Re: Warnings for excessive macro arguments

[Eric Blake]

On 08/23/2010 03:45 PM, Ralf Wildenhues wrote:
> * Yavor Doganov wrote on Mon, Aug 23, 2010 at 09:42:54PM CEST:
> > Eric Blake wrote:
> > > I have to wonder if there are cases where rejecting extra arguments
> > > might not be advisable,
> >
> > No, the behavior should remain the same (silently ignoring them); any
> > change smells like a dangerous path to follow.  The only difference
> > would be the warnings (well, errors with -Werror, unfortunately).
>
> The problem with this approach is that we then cannot ever extend a
> macro by adding new detected arguments, without also breaking backward
> compatibility that way.  That means, each feature addition requiring
> more arguments would now also require a new macro name, wasting more
> good names in the process.  :-/

Requiring a minimum number of arguments is problematic - existing code 
cannot be compiled with a newer autoconf that requires a larger minimum. 
 That is, we intentionally document that AC_RUN_IFELSE([arg1]) and 
AC_RUN_IFELSE([arg1], [], [], []) are identical, therefore AC_RUN_IFELSE 
must not print a warning if $# is less than 4.  Or, in m4 terms, we 
prefer that choices be made on whether $4 is the empty string, and not 
on whether $# == 4; when $4 is a documented argument.  The m4 builtin 
regexp is an example where regexp([abc],[b]) behaves much differently 
than regexp([abc],[b],[]), because it specifically checks $#, but that's 
confusing enough that you can see why we don't encourage those semantics 
in other autoconf macros.

Meanwhile, I don't see how warning on unexpected arguments is a problem 
- code that wants to be portable to both old and new autoconf will only 
use as many arguments as the old version supported.  On the upgrade 
path, if 2.67 supports only ABC(arg1), but 2.68 supports ABC(arg1, 
arg2), then calling ABC(arg1, oops) in 2.67 would be a syntax warning, 
while in 2.68 it would silently assume you wanted to use the new 
argument feature.  But it does imply that to get the full benefit of 
those portability warnings that you would have to use the oldest version 
of autoconf that you plan to be portable to (but also one new enough 
that it actually detects and prints warnings in the first place).

--
Eric Blake   address@hidden    +1-801-349-2682
Libvirt virtualization library http://libvirt.org