vprintf()

[Don Libes]

In article <924 at nlsun1.oracle.nl> bengsig at oracle.nl (Bjorn Engsig) writes:
>Doug:  You don't have to include <stdarg.h> to get vprintf() properly defined,
>       just <stdio.h>...
>Henry: Can you prove this?

>Isn't it wonderful for the rest of us that the guru's also disagree sometimes?
>What about Karl or Chris, don't you have a comment to this interesting discus-
>sion.  I would very much like to know if I need to include <stdarg.h> in all
>those sources where I only need to define vprintf() but don't call it :-)

This month's C Users Journal has an article by Plauger (X3J11 secretary)
which discusses this very point at length.  In brief, he agrees with
Doug, however the article goes into some detail about why this is an
important issue and some suggested implementation notes.

Since the standard was passed, Plauger has been writing interpretation
articles in CUJ that I've found fascinating and very well written.  For
instance, they include a lot of historical background, misperceptions,
caveats, implementation suggestions and examples.

Don Libes          libes at cme.nist.gov      ...!uunet!cme-durer!libes