SZEDER Gábor <szeder.dev@xxxxxxxxx> writes: >> > Let parse_options() handle unknown options instead, which, besides >> > simpler code, has the additional benefit that it prints not only the >> > usage but an "error: unknown option `foo'" message as well. >> >> Yes, I agree that getting rid of KEEP_UNKNOWN is a very good idea >> for this reason. But I suspect that we still need the "did we get >> an extra argument we do not know what to do with?" check. > > Uh, indeed. I got too trigger-happy with deleting lines. > Updated patch below. OK. I suspect that a test would have caught the breakage in the original. Would it make sense to add one now? Thanks, will queue.
