Michael Daniels wrote: > > 1) It very much looks like it has not been tested. > > Dammit. I did test a very similar patch (differing only in the placement of > &&), > then changed the placement of the && to conform to the GNU coding standards. > And I didn't test because surely that tiny change can't matter.
Yeah, I can relate to that experience :) I've made the same mistake (tweaking after testing) many times too. It isn't obvious that one needs backslashes on the line before '&&'. The better habit is "tweak then test", rather "test then tweak". But yes, often it turns into "tweak then test then tweak then test again"... Bruno
