RE: [boost] Re: shifted_ptr: why?

From: Powell, Gary (powellg_at_[hidden])
Date: 2003-10-01 16:34:45

Gennadiy > Do you want me to continue documenting shifted_ptr as it is?  

Abrahams >Yes, please.

> Do you think the framework will be able to share different
> allocations between different pointers?  

Abrahams >"The framework" is mere vapor as of today.  There's been some work on
Abrahams >it, but it would be crazy to hold up your own work waiting for it.

Here I agree with Dave, boost is not the C++ Library Standard's committee. IMO boost has room for multiple implementations of libraries and concepts. Whether to accept or reject a library should be based on whether the code and documentation is up to "boost standards" (vague), and whether there is any other perceived interest in the library.

In this case I'm voting that the documentation is not up to snuff. And therefore am rejecting it. I'm also rejecting it in part due to the interface which would easily allow raw pointers of the wrong kind to be added to it.

Both of these things can be fixed and it looks like they will and then I expect it will be back.

If later there is a general purpose Smart Pointer library that supercedes this one, we can deprecate it or continue to let Philppe maintain it. Its similar to boost::bind and boost::lambda::bind and boost::compose. And soon perhaps boost::fc++.

  -Gary-