Hi Paul (delayed answer, bugger of a month)



 RM>  char *s = "1245 2330 0400";

 RM>  int l,m,n;

 RM>  sscanf(s, "%i %i %i", &l,&m,&n);



 PE> I've never used %i before, I always use %d, and I was pretty sure

 PE> %d always treated the things as decimal.  However, I was under the

 PE> impression that %i was the same as %d.



Interesting. I've never used %d before. I started with Turbo C 1.0,

and afai remember, %d barely got a mention there. I've used %i

exclusively since then, and have just found out why %d was

invented...



I found my error the next day, btw. Just had to rtfm. The exact

nature of the problem eluded me for a while, cos I was actually

using something that looked like



 char *s = "0950 1245 2330";

 int l,m,n;

 sscanf(s, "%i %X %i", &l,&m,&n); /* the %X is
needed, trust me */



which produced l=0, m=950hex, n=1245. Infuriating. It was late, I

was tired, it was time to do mail, so I took the lazy way out and

asked here...



 PE> So, resorting to the ISO

 PE> standard (did you know you can buy this for about $70, and it is

 PE> an EXCELLENT reference, a must for every C programmer)



how long before they print a new one? I'm not blowing seventy bucks

only to find a new version has just appeared and I can burn the old

one. And I'd need to get another one for C++, too (there seems to

be an ANSI standard for C++ at last)



 PE> Use d instead of i and she'll be hunky dory.



Done. All hunkys are dorys.



 PE> BTW, my ISO C standard is no longer in the other room.



Propping up a crooked table, perhaps?



Cheers, and ta (tks also to Nhan and David)



--- PPoint 1.88