On Jul 17, 2019, Martin Sebor <mse...@gmail.com> wrote: > Isn't this test sufficient to avoid the problems?
> if (!k && kmax > 1) > continue; It is, unless someone (i) doesn't realize attributes that are present in the type can't be present in the decl, (ii) misreads the '!k' as just 'k', and (iii) uses the wrong toolchain to confirm the consequences of the misreading ;-/ doh! > Can you put together a test case that does do the wrong thing? I'm afraid not, all of the additional errors I observed were correctly covered by the test I misunderstood once I grasped the actual logic. Sorry about the, erhm, false positive ;-) > The change looks cleaner than the cumbersome code that's there > now so I have no problem with it but I'm not sure it does more > than the test above Would you like me to put it in regardless? Does it make sense to put the testcase in anyway? -- Alexandre Oliva, freedom fighter he/him https://FSFLA.org/blogs/lxo Be the change, be Free! FSF Latin America board member GNU Toolchain Engineer Free Software Evangelist Hay que enGNUrecerse, pero sin perder la terGNUra jamás - Che GNUevara
