In MSDN length_error sample,
it uses internal macro _SIZT, other than size_t. I have found in file xmemory, _SIZT is defined to size_t.
So, my questions are
1. why not use size_t directly, why use an internal name (begins with _)?
2. Do we have to use _SIZT when we define a customized allocator?
thanks in advance,