Date: Tue, 31 Jul 2001 18:34:31 -0700 (PDT) From: Matt Dillon <dillon@earth.backplane.com> To: John Baldwin <jhb@FreeBSD.ORG> Cc: "David O'Brien" <obrien@FreeBSD.ORG>, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, Garrett Wollman <wollman@khavrinen.lcs.mit.edu> Subject: Re: cvs commit: src/usr.sbin/newsyslog Makefile newsyslog.c Message-ID: <200108010134.f711YVt21651@earth.backplane.com> References: <XFMail.010731084518.jhb@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
:s = "-4", x = -4 : :Granted, when x is unsigned, you get: : :s = "-4", x = 4294967292 : :Which will be large enough for the cases assumed here. : :Does go to show that strtol() wouldn't fail in the negative case however as I :thought. Hmm. This means that you probably could just have done :s/strtoul/strtol/ to get rid of the warning. :) : :-- : :John Baldwin <jhb@FreeBSD.org> -- http://www.FreeBSD.org/~jhb/ strtol() will fail for any value > 2^31, as documented: #include <stdio.h> int main(int ac, char **av) { int x; x = strtol("0x80000000", NULL, 0); printf("%08x\n", x); return(0); } earth:/home/dillon> ./x 7fffffff This has an unfortunate result that is far worse then strtoul()'s unfortunate handling of '-' for -0x80000001 through -0xFFFFFFFF. So, in general, I would recommend using strtoul() exclusively and then casting the result to a signed quantity if you wanted a signed result. Personally I prefer to old assembly 'convert the string to an int and just ignore the overflow' method. strtol() has !@#$%@!#$ me up too many times. -Matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200108010134.f711YVt21651>