From: David Abrahams (abrahams_at_[hidden])
Date: 2001-03-12 19:15:26
----- Original Message -----
From: "Peter Dimov" <pdimov_at_[hidden]>
> From: "David Abrahams" <abrahams_at_[hidden]>
> > Why would anyone want to make this change? I introduced the macro, but
> > was precedent for it in Alexey's work. It was well-considered as
> > far as I can tell. Why mess with it?
> Consistency, I suppose. Macro arguments are traditionally separated with a
> comma. Why introduce another "convention" on a case by case basis?
Ah, but you only beg the question, because the way it is now, there are only
two macro arguments.
> Not that it matters, of course. This is an implementation detail. Users
> aren't supposed to use (or even see) the macro... or are they?
No, they're not. They are indeed an implementation detail.
I appreciate Daryle's arguments for a slight safety improvement.
If the macro was for our users, his arguments would balance about evenly
with the argument for readability.
As it stands, I have no strong opinion, but am inclined to leave things
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk