On 14 February 2012 10:35, Dr. Alexander Klein
<[email protected]> wrote:
> >> accumulating results of repeated function evaluations (inject_into),
> >
> > accumarray?
> Will this do something like
> inject_into (@(x, y){x{:},rand(y)},num2cell(1:10),{})
No, I suppose I misunderstood what you meant with "accumulate". But I
don't find this function call more convenient than
cellfun(@(x) rand(x), num2cell(1:10), "unif", 0);
It seems like extra syntax for no additional benefit.
> Really, all those functions are meant for the most part to provide
> ways to get away without loops, maybe to avoid indexing errors along
> the way, even when objects are complex, and predicates ignorant
> about of (cell) arrays. Nothing more, and nothing less.
But Octave already has ways to do this:
http://www.gnu.org/software/octave/doc/interpreter/Vectorization-and-Faster-Code-Execution.html
> I really liked the ease of use with collections in Smalltalk, but
> then again I'm too nostalgic, maybe?
This seems to be a bit too much like a square peg for a round hole.
Octave is, somewhat intentionally, a very limited language. If you're
trying to use Smalltalk message-passing in Octave, you will run into
many frustrations. I still think that using reject and select instead
of logical indexing is strange for Octave.
A further problem is that cellfun suffers a performance penalty when
using anonymous functions, and this is somewhat inevitable due to the
design of how function in Octave work.
But perhaps there are benefits here I don't see. If Carnë thinks these
functions should go into general or miscellaneous, I suppose there is
no harm in doing so.
- Jordi G. H.
------------------------------------------------------------------------------
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d
_______________________________________________
Octave-dev mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/octave-dev