Subject: Re: [ub] Justification for < not being a total order on pointers?
From: Jason Merrill (jason_at_[hidden])
Date: 2013-10-16 16:49:03
On 10/15/2013 06:39 PM, Nevin Liber wrote:
> The current rule of "calling operator< on pointers can invoke ub at the
> drop of a hat", while historically necessary, is a horrible, horrible rule.
Where is this rule? What I see in the current WP is
"Otherwise, the result of each of the operators is unspecified."
And unspecified is significantly different from undefined.
SG12 list run by herb.sutter at gmail.com