Re: __VA_ARGS__ patch for cccp.c

[Marc Lehmann <pcg at goof dot com>]

On Mon, Mar 09, 1998 at 06:34:04PM -0800, Jim Wilson wrote:
> We really should have a way of specifying the c9x standard before adding
> features that are specific to it.  I'll bring that up in another message.

yes, please, that's my bigest concern. there was quite some discussion,
but no result...

> This looks mostly OK.  However, I don't see anything in the C9X draft about
> using ## to delete a preceeding argument if the ... part is empty.  This stuff

Yes...

#define msg(frmt,...)

just doesn't work (as in msg("hi")), you have to use

#define msg(...)

and hope that functions you are calling happen to have the right argument
order. bad bad.

, ## etc.. is really missing from c9x.

> should not be enabled if using the c9x style variable argument macros.

I don't agree. gcc should be able to grok iso-c _by default_,
if possible every version of it. selecting between different standards
imho only makes sense when:

- two standards (or versions of it) disagree on the same syntax.
- you want strict conformance and use gcc as a tool to warn
  about possible (conformance) problems (like -pedantic should do).

I agree that the patch behavious should be selectable by (which??)
switch. It's time to agree on one.

[ varargs enabled with -pedantic}
> That looks like a bug to me.  This patch seems to fix it.

fine! I fell into that trap a week ago ;)

      -----==-                                              |
      ----==-- _                                            |
      ---==---(_)__  __ ____  __       Marc Lehmann       +--
      --==---/ / _ \/ // /\ \/ /       pcg@goof.com       |e|
      -=====/_/_//_/\_,_/ /_/\_\                          --+
    The choice of a GNU generation                        |
                                                          |