... resent, because message apparently bounced.
On 09/14/16 21:22, Bernd Edlinger wrote: > On 09/14/16 20:11, Jason Merrill wrote: >>> >>> Yes. The reasoning I initially had was that it is completely >>> pointless to have something of the form "if (x ? 1 : 2)" or >>> "if (x ? 0 : 0)" because the result does not even depend on x >>> in this case. But something like "if (x ? 4999 : 0)" looks >>> bogus but does at least not ignore x. >>> >>> If the false-positives are becoming too much of a problem here, >>> then I should of course revert to the previous heuristic again. >> >> I think we could have both, where the weaker form is part of -Wall and >> people can explicitly select the stronger form. >> > > > Yes, agreed. So here is what I would think will be the first version. > > It can later be extended to cover the more pedantic cases which > will not be enabled by -Wall. > > I would like to send a follow-up patch for the warning on > signed-integer shift left in boolean context, which I think > should also be good for Wall. > (I already had that feature in patch version 2 but that's meanwhile > outdated). > > > Bootstrap on x86_64-pc-linux-gnu and reg-testing not yet completed. > Is it OK for trunk when reg-testing finished? > > > Thanks > Bernd.
