|
|
Ublas : |
Subject: Re: [ublas] Patch/proposal for overloading operator * in ublas
From: K.M.A.Chai_at_[hidden]
Date: 2009-09-04 06:00:49
Hi Gunter,
Thanks for the interest. My feeling is that ublas and umatlab
really have different aims, so most of parts of umatlab may be out of
place in ublas (though, lu.hpp also seems out of place in ublas).
That said, it will be nice if ublas can be easily extendable. A lot of
people has been asking about overloading * in ublas. However, currently
ublas is pretty greedy: it tries to handle all of operator*, so that
extending operator* in a naive way (by the end user) will lead to the
compiler complaining. Attached are two patches to make ublas less greedy
in this aspect.
There are some other patches for ublas: I'll submit them once I recall
why they're there at all :)
Another possible addition would be the parser that breaks down an
expression into the consitutents for blas calls.
Kian Ming (Adam)
On Thu, 3 Sep 2009, Gunter Winkler wrote:
> Am Tuesday 18 August 2009 schrieb K.M.A.Chai_at_[hidden]:
>> Hi,
>>
>> Perhaps this is a nice time to introduce a matlab-like extension
>> to ublas that I've been working on:
>>
>> http://homepages.inf.ed.ac.uk/s9810791/umatlab.tbz
>>
>> ----
>> This package provides "syntactic sugar" for doing matrix programming
>> in C++. Among others, it is built on top of
>>
>> - ublas in boost
>> - numeric-bindings in boost
>
> This looks very interesting. Do you have contributions which should be
> included in the official uBLAS release?
>
> Since the development of uBLAs is quite frozen for a long time it would
> be nice to have some new features.
>
> mfg
> Gunter
>
>
>
-- The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.