Re: [boost] Re: [boost.tr1?] request for a hash<> implementation

From: JOAQUIN LOPEZ MU?Z (joaquin_at_[hidden])
Date: 2005-02-05 12:09:01

----- Mensaje original -----
De: Daniel James <daniel_at_[hidden]>
Fecha: Sábado, Febrero 5, 2005 5:36 pm
Asunto: [boost] Re: [boost.tr1?] request for a hash<> implementation

> John Maddock wrote:
> > Anyone else want to throw in their thoughts?  What's the 
> timescale on 
> > the unordered container development, how much is there left to do?
> 
> I'll upload a new version to the file vault later today, or maybe 
> tomorrow, which should be compliant to the draft. I haven't 
> written any 
> documentation yet, I guess I'll probably just write an 
> introduction and 
> refer to the draft for reference (as Beman Dawes suggested), if 
> that's okay.
> 
> There are a few areas which could do with some improvement, such 
> as the 
> hash functions, which are pretty basic. But I guess they can be 
> left for 
> later if you want to get it in soon.
> 

Hi Daniel,

What I'm interested in in the short term is the hash<> functor
alone. Of course unordered containers are much more interesting,
but this will have to undergo the usual review process, I guess.
Do you think we can put your hash<> implementation under
boost/functional/hash.hpp and have some short docs for it?
Are you planning to improve the hashing algorithms (in particular
for strings, which I guess it's the most improvable part)?
If I could have this in time for Boost 1.33 I'd be extremely
grateful, since Boost.MultiIndex hashed indices (practically
finished now) need a hash<>, and I don't think it is convenient
that each library ships with its own version.

Thank you,

Joaquín M López Muñoz
Telefónica, Investigación y Desarrollo