From: Peter Dimov (pdimov_at_[hidden])
Date: 2005-08-17 07:10:47
Peter Dimov wrote:
> This is very disturbing; it basically breaks all SFINAE'd operators.
> If this is what the language currently mandates, it needs to be fixed.
As Mike Miller pointed out on the core reflector, this is issue #488:
and the tentative resolution is to treat the use of a type without linkage
as a deduction failure.
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk