Referencecounted object vs. referencecounted handles

[Steven T. Hatton]

It looks as if boost::intrusiv e_ptr<> didn't make the cut for C++0X.

http://www.open-std.org/JTC1/SC22/WG21/docs/papers/2005/n1745.pdf

I guess I should take this as an indication that C++ programmers favor
referencecounte d handles over reference counted objects.

This was an interesting read, and seems to propose the kind of complete
solution I was looking for. I don't know if it would be worth the trouble,
however:

http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2004/n1671.pdf

Has anybody else used the Boost Smart Pointers, and care to share opinions
or observations on this topic?
--
If our hypothesis is about anything and not about some one or more
particular things, then our deductions constitute mathematics. Thus
mathematics may be defined as the subject in which we never know what we
are talking about, nor whether what we are saying is true.-Bertrand Russell

[Pete Becker]

Re: Referencecounte d object vs. referencecounte d handles

Steven T. Hatton wrote:
[color=blue]
> It looks as if boost::intrusiv e_ptr<> didn't make the cut for C++0X.[/color]

boost::intrusiv e_ptr hasn't been proposed, neither for TR1 nor for C++0x.
[color=blue]
> http://www.open-std.org/JTC1/SC22/WG...2005/n1745.pdf
>[/color]

That's TR1.

--

Pete Becker
Dinkumware, Ltd. (http://www.dinkumware.com)

[Steven T. Hatton]

Re: Referencecounte d object vs. referencecounte d handles

Pete Becker wrote:
[color=blue]
> Steven T. Hatton wrote:
>[color=green]
>> It looks as if boost::intrusiv e_ptr<> didn't make the cut for C++0X.[/color]
>
> boost::intrusiv e_ptr hasn't been proposed, neither for TR1 nor for C++0x.[/color]

I assumed that everything in the boost smart pointer library was a potential
candidate for inclusion in the standard.
[color=blue][color=green]
>> http://www.open-std.org/JTC1/SC22/WG...2005/n1745.pdf
>>[/color]
>
> That's TR1.[/color]

That's what the document says. It also says the items described therein had
potential for inclusion in the standard. I merely assumed that if
boost:instrusiv e_ptr<> wasn't there, it was unlikely to be included in
C++0X.

--
If our hypothesis is about anything and not about some one or more
particular things, then our deductions constitute mathematics. Thus
mathematics may be defined as the subject in which we never know what we
are talking about, nor whether what we are saying is true.-Bertrand Russell

[Pete Becker]

Re: Referencecounte d object vs. referencecounte d handles

Steven T. Hatton wrote:
[color=blue]
> Pete Becker wrote:
>[color=green]
>>
>>That's TR1.[/color]
>
>
> That's what the document says.[/color]

Funny how that works.
[color=blue]
> It also says the items described therein had
> potential for inclusion in the standard. I merely assumed that if
> boost:instrusiv e_ptr<> wasn't there, it was unlikely to be included in
> C++0X.
>[/color]

Bad assumption.

--

Pete Becker
Dinkumware, Ltd. (http://www.dinkumware.com)

[Panjandrum]

Re: Referencecounte d object vs. referencecounte d handles

Pete Becker wrote:[color=blue]
> boost::intrusiv e_ptr hasn't been proposed, neither for TR1 nor for C++0x.[/color]

Sometimes there is even good news from the Standards committee.

[Steven T. Hatton]

Re: Referencecounte d object vs. referencecounte d handles

Panjandrum wrote:
[color=blue]
> Pete Becker wrote:[color=green]
>> boost::intrusiv e_ptr hasn't been proposed, neither for TR1 nor for C++0x.[/color]
>
> Sometimes there is even good news from the Standards committee.[/color]

Why do you say that? Thus far, I haven't seen anybody provide any
substantive comment on the use of boost::intrusiv e_ptr<>.

--
If our hypothesis is about anything and not about some one or more
particular things, then our deductions constitute mathematics. Thus
mathematics may be defined as the subject in which we never know what we
are talking about, nor whether what we are saying is true.-Bertrand Russell

[Panjandrum]

Re: Referencecounte d object vs. referencecounte d handles

Steven T. Hatton wrote:[color=blue]
> Thus far, I haven't seen anybody provide any
> substantive comment on the use of boost::intrusiv e_ptr<>.[/color]

Guess why?

[Steven T. Hatton]

Re: Referencecounte d object vs. referencecounte d handles

Panjandrum wrote:
[color=blue]
> Steven T. Hatton wrote:[color=green]
>> Thus far, I haven't seen anybody provide any
>> substantive comment on the use of boost::intrusiv e_ptr<>.[/color]
>
> Guess why?[/color]

Nobody knows how to use them.
--
If our hypothesis is about anything and not about some one or more
particular things, then our deductions constitute mathematics. Thus
mathematics may be defined as the subject in which we never know what we
are talking about, nor whether what we are saying is true.-Bertrand Russell

[Swampmonster]

Re: Referencecounte d object vs. referencecounte d handles

Steven T. Hatton wrote:[color=blue]
> Panjandrum wrote:
>
>[color=green]
>>Steven T. Hatton wrote:
>>[color=darkred]
>>>Thus far, I haven't seen anybody provide any
>>>substantiv e comment on the use of boost::intrusiv e_ptr<>.[/color]
>>
>>Guess why?[/color]
>
>
> Nobody knows how to use them.[/color]

I know how to use them and I do use them. Though only for COM
stuff. I think maybe the reason why so few people care for
reference counted objects, or better: for a predefined
smart pointer for reference counted objects, is, that:
a) few people use ref counted objects
b) even fewer libs exist that use ref counted objects
c) it's so easy to roll your own implementation

And ref counted objects tend to get really bitchy when you
build cycles. Therefor their usefulness is very limited in
many situations.