Re: [Update REQ 5397]: hardy mystery

[stuart at sfb288.math.tu-berlin.de]

Just for the record, I don't think the HP's behavior on
   fprintf(stdout, "...", p, *p++, *p++, p);
and such stuff should be considered a bug -- C doesn't guarantee any
particular order of evaluation of parameters of a function,
neither left-to-right nor right-to-left, so it's not safe to mention
both p and *p++ in the same breath, let alone two *p++'s.  To get well-defined
behavior you'd probably want to write something like
   fprintf(stdout, "...", (int)p);
   fprintf(stdout, "...", *p++);
   fprintf(stdout, "...", *p++);
and so on, or else write something of the form
   fprintf(stdout, "...", (int)p, p[0], p[1], p[2], (int)(p+3));
   p += 3;

- - Stuart