Re: .ONESHELL enhancement?

[Ralf Wildenhues]

* Matt McCutchen wrote on Mon, Oct 05, 2009 at 08:50:42AM CEST:
> On Mon, 2009-10-05 at 08:39 +0200, Ralf Wildenhues wrote:
> > Well, do it only if the command line length would otherwise be exceeded,
> > of course.  That way you avoid any regressions that could stem from
> > detectability of the way it was invoked.
> 
> It sounds like your argument is, if the existing technique would fail
> due to the length limit, we aren't breaking anything by switching to a
> different technique.  I don't buy that because:
> 
> 1. It may be hard to predict 100% accurately whether a particular
> command on a particular OS would exceed the limit,

If execve fails with E2BIG, you can be pretty sure of what happened.

> and in any case I'm skeptical about adding such code to make.

> 2. A user might rather see the original "argument list too long" error
> than have make switch to a different approach that potentially fails
> later in a different way.

I can understand such reasoning only for portability testing.  But for
those, testing on Linux isn't sufficient anyway, because its limit is
far higher than that of some other systems.  All you'd trade is a
failure with possibly success, and possibly a different failure.

I still don't see a reason not to do better on Linux; but I think I've
made my point on this now.

Cheers,
Ralf