On Saturday 20 November 2010, Ralf Wildenhues wrote: > * Stefano Lattarini wrote on Fri, Nov 19, 2010 at 09:18:33PM CET: > > I've realized that my patch on instspc.test split & refactoring has > > broken the "release-stats" target, since now there are other generated > > tests besides the `*-p.test' tests. > > > > The attached patch fix this glitch in a "quick & dirty" way; a better > > fix would probably involve a refactoring of the whole "release-stats" > > target, but that can be done later IMHO. > > > > So, OK to apply the attached patch to a temporary branch based off > > of commit `v1.11-395-ge118126' "Overhauled and modularized tests in > > `instspc.test'.", and merge to master? > > This will break again when we add the next set of generated tests. > True. That's why in the refactoring I have in mind we would define a new variable `$(generated_tests)' in tests/Makefile.am, and fetch its value from top-level Makefile.
> How about determining them by grepping the test source for 'GENERATED > AUTOMATICALLY' in the test source? > That might be a bit fragile too in the long run, but it's definitely better than my hack in the short run. So I'll make the change. Thanks, Stefano