[CMake] Re: 'make test' fails on x86/Solaris

[Alan W. Irwin]

On 2007-03-16 08:20-0400 Bill Hoffman wrote:

> I am not sure everyone would be happy if make test depend on all.   I don't 
> think that the typical unix make check will build anything will it?

I don't know about other projects, but all the three active projects
(PLplot, libLASi, and FreeEOS) I am working on test using example code that
must be built against the project libraries.  So, yes, make must be run (to
build the project libraries) before ctest (or the equivalent "make test")
can be run.  I suspect this must be true of most projects that create
libraries.  Unless the tests are weak, they will involve the libraries
created by the project.

> The idea 
> was to have the standard:
>
> make
> make test
> make install
>
> I guess the downside would be extra depend checking when running tests on 
> large projects.

I have just proved by example that you can run "make install" right after
"cmake" in an initially empty build directory.  In other words, the
"install" target depends on the "all" target.  So assuming most projects
need the build in order to do the testing (which is my experience for my
library-oriented projects) I think the "test" target should be treated the
same way.

Alan
__________________________
Alan W. Irwin

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state implementation
for stellar interiors (freeeos.sf.net); PLplot scientific plotting software
package (plplot.org); the Yorick front-end to PLplot (yplot.sf.net); the
Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project
(lbproject.sf.net).
__________________________

Linux-powered Science
__________________________