In order to improve the speed of the analysis, cov-run-desktop limits
the number of files or translation units considered. In this context, a translation unit
refers to any primary source file from the intermediate directory, along with any files
included by that primary source file. The selection process is defined using various
options that affect what code will be analyzed.
There are two methods to select which translation units will be analyzed, and it is required that the user specify one. The first method is to explicitly pass a list of source files to be analyzed at the end of the command line:
cov-run-desktop [OPTIONS] FILE1 [FILE2 [...]]
For C and C++, the files will typically be .c or .cpp source files. If you want to analyze a header file, see Coverity Desktop Analysis: User Guide.
This can also be accomplished by listing your chosen source files in a response file, and then
specifying it on the command line with the @@<response_file> syntax.
--analyze-scm-modified option. See
--analyze-scm-modified
and A note on Git superprojects
for more information on this method.