Ben Abbott | 2 Sep 03:13 2014
Picon

scripts/deprecated/nargchk.m

Rik,

nargchk.m didn't make it into scripts/deprecated.

	http://hg.savannah.gnu.org/hgweb/octave/file/be8a12acb20a/scripts/deprecated

I'd have fixed it for you, but I've never moved a file between directories before and didn't want to risk
breaking something.

Ben

Rik | 1 Sep 23:45 2014

Deprecating aliases finite and fmod

9/1/14

Maintainers,

Is there any objection to deprecating the function aliases of finite and
fmod in the 4.2 release?

finite/isfinite are the same function, but isfinite is what Matlab uses and
is consistent with the rest of the predicate tests which start with "is*"
such as isnumeric, islogical, isnan, etc.

fmod/rem are the same function, but rem is what Matlab uses.  There is also
the additional complication that Octave supports a mod function which is
different from rem/fmod.  When I see fmod I think it should be closer to
mod than to rem.

--Rik

Julien Bect | 1 Sep 18:42 2014
Picon
Picon

Email addresses for package authors and maintainers

Hello all,

For most (all ?) Octave-Forge packages, the "package_box_contents" 
container looks like this:

---------------------------------------------
<div class="package_box_contents">
<table>
<tr><td class="package_table">Package Version:</td><td>1.2.0</td></tr>
<tr><td class="package_table">Last Release 
Date:</td><td>2013-11-25</td></tr>
<tr><td class="package_table">Package Author:</td><td>various 
authors</td></tr>
<tr><td class="package_table">Package Maintainer:</td><td>Mike Miller 
<mtmiller <at> ieee.org></td></tr>
<tr><td class="package_table">License:</td><td><a href="COPYING.html 
<view-source:http://octave.sourceforge.net/communications/COPYING.html>">GPLv3+</a></td></tr>
</table>
</div>
---------------------------------------------

As you can see, the email address is included in the source but not 
displayed, because <...> is interpreted as an (incorrect) HTML tag.

There are several ways for fixing this:

a) Replace <...> by &lt;...&gt; to display the email address.

b) Remove the email address from "package_box_contents".

(Continue reading)

Rik | 1 Sep 03:35 2014

Re: nchoosek

On 08/31/2014 09:00 AM, octave-maintainers-request <at> gnu.org wrote:
> Le 31/08/2014 05:43, Rik a écrit :
>> nchoosek ([4 5], 8)
>
> With Matlab R2012a:
>
> -----------------------------------------------------------------
> >> nchoosek (5, 8)
>
> Error using nchoosek (line 48)
> K must be an integer between 0 and N.
>
> >> nchoosek ([4 5], 8)
>
> ans =
>
>    Empty matrix: 0-by-8
> -----------------------------------------------------------------
>
> I find this behaviour of Matlab rather unfortunate. Actually, I would
> have defined the output to be 0 and [0 0].
>
> Having nchoosek (n, k) be 0 when k > n is a very useful convention that
> make lots of combinatorial formulas easier to write (and prove). 

nchoosek is really two functions in one depending on whether the first
input is scalar.  If the first input is a scalar then the calculation is
the binomial coefficient so

nchoosek (5, 8)
(Continue reading)

Rik | 31 Aug 05:43 2014

Help with nchoosek

Could someone run the following under Matlab?

nchoosek (5, 8)
nchoosek ([4 5], 8)

I expect an error on the first code sample and am curious about whether the
second returns anything or also emits an error.

Thanks,
Rik

Rik | 31 Aug 04:23 2014

Re: Error with test script

On 08/29/2014 11:58 PM, octave-maintainers-request <at> gnu.org wrote:
Subject:
Octave becoming too MATLAB compatible ?
From:
Ulf Griesmann <ulfgri <at> gmail.com>
Date:
08/29/2014 01:21 PM
To:
octave-maintainers <at> gnu.org
List-Post:
<mailto:octave-maintainers <at> gnu.org>
Content-Transfer-Encoding:
8bit
Precedence:
list
MIME-Version:
1.0
Message-ID:
<5400E0D8.9020500 <at> gmail.com>
Content-Type:
text/html; charset=utf-8
Message:
2


I am sure the following script used to work in Octave:

-----------------------
# test script
say_hello;

function say_hello
    fprintf('Hello !\n');
endfunction
-------------------------

Now I get the following error in both 3.8.2 and the development version:

octave:1> testscript
error: 'say_hello' undefined near line 5 column 1
error: called from:
error:   /home/ulfg/Octave/testscript.m at line 5, column 1
octave:1>

I don't know if this is a bug or intentional.

I don't think this was ever legal in Octave.  I just tested with versions 3.2.4, 3.4.3, 3.6.4, 3.8.0 and they all give the same error as with 3.8.2.

--Rik

Ulf Griesmann | 29 Aug 22:21 2014
Picon

Octave becoming too MATLAB compatible ?


I am sure the following script used to work in Octave:

-----------------------
# test script
say_hello;

function say_hello
    fprintf('Hello !\n');
endfunction
-------------------------

Now I get the following error in both 3.8.2 and the development version:

octave:1> testscript
error: 'say_hello' undefined near line 5 column 1
error: called from:
error:   /home/ulfg/Octave/testscript.m at line 5, column 1
octave:1>

I don't know if this is a bug or intentional.

-- Ulf Griesmann \\ tel: +1 301 339 4962 16833 Westbourne Terrace \\ Gaithersburg, MD 20878-2033 \\ e-mail: ulfgri <at> gmail.com
Rik | 29 Aug 18:35 2014

clearing intermediate variables at end of functions?

8/29/14

All,

I was just cleaning up nchoosek.m and I came across this construct at the
end of the function:

--- Begin Code ---
    clear cA b;
    C = C.';
  endif

endfunction
--- End Code ---

Is there any reason why it should be necessary to clear the intermediate
variables cA and b manually?  They are local to the m-file and as soon as
the endfunction is hit they will go out of scope and the memory should be
reclaimed.  I thought maybe, just maybe, it was because they were large and
the original programmer was worried that the transpose operation was going
to create an intermediate copy of C.  But this shouldn't be the case,
should it?  Don't we do something intelligent like simply reverse the row,
column attributes on the existing C matrix rather than creating a full
blown copy?

--Rik

frank wang | 29 Aug 12:49 2014
Picon

From: frank wang


Sup octave

http://barclaismiddlemarket.com/however.php?teavp3113dha

frank wang

Rohit Gupta | 28 Aug 14:23 2014
Picon

error installing imfill

I am using octave and want to use imfill.I am encountering the following error:

pwd is C:\Users\rohit.gupta

and I have saved imfill.m in the above mentioned directory.

But when i am trying to run imfill its throwing error:


error:the number of arguments must be 2 to 3


How can i solve this

Thank you
Colin Macdonald | 28 Aug 11:40 2014
Picon
Picon

Re: Octsympy-0.1.0, symbolic package

On 28/08/14 01:26, Ben Abbott wrote:
> I'm happy to try some tests on Mac OSX.  I can confirm that I'm
> able to install OctSymPy 0.1.0 running Octave 3.8.2 on Mac OS
> 10.9.4.
> 
> I ran the test suite.
> 
> ***** Ran all tests, 250.122 seconds (208.481s CPU) *****
> 
> There were several expected failures, but I didn't notice any other
> failures.

Thanks Ben!  The tests should get a bit quieter w/ future versions of
sympy.  Perhaps I should comment a few known failures out until then.

Did the unicode pretty print work on Mac?  Compare:

>> syms x
>> A = sin(x/2)
>> disp(A)
>> disp(A, 'ascii')

It does not appear to work for me on Windows 3.8.2 MXE GUI.  I think
even the editor cannot display utf-8 .m files properly (could be a
local font problem I suppose).  Perhaps that's an Octave bug.

thanks,
Colin
Attachment (0xC5326EE5.asc): application/pgp-keys, 2218 bytes

Gmane