In early version, the serialzation of pointers was overloadable at a higher level. After a lot of discussion, I was convinced to change to the current system. The achieved a real benefit - unfortunately I don't remember what it was. The previous overload was still in there - but undocumented. More recently, I deleted this "middle" layer. Robert Ramey Andreas Brinck wrote:

Hi,

...

For this to work, one needs a SFINAE compiler, right?

yes, perhaps the code could be modified to use the boost::enable_if utility to make it a bit clearer what's going on.

...

I faintly remember that early versions of the documentation (during review) described a completely different type of support for classes with operator new. I don't recall the details but I believe that one was able to specialize a template for types with operator new. While being slightly more cumbersome, this sort of support would work with virtually any compiler, not just the ones that support SFINAE. It would be interesting to hear why this kind of operator new support was discontinued later.

I think you answered your own question. Judging from the lack of responses to my post, serializing objects with overloaded new doesn't seem to be important to most people.

I guess the maintainers of the library didn't want to complicate the interface to accomodate a small minority of users.

The SFINAE principle is probably the best way to handle it right now.

Regards

/Andreas Brinck ###########################################

This message has been scanned by F-Secure Anti-Virus for Microsoft Exchange. For more information, connect to http://www.f-secure.com/

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