shown that it is possible to achieve continuous evolution while accepting some extent of feature scattering. Examples span different domains, such as operating systems, databases, and text editors [7,9,14].…”

Introduction to the theme issue on variability modeling of software-intensive systems

“…Queiroz, Passos, Valente, Hunsen, Apel and Czarnecki ("Empirical Assessment of Two Approaches for Specifying Software Product Lines Use Case Scenarios" [5]) report the results of an empirical investigation of two different specification approaches for software product line scenarios, namely PLUSS and MSVCM. These two techniques are similar with regard to their inputs but different in respect of the compositional (MSVCM) and invasive (PLUSS) style for variability management.…”

On the Relationship Between Feature Dependencies and Change Propagation

“…In our thesis, we identify direct properties of feature dependencies to draw our conclusions. Finally, in a recent work, Queiroz et al quantify three metrics in twenty preprocessor-based systems and define thresholds to the context of feature maintenance, specifically for ripple effects observed in the evolving source code [QUEIROZ et al 2015]. However, once again, this study adapts metrics to the context of preprocessor-based systems and they neglect the impact of dependencies on code maintenance and change propagation.…”

“…Consequently, they lack of standard terminology and formalism, leading to definitions of measures, which are ambiguous and/or difficult to understand. A growing number of studies have identified structural properties of product lines [SOBERNIG et al 2014, QUEIROZ et al 2015 and metrics for product-line [ VAN DER HOEK et al 2003, LIEBIG et al 2010, APEL and BEYER 2011 to overcome the problems aforementioned. Their common goal is the association between: (i) the quantification of characteristics governing either product lines or programming techniques, and (ii) their impact on software maintainability.…”