Re: [boost] [SoC][string_cvt] The proposal for string conversionslibrary.

From: Jeff Garland (jeff_at_[hidden])
Date: 2006-05-01 08:24:33

On Mon, 1 May 2006 08:05:52 -0400, Beman Dawes wrote
> "Jeff Garland" <jeff_at_[hidden]> wrote in message 
> news:20060501040735.M23040_at_crystalclearsoftware.com...
> >...
> > One other proposal you should be aware of is Pete Becker's n1982 for 
> > simple
> >> numeric to string conversions:
> >
> > http://www.open-std.org/JTC1/SC22/WG21/docs/papers/2006/n1982.html
> >
> 
> >...
> > While this is a very C like proposal it fills a basic gap in C++ for 
> > basic,
> > high performance, (w)string to number conversions.  My understanding is 
> > this
> > the committee is pretty fond of this proposal -- perhaps someone that was 
> > in
> > Berlin can say more (Beman, Howard?).
> 
> At least at the moment, the LWG intends to go forward with both
> proposals. I'm personally not convinced that is a good idea. Sounds a
> bit too much like design-by-committee. Time will tell.

Has anyone considered the possibility that Pete's proposal might serve as the
core for a more efficient lexical_cast?  I guess internationalization / locale
 issues prevent that scenario?

> >> Possible mentors for this project could be authors of the "Lexical
> >> Conversion Library Proposal for TR2" proposal - Kevlin Henney and/or
> >> Beman Dawes.
> >
> >I'm not sure Kevlin or Beman is available, but I'm certain we can find a 
> >good
> >mentor if the proposal is accepted.
> 
> I'm trying to focus this summer on moving Boost libraries into TR2,
>  so am not signing up for mentoring. Kevlin is always overcommitted too.

Yep, that's what I figured.
 
> OTOH, it really would be helpful is someone looked at the entire
> conversion picture, and tried to synthesize a complete solution.

I agree.  And I presume you will want to make time to review any designs in
this area even if it isn't officially mentoring?  

Jeff