On Wed, Sep 2, 2009 at 7:23 AM, Daniel James
template< typename CompatibleKey,typename CompatibleHash,typename CompatiblePred > iterator find( const CompatibleKey& k, const CompatibleHash& hash,const CompatiblePred& eq)const;
which is a pure extension without backwards compatibility problems. This is seemingly what Boost.Intrusive is doing.
Yes, I think I'll add that to the new version which should be released in 1.41 or 1.42.
This would indeed cleanly solve my problem. FTR, I tend to lean on Joaquín's side of the argument. If the provided explicit Hash/Pred template parameters of unordered_set do provide overloads for compatible keys, then it's "by design" that find() should use them. If previous code that was relying on an implicit conversion to key_type starts using added overload on Pred/Hash, this simply means that the container designer made this choice explicitly, and I don't see why that a bad thing. After all the overloads were added for avoiding the key creation, which is exactly what I was trying to do here. In Joaquín's proposal, all clients using find() implicitly start using the optimized find() avoiding the key_type conversion once the container is modified and the client code is recompiled. No code necessary. That's a good thing to me. But that's just me ;) --DD