RE: [boost] Review request: minmax library

From: Yitzhak Sapir (yitzhaks_at_[hidden])
Date: 2002-07-25 09:07:27

> -----Original Message-----
> From: Thomas Witt [mailto:witt_at_[hidden]]

> Herve,
> 
> On Monday 22 July 2002 17:17, Herve Bronnimann wrote:
> > Dear review wizard: I'd like to request a formal review of 
> the minmax
> > library that's been sitting in the sandbox for a little while now.
> > Please contact me to arrange a schedule.
> 
> I am a bit concerned regarding this one. I am unsure whether 
> the size and 
> scope of the library justifies the effort of a formal review. 
> Though I might 
> be wrong. IIRC somebody proposed to integrate this with a 
> possible boost 
> algorithm library.
> 
> Comments anybody ?
> 
> Thomas
> 
> Boost Review Wizard

Maybe there should be procedures for a formal review of portions/functions into utility/algorithm libraries?  I did see some boost functions in an algorithm header file called none/all, and after seeing them I thought they were already part of the boost libraries that I tried using them remembering I saw them in a boost header.  I was surprised when I found out they weren't.  They seemed very useful and the type of thing people would either write themselves because it wasn't available or would do it the long way each time many times, both reasons for inclusion in a library.

Doing a file search, I see them now in a file in the detail subdirectory of boost and they seem part of a bigger "sequence algorithms" library that has "not yet undergone formal review, but no review is scheduled.  Why is that?  Is there a formal review planned or is there work in progress on this?  If so, couldn't it be submitted in parts (as an "algorithm" library has very large scope and could consist of many small parts, just like "utility")?  If not, why not?  Is everyone just waiting for someone to take up work on this?  I am talking about all the functions in boost/detail/algorithm.hpp that are defined on iterators.