PE> That is not true.  This program is strictly conforming...
PE> void doprint(void);
PE> Despite the fact that doprint() is not defined by ISO.  BFN.  Paul.

TS> What do you want to tell me? Of course you may define functions as you 
TS> like, and "doprint" doesn't fall into the namespace of
system or compiler.

It was in response to the below...

ML>> standard, non-ANSI library functions have an underscore as the leading
ML>> symbol in the function name.

PE> There is no requirement for this, unless they interfere with the

TS> One is free to use functions named as one likes but the resulting programs 
TS> are not longer allowed to be called (strictly) conforming :-)

PE> That is not true.  This program is strictly conforming...


My point is that there is absolutely nothing wrong with Borland
calling their extensions things like cprintf(), despite the fact
that there's no underscore.  You don't want them, don't include
them in.  Borland's cprintf() is no different from my doprint()
[or it would be if their cprintf() was written in pure ISO C].

Maybe I misunderstood the point you were trying to make above.
BFN.  Paul.
@EOT:

---