On Tuesday 01 March 2011, Peter Rosin wrote: > Den 2011-03-01 10:40 skrev Stefano Lattarini: > > On Tuesday 01 March 2011, Peter Rosin wrote: > >> Good, I pushed with that squashed in. I didn't dare a merge to > >> master as the conflicts looked scary. > >> > > I've taken a look and done the merge. Luckily, the conflicts were > > in fact strictly spurious (but yes, they truly look scary if you're > > not intimate with the latest changes in tests/defs, which indeed > > entailed quite a bit of code moving; sorry about that). > > Thanks! I managed to guess that it probably just looked scary, but > the thing that held me back was the fact that the new ltinit.test > was written for maint, and I didn't know what parts of it needed > adjustment for master. > *Ideally*, no one should. Every test that works for maint should also work for master without modifications (unless, of course, it tests an automake behaviour that has been modified in master).
> I did spot the obsolete set -e but didn't > know what else I was missing (probably nothing, but I wasn't > sure...). Anyway, should I push the below or is the plan to > clean up all new 'set -e' fallouts "later"? > That was my intention, yes. But if you want to clean up all the existing 'set -e' fallouts, that would be nice and well-received (and in fact, "git grep '^set -e$' master:tests" shows there is a fair numer of such fallouts already). Otherwise, I can still do that (at a later time). Thanks, Stefano
