Empirical analysis on the correlation between GCC compiler warnings and revision numbers of source files in five industrial software projects

This article discusses whether using warnings generated by the GNU C++ compiler can be used effectively to identify source code files that are likely to be error prone. We analyze five industrial projects written in C++ and belonging to the telecommunication domain. We find a significant positive co...

Full description

Saved in:
Bibliographic Details
Published inEmpirical software engineering : an international journal Vol. 12; no. 3; pp. 295 - 310
Main Authors Moser, Raimund, Russo, Barbara, Succi, Giancarlo
Format Journal Article
LanguageEnglish
Published Dordrecht Springer Nature B.V 01.06.2007
Subjects
C++ (programming language)
Compilers
Empirical analysis
Error analysis
Software
Source code
Online AccessGet full text

Cover

More Information
Summary:This article discusses whether using warnings generated by the GNU C++ compiler can be used effectively to identify source code files that are likely to be error prone. We analyze five industrial projects written in C++ and belonging to the telecommunication domain. We find a significant positive correlation between the number of compiler warnings and the number of source files changes. We use such correlation to conclude that compiler warnings may be used as an indicator for the presence of software defects in source code. The result of this research is useful for finding defect-prone modules in newer projects, which lack change history.
Bibliography:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 23
ISSN:1382-3256
1573-7616
DOI:10.1007/s10664-006-9029-x