On Tuesday 15 February 2011, Ralf Wildenhues wrote:
> * Stefano Lattarini wrote on Tue, Feb 15, 2011 at 10:21:36AM CET:
> > On Monday 14 February 2011, Ralf Wildenhues wrote:
> > > > --- /dev/null
> > > > +++ b/tests/instspc-data.test
> > >
> > > > +# Helper testcase which generate input data for the other test
> > > > +# `instspc-*.test'. It basically delegates the work to the helper
> > > > +# script `instspc-test.sh'.
> > >
> > > As an alternative to a helper testcase, this could also just be a helper
> > > script whose run is a prerequisite to the instspc*.log files. That way
> > > you don't have a bogus test result.
>
> > I had already tried a similar approach in the first version of the patch:
> > <http://lists.gnu.org/archive/html/automake-patches/2010-11/msg00152.html>
> > but IMHO that turned out to be slightly more fragile and more complex than
> > the current approach. So I'd rather not go back there.
>
> It is not immediately obvious why the approach was more fragile; care to
> elaborate on that? Asking just out of curiosity.
>
Well, basically I wanted to rely on the code in `tests/defs' (in order to
avoid complications and duplications), but since that code is really meant
only to be run inside a testcase, it wasn't obvious whether I could safely
use it in a helper script. Making that helper script a fully-fledged
testcase helped with this. Also, I find it simpler to have a simple
dependency:
$(instspc_tests:.test=.log): instspc-tests.sh instspc-data.log
rather than a dependency + hand-written rule:
$(instspc_tests:.test=.log): instspc-tests.sh instspc-data.dir/.dirstamp
instspc-data.dir/.dirstamp:
srcdir=$(srcdir) $(SHELL) $(srcdir)/tests/instspc-setup
This are small things, agreed, and that's why I said "... *slightly* more
fragile and more complex than the current approach".
> Sorry for overlooking that you had already tried this.
>
> Cheers,
> Ralf
>
Regards,
Stefano
