On Thu, Jan 3, 2013 at 10:34 PM, Lennart Poettering <lenn...@poettering.net> wrote: > On Sat, 29.12.12 02:10, Thomas H.P. Andersen (pho...@gmail.com) wrote: > >> Another item from the todo > > Heya! > > Hmm, so I commited a patch for using assert_static() a few days before > you did your patches but unfortunately never commited it. > > It works a bit differently from your patch, i.e. keeps assert_cc() as > special version of assert_static() around which only takes one rather > than two parameters and generates the message from the passed > expression. > > I am tempted to just leave this code now as I commited it, since having > to pass an explicit string to assert_static() is sometimes a bit > redundant i'd say, and doesn't really make things more readable unless > the expression that is tested is really complicated. > > Hence, I'd say that we should use assert_cc() and assert_static() from > now on like this: > > assert_cc() for simple expressions where the expression is readable > enough as is. > > assert_static() for complex expressions where it is worth specifying a > human readable string. > > I hope this makes sense?
yeah, makes sense. It also does not break the build for older compilers :) Do you see warnings about mixed declarations and code? I had to move some of the asserts to avoid that with my patch. _______________________________________________ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel