Path: csiph.com!eternal-september.org!feeder.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail From: Keith Thompson Newsgroups: comp.lang.c Subject: Re: usecases of void foo(char **b) {} Date: Wed, 16 Nov 2016 09:49:26 -0800 Organization: None to speak of Lines: 36 Message-ID: References: <89e9cc99-3b51-44f1-91bc-f74b6f2b7f5a@googlegroups.com> <34d35fa8-fe94-4665-b0e1-21de4f507fe4@googlegroups.com> <52475847-d45a-4c24-9661-1ebe869759b3@googlegroups.com> <6c6a30d0-94cd-42a6-884a-2ac99995e681@googlegroups.com> <62d12ca1-490e-447d-b867-69b1458517be@googlegroups.com> <939b0c10-54c9-408d-89c4-34868c3f63af@googlegroups.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Injection-Info: mx02.eternal-september.org; posting-host="4ec4d910e7b984fdf1507bf694506dd1"; logging-data="18752"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX1+RgWaP2XiEXCwh81mz5qyQ" User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.1 (gnu/linux) Cancel-Lock: sha1:EBDxRd5pf+v6WdbKVat4vhKq6uY= sha1:LsOV5rxeJl+ytInKyt7krqV13fw= Xref: csiph.com comp.lang.c:96706 supercat@casperkitty.com writes: > On Wednesday, November 16, 2016 at 10:37:19 AM UTC-6, Keith Thompson wrote: [...] >> Surely the phrase "not defined by the language" is clear enough. Apparently not. > The language is usable on platforms which implement silent wraparound in > hardware and also on platforms that don't. Further, even on silent- > wraparound hardware, some implementations had documented options to trap > on overflow. > > Most implementations, however, ran on silent-wraparound hardware and had > nothing in their code-generation logic which would cause them to deviate > from such semantics, but didn't bother to explicitly document the silent- > wraparound behavior that resulted. > > What would you think Ritchie would have said if he merely wanted to indicate > that implementations which run on non-silent-wraparound platforms or document > other behaviors for overflow are not required to have silent-wraparound > semantics, but *didn't* want to suggest that such semantics should not be > expected on commonplace two's-complement platforms whose documentation does > not say anything about integer overflow? I don't know, probably something other than "not defined by the language". More specifically, I'm fairly sure he would have used shorter sentences to say whatever it is you imagine he might have said in some alternate universe. In any case, I'm not interested in helping you put words in Dennis Ritchie's mouth. -- Keith Thompson (The_Other_Keith) kst-u@mib.org Working, but not speaking, for JetHead Development, Inc. "We must do something. This is something. Therefore, we must do this." -- Antony Jay and Jonathan Lynn, "Yes Minister"