On Tue, 2011-07-26 at 11:38 +0200, Patrick Ohly wrote:
On Di, 2011-07-26 at 11:28 +0200, Murray Cumming wrote:
> On Tue, 2011-07-26 at 11:22 +0200, Patrick Ohly wrote:
> > On Di, 2011-07-26 at 11:08 +0200, Murray Cumming wrote:
> > > The test/dbus-session script doesn't seem to be used by anything. Is
it
> > > used by some separate testing script or meant to be used only manually?
> > >
> > >
https://meego.gitorious.org/meego-middleware/syncevolution/blobs/master/t...
> >
> > It's still used in the nightly testing. Normal users probably don't
need
> > it much anymore since --start-dbus-session was added to
> > syncevo-dbus-server.
>
> So there is a nightly testing script somewhere that uses it? Is that in
> git anywhere?
test/runtests.py is the main script. We have wrapper scripts that are
not public, partly because they contain confidential information like
the credentials on the various SyncML servers, partly because it's not
applicable to other machines.
Here's one example run which shows how dbus-session.sh is used inside a
chroot:
http://runtests.syncevolution.org/2011-07-25-21-30_all/head-lucid-amd64/5...
Thanks.
While we are on the subject, is there any particular reason that, when
--enable-unit-tests is used, the tests are installed?
Presumably the automated testing has to build syncevolution from source
anyway, so can't it just run the tests in the build directory? That
would also avoid any worry about installation of previous versions
affecting later versions.
That would make it easier to
1. make the tests run during "make check" only, which is more correct.
2. Always run those tests against only the locally-built syncevolution,
in a private D-Bus session. That makes the testing far easier to do by
more people, more reliably.
2. Remove the --enable-unit-tests configure option, so that that main
build does not include the unit-tests code, allowing autotools to
create a separate build with ENABLE_UNIT_TESTS defined, which the builds
will link against. That will almost double the build time, but it seems
correct.
--
murrayc(a)murrayc.com
www.murrayc.com
www.openismus.com