Re: [boost] reinterpret_cast : differents structs or classes withidentical structural type

From: Jean-Francois Bastien (jfbastien_at_[hidden])
Date: 2007-10-02 17:16:23

> It does. It says that members of a struct or class (even 
> non-POD ones, which I think is a useless restriction) that 
> have no intervening access specifier are stored in memory in 
> the same order as they are declared.

[snip]

> What I personally don't understand is why private members - 
> or in fact all members of non-PODs - can't always be 
> re-ordered. It's not like anyone should ever depend on their 
> order. It probably was C compatibility that there were any 
> guarantees at all.

Binary compatibility is one reason: you need a guideline that allows
compilers to agree on class layout. It's a pain, but it's necessary if
ever you're to compile a .h in two separate places (and even in two
different compiler versions, to some extent) and have the same class
layout.

JF