[boost] Re: Design discussion: easy i/o

From: Moore, Paul (Paul.Moore_at_[hidden])
Date: 2000-03-08 07:42:41

From: Dave Abrahams [mailto:abrahams_at_[hidden]]
> > This relies on the rule that temporaries are destroyed at 
> > the end of the full expression where they are created. As far
> > as I can see, this rule means that the example works, but it
> > also means that it's a little fragile in practice.
> 
> Why does that rule mean your technique is fragile in practice? I don't
> follow.

Hmm. You're right. Now that the standard exists and says what it does, I
can't think of an example which could give problems. Before temporary
lifetimes were standardised, I seem to recall that destruction at the end of
the block was an option - which would of course produce completely the wrong
results.

Actually, if I embed a print inside another expression, say

   func(((MB() << "Calling func"), arg));

then the "full expression" is the function call, so destruction happens
later than I intend. But that's a pathological case, and I have no sympathy
for anyone who tries it :-)

Paul.