Re: [boost] Re: in/out parameters, codingstylesandmaintenance

From: Justin M. Lewis (boost_at_[hidden])
Date: 2003-05-03 21:46:23

----- Original Message ----- 
From: "Gregory Colvin" <gregory.colvin_at_[hidden]>
To: "Boost mailing list" <boost_at_[hidden]>
Sent: Saturday, May 03, 2003 5:53 PM
Subject: Re: [boost] Re: in/out parameters, codingstylesandmaintenance

> On Saturday, May 3, 2003, at 18:07 America/Denver, Justin M. Lewis
> wrote:
> > From: "Gregory Colvin" <gregory.colvin_at_[hidden]>
> >> On Saturday, May 3, 2003, at 17:52 America/Denver, Noel Yap wrote:
> >>> Gregory Colvin wrote:
> >>>>
> >>>> On Saturday, May 3, 2003, at 17:13 America/Denver, Noel Yap wrote:
> >>>>
> >>>>> Noel Yap wrote:
> >>>>>>   std::auto_ptr< T > t( new T() );
> >>>>>>   f( boost::dumb_ptr< T >( t.get() ); // clearly an in/out
> >>>>>> parameter
> >>>>>
> >>>>> Or if you don't want to dynamically allocate t:
> >>>>>
> >>>>>   T t;
> >>>>>   f( boost::dumb_ptr< T >( &t ) ); // clearly an in/out parameter
> >>>>>
> >>>>>   static T t;
> >>>>>   f( boost::dumb_ptr< T >( &t ) ); // clearly an in/out parameter
> >>>>
> >>>> Not so clear -- it depends on whether f() cares about the initial
> >>>> value of t.
> >>>
> >>> So you're saying the parameter can be an out-only parameter?  If so,
> >>> for
> >>> out parameters, use:
> >>>
> >>>   boost::dumb_ptr< T > f();
> >>>   t = f();
> >>>
> >>> IMHO, the intent is much more clear than:
> >>>
> >>>   f( out< T > t );
> >>>   f( out< T >( t ) );
> >>
> >> Not to me it isn't.
> >>
> >>>>  Also, not in my opinion any more clear than:
> >>>>
> >>>>     f(&t);
> >>>
> >>> I agree, but the OP doesn't like raw pointers from what I've gathered
> >>> from the thread.
> >>
> >> Who is OP?
> >>
> >> Anyway, I dislike out parameters more than I do raw pointers.
> >>
> > That doesn't really make sense, since out params aren't always
> > avoidable,
> > nor are in_out params.  I mean, disliking them aside, they're necessary
> > sometimes.
>
> Agreed.  And when they are necessary I am happy enough with
> pointers -- or at least my unhappiness is not reduced by
> any of various wrappers proposed here.
>

The difference here is, at the call site you still get an explicit
description of how the data is being used, with just a pointer, there's no
information given.

>
> _______________________________________________
> Unsubscribe & other changes:
http://lists.boost.org/mailman/listinfo.cgi/boost
>