Discipline Matters: Refactoring of Preprocessor Directives in the #ifdef Hell

Overview

The preprocessor is used in almost all C projects to support variability and portability. However, many researchers and practitioners criticize the C preprocessor because of its negative effect on code understanding and maintainability, and its error proneness. More importantly, the use of the preprocessor hinders the development of tool support that is available in other languages, such as automated refactoring. Developers aggravate these problems when using the preprocessor in undisciplined ways (e.g., conditional blocks that do not align with the syntactic structure of the code). To address these problems, we proposed a catalog of refactorings to resolve undisciplined preprocessor usage in a previous study, and we offered an Eclipse plug-in that automatizes our refactorings. In this article, we evaluated our catalog regarding the number of application possibilities in practice, opinion of developers, and behavior preservation. Overall, we found 5670 application possibilities for our refactorings in 63 real-world projects. We performed an online survey among 202 developers, and we submitted 28 patches to convert undisciplined into disciplined directives. According to our results, most developers prefer to use the refactored (i.e., disciplined) version of the code instead of the original code with undisciplined preprocessor usage. To verify that our refactorings are indeed behavior preserving, we applied them to more than 36 thousand programs generated automatically using a model of a subset of the C language, and run the same test cases in the original and refactored programs. Furthermore, we applied the refactorings in three real-world projects: BusyBox, OpenSSL, and SQLite. This way, we detected and fixed a few behavioral changes introduced by our refactorings, the majority caused by unspecified behavior in the C language.

Number of Application Possibilities for our Refactorings

We investigate whether the undisciplined directives considered by our refactorings appear in real-world C projects. Our goal is to show that there are several possibilities to apply our refactorings in many real-world projects. To this end, we perform a large-scale study to learn whether developers can apply our refactorings in several places and in different projects. Please download the results here.

Model of the C Language

We create a model of a subset of the C language in Alloy to generate configurable programs with an opportunity to apply our refactorings. Download the complete C model in Alloy.

Patches Submitted to the Projects

To further understand the real-world relevance of our refactorings (RQ2), we submitted 28 patches to popular C projects. We want to know whether undisciplined directives are important enough to motivate developers to change code only to remove these directives. For selecting projects, we used SHTorrent,5 identifying active projects that heavily use pull requests on GitHub. We submitted patches to the 28 most active projects with application possibilities for our refactorings. Overall, developers accepted 21 (75%) patches, one each in: Angband; Amxmodx; Asfmapready; Collectd; Curl; Dmd; Libpng; Linux; Mapserver; Machinekit; Mongo; Opensc; Openssl; Opentx; Ossec-hids; Retroarch; Sleuthkit; Syslog-ng; Taulabs; Uwsgi; and Wiredtiger.

Contact