Re: [xml] Regression Testing Questions - Starting over

[Daniel Veillard <veillard redhat com>]

On Wed, Nov 26, 2008 at 08:52:22AM -0800, Rush Manbert wrote:
> I now know that my previous problems were because I was trying to run  
> configure in a directory other than the distribution directory.

  ah, yes no chance that will work !

> I can now run all 4 of the test programs and all (?) of the tests that  
> make will run, and they all seem to pass. Some of the tests run by make 
> generate a lot of error output, but the make does not fail.
>
> So I guess my real question is how do I get the best test coverage?
>
> Right now I am doing the following:
> ./configure
> make
> make runtests
> make testall
>
> because that seems to run all of the available tests, even though it  
> also seems to have some overlap.
>
> Are runtests and testall the targets to use?

I would say use:  make check
I try to get as much coverage as possible that way.

there are overlaps between the various test commands, yes
see also 'configure --with-coverage' and 'make cov' if you
have a recent gcc, lcov and genhtml on your platform (works on
some linuxes)

Daniel

-- 
Daniel Veillard      | libxml Gnome XML XSLT toolkit  http://xmlsoft.org/
daniel veillard com  | Rpmfind RPM search engine http://rpmfind.net/
http://veillard.com/ | virtualization library  http://libvirt.org/