octave-dev — GNU Octave developers' list

[OctDev] interp2.m

[Tom H. <to...@ku...>]

The interp2.m script doesn't work if XI and YI are not meshgrids.
For example, it grabs the first row of XI only to sort in with the original
grid (which must be a monotone meshgrid).  If the other rows have different mins 
and maxes it'll get incorrect results.

Has anybody noticed and/or fixed this already?

It seems like all that might be necessary is to find the global min and max of 
the XI and YI arrays, instead of just the first row/column, but I haven't played 
with the code enough to know for sure yet.
-- 
Dr. Tom Holroyd
"A man of genius makes no mistakes. His errors are volitional and
are the portals of discovery." -- James Joyce

[OctDev] unsubscribe [pmastroianni@teleflex.bc.ca]

[Ray W. <rw...@te...>]

unsubscribe [pma...@te...]

oct...@li... wrote:

>Send Octave-dev mailing list submissions to
>	oct...@li...
>
>To subscribe or unsubscribe via the World Wide Web, visit
>	https://lists.sourceforge.net/lists/listinfo/octave-dev
>or, via email, send a message with subject or body 'help' to
>	oct...@li...
>
>You can reach the person managing the list at
>	oct...@li...
>
>When replying, please edit your Subject line so it is more specific
>than "Re: Contents of Octave-dev digest..."
>
>
>Today's Topics:
>
>   1. FANN binding (=?ISO-8859-1?Q?S=F8ren_Hauberg?=)
>
>--__--__--
>
>Message: 1
>Date: Fri, 03 Jun 2005 15:21:40 +0200
>From: =?ISO-8859-1?Q?S=F8ren_Hauberg?= <so...@ha...>
>To:  oct...@li...
>Subject: [OctDev] FANN binding
>
>Hi
>As some of you might know from reading he...@oc..., I've recently=20
>released a binding for the neural network library FANN.
>It has been suggested to me that I tried to get this binding into=20
>octave-forge (in Extras) so I wouldn't have to deal with the black=20
>voodoo called autotools.
>
>Two questions:
>1) Can the binding enter octave-forge?
>2) If so, would anybody be willing to give me a hand with the above=20
>mentioned voodoo?
>
>/S=F8ren
>
>P.S. I should mention that the binding is at http://hauberg.org/octave-fa=
>nn
>
>
>
>--__--__--
>
>_______________________________________________
>Octave-dev mailing list
>Oct...@li...
>https://lists.sourceforge.net/lists/listinfo/octave-dev
>
>
>End of Octave-dev Digest
>  
>

[OctDev] FANN binding

[<so...@ha...>]

Hi
As some of you might know from reading he...@oc..., I've recently=20
released a binding for the neural network library FANN.
It has been suggested to me that I tried to get this binding into=20
octave-forge (in Extras) so I wouldn't have to deal with the black=20
voodoo called autotools.

Two questions:
1) Can the binding enter octave-forge?
2) If so, would anybody be willing to give me a hand with the above=20
mentioned voodoo?

/S=F8ren

P.S. I should mention that the binding is at http://hauberg.org/octave-fa=
nn

[OctDev] Re: Problem compiling fsolve.cc from octave-forge/FIXES

[Paul K. <pki...@us...>]

I have no interest in maintaining this function.  I am removing it from 
the build system for now, and will purge it completely if nobody steps 
forward to update it.

IIRC, the purpose of this function is to pass extra arguments to 
fsolve, but we can do this now with the new syntax:

	@(x) f(x,a,b,c)

- Paul

On May 10, 2005, at 5:50 AM, Michael Kopp wrote:

> Hi all,
>
> I was just trying to compile fsolve.cc from octave_forge/FIXES
> (a link to the file can be found here: 
> http://octave.sourceforge.net/index/optimization.html#Rootfinding).
>
> I used the command "mkoctfile fsolve.cc" on a PC with Suse Linux 9.1. 
> My version of gcc is 3.3.3.
>
> I got the following error message:
>
> **********************************************************************
> fsolve.cc:260: error: cannot convert `double (NLEqn_options::*)() 
> const' to `double (NLEqn_options::*)()' in initialization
> **********************************************************************
>
> The relevant code from fsolve.cc seems to be
>
> **********************************************************************
> struct NLEQN_OPTIONS
> {
>   const char *keyword;
>   const char *kw_tok[MAX_TOKENS + 1];
>   int min_len[MAX_TOKENS + 1];
>   int min_toks_to_match;
>   d_set_opt_mf d_set_fcn;
>   d_get_opt_mf d_get_fcn;
> };
>
> static NLEQN_OPTIONS fsolve_option_table [] =
> {
>   { "tolerance",
>     { "tolerance", 0, },
>     { 1, 0, }, 1,
>     &NLEqn_options::set_tolerance,
>     &NLEqn_options::tolerance, },   // here is the problem
>
>   { 0,
>     { 0, 0, },
>     { 0, 0, }, 0,
>     0, 0, },
> };
> **********************************************************************
>
> The problem is in the line &NLEqn_options::tolerance. If I outcomment 
> this, the rest compiles fine. From 
> http://pareto.uab.es/mcreel/OctaveClassReference/html/index.html
> I get the following information:
>
> **********************************************************************
> NLEqn_options Class Reference
> Public Member Functions
>  	NLEqn_options (void)
>  	NLEqn_options (const NLEqn_options &opt)
> NLEqn_options & 	operator= (const NLEqn_options &opt)
>  	~NLEqn_options (void)
> void 	init (void)
> void 	copy (const NLEqn_options &opt)
> void 	set_options (const NLEqn_options &opt)
> void 	set_default_options (void)
> void 	set_tolerance (double val)
> double 	tolerance (void) const
> **********************************************************************
>
> So the tolerance function is indeed declared with const. 
> Unfortunately, my rudimentary knowledge of C++ does not allow me to 
> find a workaround other than outcommenting the entire line. Does 
> anyone have a better solution?
>
> Thanks in advance,
>
> Michael.
>
>
>
>
> -- 
> Michael Kopp
> Department Biologie II
> University of Munich (LMU)
> Grosshaderner Strasse 2
> 82152 Plannegg-Martinsried
> Germany
>
> Phone: ++49-89-2180-74233
> Email: ko...@zi...
>
>
>
> -------------------------------------------------------------
> Octave is freely available under the terms of the GNU GPL.
>
> Octave's home on the web:  http://www.octave.org
> How to fund new projects:  http://www.octave.org/funding.html
> Subscription information:  http://www.octave.org/archive.html
> -------------------------------------------------------------
>

Re: [OctDev] Re: isdir.m

[Paul K. <pki...@us...>]

It is now index and should show up on the server soon.

Thanks,

- Paul

On May 10, 2005, at 6:05 PM, Tom Holroyd wrote:

> Hmm.  I posted isdir.m, because I couldn't find it on the octave-forge 
> webpage.  Turns out, isdir is in there, but it doesn't show up under 
> "I" here: http://octave.sourceforge.net/index/index.html.
> -- 
> Dr. Tom Holroyd
> "A man of genius makes no mistakes. His errors are volitional and
> are the portals of discovery." -- James Joyce
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by Oracle Space Sweepstakes
> Want to be the first software developer in space?
> Enter now for the Oracle Space Sweepstakes!
> http://ads.osdn.com/?ad_id=7393&alloc_id=16281&op=click
> _______________________________________________
> Octave-dev mailing list
> Oct...@li...
> https://lists.sourceforge.net/lists/listinfo/octave-dev
>

Re: [OctDev] pcre.h not found

[Paul K. <pki...@us...>]

On May 30, 2005, at 12:56 AM, Dmitri A. Sergatskov wrote:

> Paul Kienzle wrote:
>
>> So we don't need to rely on the existence of pcre-config, and I
>> shouldn't apply the patch?
>
> I do not know.
> I assume the following sequence would be good:
>
> -- check for pcre-config
> failing that, check for <pcre.h>
> failing that -- bail out.
>
> No, I do not have a patch :)
>

I did the other order.

Let me know if it works.

- Paul

Re: [OctDev] fzero.m doesn't pass varargin to feval

[Paul K. <pki...@us...>]

Ben,

I've included your changes.

In future, please send 'cvs diff -cp file', or if you are not working 
against the CVS tree, 'diff -cp file file-new'.  Also, please add 
documentation and test cases for the new functionality.

Note that we no longer need to use this hack for optional arguments.

Instead of:

	fzero(f,x0,options,a,b,c)

we can now use:

	fzero(@(x) f(x,a,b,c), x0, options)


Thanks,

	- Paul

On May 10, 2005, at 2:45 PM, Ben Barrowes wrote:

> fzero doesn't seem to pass varargin to feval. There are 6 places 
> including the function statement itself where varargin (function 
> statement) or varargin{:} (when using feval) is needed. I have tested 
> this modified fzero.m for varargin functionality and include it below.
>

Re: [OctDev] pcre.h not found

[Dmitri A. S. <das...@gm...>]

Paul Kienzle wrote:

> 
> So we don't need to rely on the existence of pcre-config, and I
> shouldn't apply the patch?

I do not know.
I assume the following sequence would be good:

-- check for pcre-config
failing that, check for <pcre.h>
failing that -- bail out.

No, I do not have a patch :)

> 
> - Paul
> 

Dmitri.
-- 

Re: [OctDev] pcre.h not found

[Paul K. <pki...@us...>]

On May 30, 2005, at 12:03 AM, Dmitri A. Sergatskov wrote:

> Paul Kienzle wrote:
>> I'll apply this patch, but I'm not too thrilled about relying on the
>> existence of pcre-config.  Do we need a fallback which will search
>> for pcre.h and libpcre if pcre-config is not present?
>> Thanks,
>>     - Paul
>
> It looks to me that it is rather safe to assume that pcre.h is in
> standard include directory -- after some googling it looks to me
> that many other packages depend on it
> (number of reports in redhat's bugzilla also confirm that).

So we don't need to rely on the existence of pcre-config, and I
shouldn't apply the patch?

- Paul

Re: [OctDev] pcre.h not found

[Dmitri A. S. <das...@gm...>]

Paul Kienzle wrote:
> I'll apply this patch, but I'm not too thrilled about relying on the
> existence of pcre-config.  Do we need a fallback which will search
> for pcre.h and libpcre if pcre-config is not present?
> 
> Thanks,
> 
>     - Paul
> 

It looks to me that it is rather safe to assume that pcre.h is in
standard include directory -- after some googling it looks to me
that many other packages depend on it
(number of reports in redhat's bugzilla also confirm that).

Dmitri.
-- 

Re: [OctDev] pcre.h not found

[Paul K. <pki...@us...>]

I'll apply this patch, but I'm not too thrilled about relying on the
existence of pcre-config.  Do we need a fallback which will search
for pcre.h and libpcre if pcre-config is not present?

Thanks,

	- Paul

On Apr 13, 2005, at 5:14 PM, Stefan van der Walt wrote:

> Attached is a much simpler patch that should resolve the situation,
> irrespective of any shuffling done by distros.
>
> I think it is safe to assume that pcre.h is present if pcre-config is?
> Otherwise we can explicitly check, using the output from
>
> `pcre-config --cflags`
>
> Regards
> Stefan
>
> On Wed, Apr 13, 2005 at 10:52:06PM +0200, Stefan van der Walt wrote:
>> On Wed, Apr 13, 2005 at 08:33:22AM -0600, Dmitri A. Sergatskov wrote:
>>> On 4/13/05, Stefan van der Walt <st...@su...> wrote:
>>>> How about the following patch, as well as renaming pcregexp.cc to
>>>> pcregexp.cc.in and doing a variable substitution:
>>>>
>>> ...
>>>
>>> At this moment I am not even sure that this is not a bug with the 
>>> pcre-devel
>>> package. Perhaps, they should have made a symlink
>>> pcre.h -> pcre/pcre.h ?
>>
>> I think that Debian (and some other distros) modify the path to
>> pcre.h.  I notice now that there is a `pcre-config' available.  Maybe
>> we should rather use that to determine compiler flags.

Re: [OctDev] Re: delaunay

[Paul K. <pki...@us...>]

On Apr 28, 2005, at 5:15 AM, Rafael Laboissiere wrote:

> * Dmitri A. Sergatskov <das...@gm...> [2005-04-27 07:49]:
>
>> I noticed that "make install" in octave-forge would install delaunay.m
>> (and related) even though qhull is not present and delaunayn.oct has
>> not being built. Perhaps this is a bug in the octave-forge configure
>> script?
>
> Yes, it seems to be a bug.  Unfortunately, I am not very used to the
> octave-forge configure/Makefile scheme and I would prefer that someone
> else review this issue.

Install is mostly handled by main/Makefile, which walks through every
subdirectory without a NOINSTALL file and runs octinst.sh on it.  This
installs every oct-file and m-file.

The quick hack is to figure out how to make ./configure touch or remove
main/geometry/NOINSTALL depending on whether qhull is present.  You
can't do it from main/geometry/Makefile since this will not be used
if main/geometry/NOINSTALL is present.  This isn't a very good solution.

Perhaps it is time to rethink the idea of automatically installing every
m-file and oct-file in a directory, and instead go back to explicitly
listing the files which need to be installed.

In the meantime, I'm inclined to live with a non-working delaunay.m
installed when qhull is not present.

- Paul

Re: [OctDev] conftest.cc has problems with tetex-3.0

[R S A. M. <rsa...@sa...>]

Paul Kienzle wrote:

> The problem I believe is that texi2html on your system is leaving an 
> empty conftest directory.  If so, then this patch to configure.base 
> should fix it.  You will need to rerun ./autogen.sh before ./configure 
> after applying this change.  If it does not fix it, then try the same 
> thing in the texi2dvi and makeinfo tests.
>
> Please let us know which one works.
>
> - Paul
>
> diff -c -p -r1.41 configure.base
> *** configure.base      1 Mar 2005 16:33:25 -0000       1.41
> --- configure.base      29 May 2005 14:58:39 -0000
> *************** if [ test -n "$TEXI2HTML" ]; then
> *** 454,459 ****
> --- 454,460 ----
>                 AC_MSG_RESULT(no)
>         fi
>         rm -f conftest.*
> +       rm -rf conftest
>   fi
>
>   dnl Test for the dvipdf program
>
>
> On May 11, 2005, at 10:24 PM, R S Ananda Murthy wrote:
>
>> Hello:
>>
>> I am trying to build octave-forge-2004.11.16 on Slackware 10.1. I 
>> have found that ./configure recognizes all required libraries 
>> properly only when I have tetex-2.0.2 installed on the system. I 
>> removed tetex-2.0.2 and installed tetex-3.0 which is the latest 
>> version. After this, if I try ./configure, it is giving several error 
>> messages. How to correct this? Can we not make ./configure work with 
>> tetex-3.0?
>>
>> Thanks for your help.
>>
>> Anand
>>
>>
>> -------------------------------------------------------
>> This SF.Net email is sponsored by Oracle Space Sweepstakes
>> Want to be the first software developer in space?
>> Enter now for the Oracle Space Sweepstakes!
>> http://ads.osdn.com/?ad_id=7393&alloc_id=16281&op=click
>> _______________________________________________
>> Octave-dev mailing list
>> Oct...@li...
>> https://lists.sourceforge.net/lists/listinfo/octave-dev
>>
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by Yahoo.
> Introducing Yahoo! Search Developer Network - Create apps using Yahoo!
> Search APIs Find out how you can build Yahoo! directly into your own
> Applications - visit 
> http://developer.yahoo.net/?fr=offad-ysdn-ostg-q22005
> _______________________________________________
> Octave-dev mailing list
> Oct...@li...
> https://lists.sourceforge.net/lists/listinfo/octave-dev
>
How do I apply this patch? Can you please explain in a little more detail?

Anand

Re: [OctDev] conftest.cc has problems with tetex-3.0

[Paul K. <pki...@us...>]

The problem I believe is that texi2html on your system is leaving an 
empty conftest directory.  If so, then this patch to configure.base 
should fix it.  You will need to rerun ./autogen.sh before ./configure 
after applying this change.  If it does not fix it, then try the same 
thing in the texi2dvi and makeinfo tests.

Please let us know which one works.

- Paul

diff -c -p -r1.41 configure.base
*** configure.base      1 Mar 2005 16:33:25 -0000       1.41
--- configure.base      29 May 2005 14:58:39 -0000
*************** if [ test -n "$TEXI2HTML" ]; then
*** 454,459 ****
--- 454,460 ----
                 AC_MSG_RESULT(no)
         fi
         rm -f conftest.*
+       rm -rf conftest
   fi

   dnl Test for the dvipdf program


On May 11, 2005, at 10:24 PM, R S Ananda Murthy wrote:

> Hello:
>
> I am trying to build octave-forge-2004.11.16 on Slackware 10.1. I have 
> found that ./configure recognizes all required libraries properly only 
> when I have tetex-2.0.2 installed on the system. I removed tetex-2.0.2 
> and installed tetex-3.0 which is the latest version. After this, if I 
> try ./configure, it is giving several error messages. How to correct 
> this? Can we not make ./configure work with tetex-3.0?
>
> Thanks for your help.
>
> Anand
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by Oracle Space Sweepstakes
> Want to be the first software developer in space?
> Enter now for the Oracle Space Sweepstakes!
> http://ads.osdn.com/?ad_id=7393&alloc_id=16281&op=click
> _______________________________________________
> Octave-dev mailing list
> Oct...@li...
> https://lists.sourceforge.net/lists/listinfo/octave-dev
>

Re: [OctDev] strtok vs column arrays

[Paul K. <pki...@us...>]

In the past my solution to the fread problem has been to append a 
transposition operator ' to the fread call.  Looking at recent matlab 
docs, it looks like you can specify '*char' or 'char=>char'.  I don't 
know if that fixes the orientation problems though.

Do you know if the strtok problem exists a lot of places in the wild?  
If so, we should probably adapt strtok to handle it.

Note that recent versions of strtok on matlab are documented to take 
cell arrays.  Matrices of characters as a representation for a list of 
strings was never a good representation.

- Paul

On May 29, 2005, at 8:59 AM, Tom Holroyd wrote:

> On Sat, 28 May 2005, Paul Kienzle wrote:
>
>> An m x 1 vector of characters should be interpreted as an array of 
>> length one strings.
>
> Good point.  strcmp treats an m x n string array as m strings.
>
> Of course strtok doesn't work for string arrays, either:
>
> octave:4> s = ['abca';'fdag'];
> octave:5> strtok(s, 'a')
> error: mx_el_ne: nonconformant arguments (op1 is 3x1, op2 is 1x3)
>
> With my patch the result is the somewhat non-obvious
>
> octave:1> s = ['abca';'fdag'];
> octave:2> strtok(s, 'a')
> ans = fbdc
>
> That's wrong, too.  It should be ['bc';'fd'], and I guess in the case 
> where the result has different length rows the short ones should be 
> padded with blanks.
>
> One could certainly argue that fread with 'char' should return 
> strings, but since fread is documented as returning columns, for the 
> other types this doesn't make sense.  The default is uchar,
> too.
>
> Perhaps the documentation for fread could include a statement that if 
> you specify 'char', you probably also want to transpose the output, to 
> make it into a "string".  Or a new 'string' type, as opposed to 
> 'char', that does this.
>
> Dr. Tom Holroyd
> "A man of genius makes no mistakes. His errors are volitional and
> are the portals of discovery." -- James Joyce
>

Re: [OctDev] strtok vs column arrays

[Tom H. <to...@ku...>]

On Sat, 28 May 2005, Paul Kienzle wrote:

> An m x 1 vector of characters should be interpreted as an array 
> of length one strings.

Good point.  strcmp treats an m x n string array as m strings.

Of course strtok doesn't work for string arrays, either:

octave:4> s = ['abca';'fdag'];
octave:5> strtok(s, 'a')
error: mx_el_ne: nonconformant arguments (op1 is 3x1, op2 is 1x3)

With my patch the result is the somewhat non-obvious

octave:1> s = ['abca';'fdag'];
octave:2> strtok(s, 'a')
ans = fbdc

That's wrong, too.  It should be ['bc';'fd'], and I guess in the 
case where the result has different length rows the short ones 
should be padded with blanks.

One could certainly argue that fread with 'char' should return 
strings, but since fread is documented as returning columns, for 
the other types this doesn't make sense.  The default is uchar,
too.

Perhaps the documentation for fread could include a statement 
that if you specify 'char', you probably also want to transpose 
the output, to make it into a "string".  Or a new 'string' type, 
as opposed to 'char', that does this.

Dr. Tom Holroyd
"A man of genius makes no mistakes. His errors are volitional and
are the portals of discovery." -- James Joyce

Re: [OctDev] strtok vs column arrays

[Paul K. <pki...@us...>]

This seems to be a request for bug-for-bug compatibility.  An m x 1 
vector of characters should be interpreted as an array of length one 
strings.  The fact that fread is returning the wrong orientation for a 
character array shouldn't require us to make an ugly interface to all 
our string functions.  I'm not going to 'fix' this, but if someone else 
feels so inclined they may.

- Paul

On May 15, 2005, at 5:24 PM, Tom Holroyd wrote:

> fread with 'char' returns a column array; eventually this got passed 
> to strtok.  In strtok, there's a
> 	skip = find(idx != 1:length(idx))
> which failed because 1) there were multiple instances of the 
> delimiter, and 2) idx is a column and the : is a row.
> For example, in matlab:
> >> s = 'abca';
> >> strtok(s, 'a')
>
> ans =
>
> bc
>
> >> strtok(s', 'a')
>
> ans =
>
> b
> c
>
> In octave:
> octave:1> s = 'abca';
> octave:2> strtok(s, 'a')
> ans = bc
> octave:3> strtok(s', 'a')
> error: mx_el_ne: nonconformant arguments (op1 is 2x1, op2 is 1x2)
>
> One way to fix this is:
> --- #strtok.m   2005-05-15 16:18:02.000000000 -0400
> +++ strtok.m    2005-05-15 16:47:53.000000000 -0400
> @@ -57,6 +57,7 @@
>        tok = str;
>        rem = "";
>      else
> +      idx = reshape(idx, [1 length(idx)]);
>        skip = find(idx != 1:length(idx)); # find first non-leading 
> delimiter
>        if isempty(skip)
>         tok = str(idx(length(idx))+1:length(str));
>
> -- 
> Dr. Tom Holroyd
> "A man of genius makes no mistakes. His errors are volitional and
> are the portals of discovery." -- James Joyce
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by Oracle Space Sweepstakes
> Want to be the first software developer in space?
> Enter now for the Oracle Space Sweepstakes!
> http://ads.osdn.com/?ad_id=7393&alloc_id=16281&op=click
> _______________________________________________
> Octave-dev mailing list
> Oct...@li...
> https://lists.sourceforge.net/lists/listinfo/octave-dev
>

Re: [OctDev] meshgrid 3D

[Paul K. <pki...@us...>]

Ben,

Please send this as a context diff (using "diff -c") to bu...@oc....

	- Paul

On May 24, 2005, at 2:14 PM, Ben Barrowes wrote:

> meshgrid did not handle 3-d gridding, here is a simple fix for:
> /usr/local/share/octave/2.9.3/m/plot/meshgrid.m
>
> ...
>
> Is this the right forum for this?
>
> Ben Barrowes
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by Yahoo.
> Introducing Yahoo! Search Developer Network - Create apps using Yahoo!
> Search APIs Find out how you can build Yahoo! directly into your own
> Applications - visit 
> http://developer.yahoo.net/?fr=offad-ysdn-ostg-q22005
> _______________________________________________
> Octave-dev mailing list
> Oct...@li...
> https://lists.sourceforge.net/lists/listinfo/octave-dev
>

[OctDev] meshgrid 3D

[Ben B. <bar...@al...>]

meshgrid did not handle 3-d gridding, here is a simple fix for:
/usr/local/share/octave/2.9.3/m/plot/meshgrid.m

## Copyright (C) 1996, 1997 John W. Eaton
##
## This file is part of Octave.
##
## Octave is free software; you can redistribute it and/or modify it
## under the terms of the GNU General Public License as published by
## the Free Software Foundation; either version 2, or (at your option)
## any later version.
##
## Octave is distributed in the hope that it will be useful, but
## WITHOUT ANY WARRANTY; without even the implied warranty of
## MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
## General Public License for more details.
##
## You should have received a copy of the GNU General Public License
## along with Octave; see the file COPYING.  If not, write to the Free
## Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
## 02110-1301, USA.

## -*- texinfo -*-
## @deftypefn {Function File} {[@var{xx}, @var{yy}] =} meshgrid 
(@var{x}, @var{y})
## @deftypefnx {Function File} {[@var{xx}, @var{yy}] =} meshgrid (@var{x})
## Given vectors of @var{x} and @var{y} coordinates, return two matrices 
corresponding
## to the @var{x} and @var{y} coordinates of a mesh.  The rows of 
@var{xx} are copies of @var{x},
## and the columns of @var{yy} are copies of @var{y}.
## @end deftypefn
##
## @seealso{sombrero, plot, semilogx, semilogy, loglog, polar, mesh,
## meshdom, contour, bar, stairs, replot, xlabel, ylabel, and title}

## Author: jwe

function [xx, yy, zz] = meshgrid (x, y, z)

  if (nargin == 1)
    y = x;
  endif
  if (nargin > 0 && nargin < 3)
    if (isvector (x) && isvector (y))
      xx = ones (length (y), 1) * x(:).';
      yy = y(:) * ones (1, length (x));
    else
      error ("meshgrid: arguments must be vectors");
    end
  elseif nargin==3
   if (isvector (x) && isvector (y) && isvector (z))
      lenx=length(x); leny=length(y); lenz=length(z);
      xx = repmat( ones (leny, 1) * x(:).', [1,1,lenz]);
      yy = repmat( y(:) * ones (1, lenx), [1,1,lenz]);
      zz = reshape(repmat(z(:).',lenx*leny,1)(:),leny,lenx,lenz);
   else
      error ("meshgrid: arguments must be vectors");
    end  
  else
    usage ("[xx, yy, zz] = meshgrid (x, y, z)");
  end
  
endfunction


Is this the right forum for this?

Ben Barrowes

[OctDev] Re: strcmp doesn't return a bool

[John W. E. <jw...@be...>]

On 20-May-2005, Tom Holroyd wrote:

| Here's one that does.  This is based on JWE's strcmp.
| 
| I collapsed the case
| 	strncmp(cell array, string matrix, n)
| to
| 	strncmp(string matrix, cell array, n)
| with the arguments reversed; the same thing could be done to strcmp.

I included the strncmp function in Octave (both 2.9.x and 2.1.x).

| I don't know if it's slower or faster, but it's easier to maintain,
| certainly.  There's a similar possible collapse in the cell, cell
| case, but it looks like it might be a bit slower.

Yes, it is likely to be slower because recursion is slow in Octave.

But if you would like to make a similar change to strcmp.m now to make
it simpler to maintain and more similar to the strncmp function, I
would take the patch.

Thanks,

jwe

Re: [OctDev] Content mail warning notification!

[Quentin S. <qsp...@ie...>]

Tre...@nu... wrote:

>blah blah blah
>
OK, these messages are getting annoying now. If I recall correctly, this 
sort of thing happens when somebody's computer gets infected with 
something. I'm not good at reading headers, but I'm guessing 
someone/something is using smtp.math.keio.ac.jp to send 
spam/viruses/whatever, and lei...@np... has a virus 
scanner that's reporting it back to the list? Is there any way for 
someone with administrative powers over this list to blacklist a 
particular computer or something so we stop getting these?

-Quentin

[OctDev] Re: strcmp doesn't return a bool

[Tom H. <to...@ku...>]

> Here's one that does.  This is based on JWE's strcmp.

Oh, sorry to followup, but I forgot: since strcmp.m is part of octave, strncmp.m 
should probably be part of octave, too; the strncmp.m in octave-forge should go 
away, now.

-- 
Dr. Tom Holroyd
"A man of genius makes no mistakes. His errors are volitional and
are the portals of discovery." -- James Joyce

[OctDev] Re: strcmp doesn't return a bool

[Tom H. <to...@ku...>]

Tom Holroyd wrote:
> The strncmp implementation in octave-forge doesn't handle cell arrays at
> all.

Here's one that does.  This is based on JWE's strcmp.

I collapsed the case
	strncmp(cell array, string matrix, n)
to
	strncmp(string matrix, cell array, n)
with the arguments reversed; the same thing could be done to strcmp.  I don't 
know if it's slower or faster, but it's easier to maintain, certainly.  There's 
a similar possible collapse in the cell, cell case, but it looks like it might 
be a bit slower.  I didn't time it, it just looks that way.  That one's minor 
anyway.

Of course these are both popular functions and as one person has already 
mentioned today are good candidates for builtins.  These guys are both different 
from C's str*cmp, so at least, the builtin can follow the logic here, which is 
compatible and correct, as far as I know.
-- 
Dr. Tom Holroyd
"A man of genius makes no mistakes. His errors are volitional and
are the portals of discovery." -- James Joyce

[OctDev] Re: print.m no longer working

[Dmitri A. S. <das...@gm...>]

On 5/19/05, Joe Koski <jko...@co...> wrote:
> on 5/19/05 9:50 AM, Michael Creel at mic...@ua... wrote:
>=20
> > Hello all,
> > I'm trying to use print to create some .eps files. It's no longer worki=
ng:
> >
> > octave:1> plot(rand(3,1))
> > octave:2> print("test.eps","-depsc2")
....

> On my Mac with octave-2.1.69 and octave-forge, I got the error message "c=
an
> only plot in 2 or 3 dimensions" when I tried to update my print and legen=
d
> to use the __gnuplot_replot__ routine in order to eliminate the warning
...

FWIW -- the current print.m from the recent octave-forge works fine
with octave-2.1.71.
[dima@tumbleweed octave]$ octave
GNU Octave, version 2.1.71 (i686-pc-linux-gnu).
Copyright (C) 2005 John W. Eaton.
....
octave:1> plot(rand(3,1))
octave:2> print("test.eps","-depsc2")
warning: in fopen near line 55, column 3:
warning: fopen: default open mode is now binary
octave:3> quit

The test.eps file is on
ftp://coffee.phys.unm.edu/pub/dima/octave/test.eps

I am using a recent gnuplot-4.1 cvs snapshot, but I doubt it matters.
=20
Sincerely,

Dmitri.

[OctDev] Re: print.m no longer working

[Joe K. <jko...@co...>]

on 5/19/05 9:50 AM, Michael Creel at mic...@ua... wrote:

> Hello all,
> I'm trying to use print to create some .eps files. It's no longer working:
> 
> octave:1> plot(rand(3,1))
> octave:2> print("test.eps","-depsc2")
> warning: in fopen near line 55, column 3:
> warning: fopen: default open mode is now binary
> error: plot: can only plot in 2 or 3 dimensions
> error: plot: can only plot in 2 or 3 dimensions
> error: called from `print' in file
> `/usr/share/octave/2.1.69/site/m/octave-forge/plot/print.m'
> 
> I guess that print.m needs some updating, if anyone knows what needs to be
> done.
> Thanks, M.       

On my Mac with octave-2.1.69 and octave-forge, I got the error message "can
only plot in 2 or 3 dimensions" when I tried to update my print and legend
to use the __gnuplot_replot__ routine in order to eliminate the warning
messages associated with use of gset, which is now deprecated, but used by
the older versions of print and legend. If I use the print and legend that
came with the Nov. 2004 octave-forge, I get warning messages, but everything
works. Have you tried to update print to use the octave-2.9.x versions of
print?

Joe