Re: [boost] [pimpl] Mini Review

Subject: Re: [boost] [pimpl] Mini Review
From: Thomas Klimpel (Thomas.Klimpel_at_[hidden])
Date: 2011-05-26 07:40:40

Artyom Beilis wrote:
> Should The library be accepted in Boost:
> ========================================
> 
> No it should not. It is "a fancy" library that was written
> without a knowledge of the problem domain. It tries
> to solve problem in smart way making it overcomplicated.
> 
> IMHO old std::auto_ptr is much more suitable
> for writing pimpl-objects then Boost.Pimple.
> 
> For easy pimpl-ideom you do not need a fancy class
> but rather a set of pimpl-suitable smart pointers.

I haven't reviewed the Pimpl library, but this comparison to "old std::auto_ptr" looks very strange to me. If I understood correctly, one of the reasons why "old std::auto_ptr" is deprecated in favor of "std::unique_ptr" is that "old std::auto_ptr" will silently break the default generated copy-constructor and assignment operator of a class containing an "old std::auto_ptr" as member. "std::unique_ptr" on the other hand will disable default generated copy-constructor and assignment operator and only default generate the "move" versions of these, so the user will at least realize that he has to manually provide correct versions of copy-constructor and assignment operator.

Did I totally misunderstood you here, or was this comparison intended as a ironic remark, or is my understanding old "old std::auto_ptr" incorrect, or ...?

Regards,
Thomas