🚀 go-pugleaf

In article <33kfu1$eff@usenet.pa.dec.com> diamond@jrd.dec.com (Norman Diamond) writes:
>In article <Cv4GDy.GMr@world.std.com> jhallen@world.std.com (Joseph H Allen) writes:
>>unsigned short a=1024;
>>unsigned short b=1024;
>>int c=311;
>>char *d="Hello there";
>>main() { printf("%x %x\n",d+c,d+a-(b-c)); }
>>
>>The fix for JOE is to change the two 'unsigned short's in the file b.h to
>>'int's.  This is truly a bizzare problem.

>The correct fix is to change the two '%x' conversions to '%hx' conversions.
>%x requires an unsigned int argument; %hx requires an unsigned short argument.

Sorry, I had noticed the "unsigned short" declarations and had not noticed
the "char *" declarations.  Someone pointed out by e-mail that these are
pointer expressions.  The correct fix is to use %p format (and pedantically,
to cast the pointer expressions to type (void *), but it is "intended" to
work without it because char * and void * have the same representation).
However, %hx or %x or %lx might work on some machines if you know the
sizes of the corresponding integral types and of char *.

[Notice to jik:  Sorry the amount of source here is only a few characters,
changing x or hx to p.  It is still source.]
--
 <<  If this were the company's opinion, I would not be allowed to post it.  >>
segmentation fault (california dumped)