C++Talk.NET C++ language newsgroups   Archives   FAQ   Search   Memberlist   Usergroups   Register   Profile   Log in to check your private messages   Log in  why nan and infs evaluate to true when used as if condition          C++Talk.NET Forum Index -> C Language (Moderated) View previous topic :: View next topic   Author Message Ganny Guest Posted: Tue May 15, 2007 9:06 pm    Post subject: why nan and infs evaluate to true when used as if condition #include <stdio.h> int main() { int a = 0; if (1.0/0.0) printf("result was true"); else printf("result was false"); } I get "result was true" in the compilers i tried - Is this the right behaviour expected? shouldn't nans and infs treated as false? thanks! -ganesh -- comp.lang.c.moderated - moderation address: clcm (AT) plethora (DOT) net -- you must have an appropriate newsgroups line in your header for your mail to be seen, or the newsgroup name in square brackets in the subject line. Sorry. Back to top Francis Glassborow Guest Posted: Thu May 17, 2007 9:11 am    Post subject: Re: why nan and infs evaluate to true when used as if condit Ganny wrote: Quote: #include <stdio.h int main() { int a = 0; if (1.0/0.0) printf("result was true"); else printf("result was false"); } I get "result was true" in the compilers i tried - Is this the right behaviour expected? shouldn't nans and infs treated as false? thanks! -ganesh Yes, but did the implementations you tested have NaNs and INFs? As it stands it is undefined behaviour and so the program can generate any output it likes not just the alternatives you provided. -- comp.lang.c.moderated - moderation address: clcm (AT) plethora (DOT) net -- you must have an appropriate newsgroups line in your header for your mail to be seen, or the newsgroup name in square brackets in the subject line. Sorry. Back to top Douglas A. Gwyn Guest Posted: Thu May 17, 2007 9:11 am    Post subject: Re: why nan and infs evaluate to true when used as if condit Ganny wrote: Quote: if (1.0/0.0) printf("result was true"); I get "result was true" in the compilers i tried - Is this the right behaviour expected? shouldn't nans and infs treated as false? NaNs and Infs aren't necessarily supported on all platforms; dividing by 0.0 has undefined behavior in general. Anyway, any not-precisely-zero value is treated as "true" in C conditional contexts. -- comp.lang.c.moderated - moderation address: clcm (AT) plethora (DOT) net -- you must have an appropriate newsgroups line in your header for your mail to be seen, or the newsgroup name in square brackets in the subject line. Sorry. Back to top Jack Klein Guest Posted: Thu May 17, 2007 9:11 am    Post subject: Re: why nan and infs evaluate to true when used as if condit On 15 May 2007 16:06:41 GMT, Ganny <sgganesh (AT) gmail (DOT) com> wrote in comp.lang.c.moderated: Quote: #include <stdio.h int main() { int a = 0; if (1.0/0.0) Since the line above produces undefined behavior, the C standard has nothing at all to say about what happens afterward. Quote: printf("result was true"); else printf("result was false"); } I get "result was true" in the compilers i tried - Is this the right behaviour expected? shouldn't nans and infs treated as false? An if() test of an arithmetic value is an explicit comparison with 0. I would expect a floating point infinity to compare unequal with 0, since it means a value to large in magnitude to be represented. Mathematically, it is as far from 0 as it is possible to be for a value of that type. A NaN cannot compare equal to 0 because a NaN is required not to compare equal to anything, even itself. -- Jack Klein Home: http://JK-Technology.Com FAQs for comp.lang.c http://c-faq.com/ comp.lang.c++ http://www.parashift.com/c++-faq-lite/ alt.comp.lang.learn.c-c++ http://www.club.cc.cmu.edu/~ajo/docs/FAQ-acllc.html -- comp.lang.c.moderated - moderation address: clcm (AT) plethora (DOT) net -- you must have an appropriate newsgroups line in your header for your mail to be seen, or the newsgroup name in square brackets in the subject line. Sorry. Back to top Guest Posted: Thu May 17, 2007 9:11 am    Post subject: Re: why nan and infs evaluate to true when used as if condit Ganny <sgganesh (AT) gmail (DOT) com> wrote: Quote: shouldn't nans and infs treated as false? No -- there's an implicit comparison to 0. Neither Inf nor NaN compares equal to zero, so both are treated as true. -Larry Jones I don't need to do a better job. I need better P.R. on the job I DO. -- Calvin -- comp.lang.c.moderated - moderation address: clcm (AT) plethora (DOT) net -- you must have an appropriate newsgroups line in your header for your mail to be seen, or the newsgroup name in square brackets in the subject line. Sorry. Back to top Hans-Bernhard Bröker Guest Posted: Thu May 17, 2007 9:11 am    Post subject: Re: why nan and infs evaluate to true when used as if condit Ganny wrote: Quote: #include <stdio.h int main() { int a = 0; if (1.0/0.0) printf("result was true"); else printf("result was false"); } I get "result was true" in the compilers i tried - Is this the right behaviour expected? It is one of many right behaviours. The code causes undefined behaviour. Which means there's strictly no way it can behave incorrectly. Quote: shouldn't nans and infs treated as false? What makes you think that? Only zero is defined to be treated as false? Why should NaN or Inf be considered zero? -- comp.lang.c.moderated - moderation address: clcm (AT) plethora (DOT) net -- you must have an appropriate newsgroups line in your header for your mail to be seen, or the newsgroup name in square brackets in the subject line. Sorry. Back to top jg Guest Posted: Thu May 17, 2007 9:11 am    Post subject: Re: why nan and infs evaluate to true when used as if condit On May 15, 9:06 am, Ganny <sggan...@gmail.com> wrote: Quote: #include <stdio.h int main() { int a = 0; if (1.0/0.0) printf("result was true"); else printf("result was false"); } I get "result was true" in the compilers i tried - Is this the right behaviour expected? shouldn't nans and infs treated as false? I think you condition is to compare NaN with zero. Since the representation of NaN isn't zero, your condition is true. I don't see any relation between false/true and Nan. Why would you think NaN and Infs should be false ? JG -- comp.lang.c.moderated - moderation address: clcm (AT) plethora (DOT) net -- you must have an appropriate newsgroups line in your header for your mail to be seen, or the newsgroup name in square brackets in the subject line. Sorry. Back to top Michael Tiomkin Guest Posted: Thu May 17, 2007 9:11 am    Post subject: Re: why nan and infs evaluate to true when used as if condit On May 15, 6:06 pm, Ganny <sggan...@gmail.com> wrote: Quote: #include <stdio.h int main() { int a = 0; if (1.0/0.0) printf("result was true"); else printf("result was false"); } I get "result was true" in the compilers i tried - Is this the right behaviour expected? shouldn't nans and infs treated as false? First, as far as I remember, only NaNs should return false after any comparison, and second, this happens only with IEEE compatible floating point implementation. 1.0/0.0 will usually give you an infinity value, with results of comparisons different from those of NaNs, e.g. inf!=0 is true. -- comp.lang.c.moderated - moderation address: clcm (AT) plethora (DOT) net -- you must have an appropriate newsgroups line in your header for your mail to be seen, or the newsgroup name in square brackets in the subject line. Sorry. Back to top John Dallman Guest Posted: Thu May 17, 2007 9:11 am    Post subject: Re: why nan and infs evaluate to true when used as if condit In article <clcm-20070515-0006 (AT) plethora (DOT) net>, sgganesh (AT) gmail (DOT) com (Ganny) wrote: Quote: #include <stdio.h int main() { int a = 0; if (1.0/0.0) printf("result was true"); else printf("result was false"); } I get "result was true" in the compilers i tried - Is this the right behaviour expected? shouldn't nans and infs treated as false? Relying on this is a bad idea to start with, but since the general definition of "what is true" in C is something rather like "not equal to zero", it seems hard to say your compilers are wrong. Whatever infinities and NaNs are, they aren't equal to zero. -- John Dallman, jgd (AT) cix (DOT) co.uk, HTML mail is treated as probable spam. -- comp.lang.c.moderated - moderation address: clcm (AT) plethora (DOT) net -- you must have an appropriate newsgroups line in your header for your mail to be seen, or the newsgroup name in square brackets in the subject line. Sorry. Back to top Keith Thompson Guest Posted: Thu May 17, 2007 9:11 am    Post subject: Re: why nan and infs evaluate to true when used as if condit Ganny <sgganesh (AT) gmail (DOT) com> writes: Quote: #include <stdio.h int main() { int a = 0; if (1.0/0.0) printf("result was true"); else printf("result was false"); } I get "result was true" in the compilers i tried - Is this the right behaviour expected? shouldn't nans and infs treated as false? Anything that compares equal to zero is false; anything else is true. In any case, your program invokes undefined behavior. The C standard doesn't require 1.0/0.0 to yield Infinity or NaN. It could yield some real value, or terminate your program, or make demons fly out of your nose. -- Keith Thompson (The_Other_Keith) kst-u (AT) mib (DOT) org <http://www.ghoti.net/~kst> San Diego Supercomputer Center <*> <http://users.sdsc.edu/~kst> "We must do something. This is something. Therefore, we must do this." -- Antony Jay and Jonathan Lynn, "Yes Minister" -- comp.lang.c.moderated - moderation address: clcm (AT) plethora (DOT) net -- you must have an appropriate newsgroups line in your header for your mail to be seen, or the newsgroup name in square brackets in the subject line. Sorry. Back to top WillerZ Guest Posted: Thu May 17, 2007 9:11 am    Post subject: Re: why nan and infs evaluate to true when used as if condit Ganny wrote: Quote: #include <stdio.h int main() { int a = 0; if (1.0/0.0) printf("result was true"); else printf("result was false"); } I get "result was true" in the compilers i tried - Is this the right behaviour expected? shouldn't nans and infs treated as false? if (exp) statement1 else statement2 is defined to execute statement1 iff exp is not equal to zero. Is inf or nan equal to zero? No. -- comp.lang.c.moderated - moderation address: clcm (AT) plethora (DOT) net -- you must have an appropriate newsgroups line in your header for your mail to be seen, or the newsgroup name in square brackets in the subject line. Sorry. Back to top Display posts from previous: All Posts1 Day7 Days2 Weeks1 Month3 Months6 Months1 Year Oldest FirstNewest First         C++Talk.NET Forum Index -> C Language (Moderated) All times are GMT Page 1 of 1   Jump to: Select a forum C/C++----------------C LanguageC Language (Moderated)C++ Language (Moderated)C++ language (comp.lang.c++)C++ language, library and standards Non-English C/C++ Newsgroups----------------C++ (German)C++ (French)  You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum Powered by phpBB © 2001, 2006 phpBB Group SEO toolkit © 2004-2006 webmedic.