build regression

[Neil Booth]

Alexandre Oliva wrote:-

> > I want to be able to *decode* all switches before *handling* any of
> > them, and to do this switch decoding has to be language and target
> > independent
> 
> I don't understand why decoding has to be language and
> target-independent in order to be able to decode before handling.

Because otherwise you wouldn't know their meaning; and whether the
following element in the argv vector is an argument or a switch.

> Could you please explain it to me?  Please start by describing whether
> you're talking about option handling in the compiler driver, in
> front end actual compilers, or both.

We already have a similar restriction in the driver; it's never been
mentioned as a problem that someone might want an -m or an -f option to
take a separate argument, for example.

Unless you can come up with existing front ends that this will cause
a problem with, I'm not going to worry about it.  The driver can
rewrite options anyway; that's what it's main purpose in life is after
all.

Neil.