Re: [PHP4BETA] Re: eval() bug?

On Fri, Jun 11, 1999 at 07:37:18PM +0300, Zeev Suraski wrote: > Really? Hmm, that's weird. I never protect against NULL's in printf()'s, > just about any implementation I saw prints (null), and doesn't crash. > > That's kind of bad. The problem is that the C standard lays down that the argument for a %s specifier shall be a pointer to an array of characters. Since NULL is not such a pointer (it's a pointer, but not to an array of characters), the call falls into "undefined behaviour." The question here is, what do you generally want to do with NULL pointers passed to library functions. If you insist that *printf should accept a NULL pointer as if it were valid, how should other functions such as strcmp behave in such a case? Sun chooses the way of everything or nothing. If your program conforms to the ANSI C standard, everything will work as expected. Once a programmer uses a construct with undefined behaviour, the Sun libc segfaults. We observed this already with a non-deterministic compare function passed to libc's qsort(). More information about this topic can be found in "Expert C Programming" by Peter v.d. Linden, a member of the compiler and OS kernel group at Sun. He details questions such as "why extern char *cp isn't equal to extern char cp[]" and other important things you won't find in common C books. -- Regards, Sascha Schumann Consultant