octave-dev — GNU Octave developers' list

[OctDev] becoming developer

[sissou <syl...@gm...>]

Hello,

Can someone add me as developper please, my sourceforge user id is sis-sou.
Thanks.

Re: [OctDev] Using the package system

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

S=C3=B8ren Hauberg wrote:

>Hi,
>
>tir, 08 08 2006 kl. 08:35 -0500, skrev Quentin Spencer:
> =20
>
>>Paul Kienzle wrote:
>>   =20
>>
>>>If the current octave-forge works with 2.9.7 then you can start checki=
ng
>>>in changes now.  Anybody tried it?  =20
>>>     =20
>>>
>>Yes, I have released the current octave-forge in Fedora Extras with=20
>>octave 2.9.7. The only problem so far is a small bug in legend.m, which=
=20
>>I have fixed in CVS.
>>   =20
>>
>So, I can start making changes in CVS? I need to be sure everybody is
>fine with this because I need to make some minor changes in the
>file/directory layout, that will break the build system.
>
>The package system assumes that
>* Files that can be directly installed (e.g. *.m files) are in a package
>subdirectory called 'inst'.
>* Files that need to be compiled are in a package subdirectory called
>'src'.
>
>So are everybody okay with me moving files around? If yes, then things
>will be broken for a while...
> =20
>

It's generally fine with me as long as the last release of octave-forge=20
still compiles with current 2.9.x octave versions. If there is a new=20
octave release that requires significant changes in octave-forge I will=20
want to see an octave-forge release. However, I think that's not so=20
likely since 2.9.x seems to be stabilizing a bit more now. So, unless=20
others object, I'd say go ahead and break things as long as they don't=20
stay broken for too long.

Quentin

Re: [OctDev] Using the package system

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

Hi,

tir, 08 08 2006 kl. 08:35 -0500, skrev Quentin Spencer:
> Paul Kienzle wrote:
> > If the current octave-forge works with 2.9.7 then you can start check=
ing
> > in changes now.  Anybody tried it?  =20
>=20
> Yes, I have released the current octave-forge in Fedora Extras with=20
> octave 2.9.7. The only problem so far is a small bug in legend.m, which=
=20
> I have fixed in CVS.
So, I can start making changes in CVS? I need to be sure everybody is
fine with this because I need to make some minor changes in the
file/directory layout, that will break the build system.

The package system assumes that
* Files that can be directly installed (e.g. *.m files) are in a package
subdirectory called 'inst'.
* Files that need to be compiled are in a package subdirectory called
'src'.

So are everybody okay with me moving files around? If yes, then things
will be broken for a while...

S=C3=B8ren


> Quentin
>=20
>=20
>=20

[OctDev] fill function

[sissou <syl...@gm...>]

Hello,

I want use the octave's fill function but the result is very slow and
not beautiful. Why the fill function doesn't use the gnuplot's
filledcurve option
 like (without color) :

/function [] = myfill(x,y)
__gnuplot_set__("style data filledcurve");
plot(x,y);
__gnuplot_set__("style data lines");

/I can't make a fill function with colors but if someone can make a
better fill function.

Re: [OctDev] Becoming a developer?

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

On Aug 8, 2006, at 6:57 AM, S=F8ren Hauberg wrote:

> Okay. My sourceforge id is "hauberg" -- could somebody add me as a
> developer?

Done.

- Paul
>

Re: [OctDev] Using the package system

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

Paul Kienzle wrote:
> On Aug 7, 2006, at 1:34 PM, S=F8ren Hauberg wrote:
>
>  =20
>> s=F8n, 06 08 2006 kl. 07:27 -0400, skrev Paul Kienzle:
>>    =20
>>> If you needed to change the package manager in order to make it work=20
>>> in
>>> octave 2.9.7 then we will have to wait for the 2.9.8 release.
>>>      =20
>> Wait before we make an octave-forge release, or wait before we check
>> changes into cvs?
>>    =20
>
> If the current octave-forge works with 2.9.7 then you can start checkin=
g
> in changes now.  Anybody tried it?
>  =20

Yes, I have released the current octave-forge in Fedora Extras with=20
octave 2.9.7. The only problem so far is a small bug in legend.m, which=20
I have fixed in CVS.

Quentin

Re: [OctDev] Becoming a developer?

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

Okay. My sourceforge id is "hauberg" -- could somebody add me as a
developer?

tir, 08 08 2006 kl. 05:38 -0400, skrev Paul Kienzle:
> Send your sourceforge user id to this list and somebody will add you.
>=20
> See the developers guide on http://octave.sf.net for details.
>=20
> 	- Paul
>=20
> On Aug 7, 2006, at 1:37 PM, S=C3=B8ren Hauberg wrote:
>=20
> > Hi,
> >   How do I get developers (write) access to the octave-forge cvs
> > repository?
> >
> > S=C3=B8ren
> >
> >
> >
> > ---------------------------------------------------------------------=
--=20
> > --
> > Using Tomcat but need to do more? Need to support web services, =20
> > security?
> > Get stuff done quickly with pre-integrated technology to make your jo=
b =20
> > easier
> > Download IBM WebSphere Application Server v.1.0.1 based on Apache =20
> > Geronimo
> > http://sel.as-us.falkag.net/sel?=20
> > cmd=3Dlnk&kid=3D120709&bid=3D263057&dat=3D121642
> > _______________________________________________
> > Octave-dev mailing list
> > Oct...@li...
> > https://lists.sourceforge.net/lists/listinfo/octave-dev
>=20
>=20
>=20

Re: [OctDev] Becoming a developer?

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

Send your sourceforge user id to this list and somebody will add you.

See the developers guide on http://octave.sf.net for details.

	- Paul

On Aug 7, 2006, at 1:37 PM, S=F8ren Hauberg wrote:

> Hi,
>   How do I get developers (write) access to the octave-forge cvs
> repository?
>
> S=F8ren
>
>
>
> =
-----------------------------------------------------------------------=20=

> --
> Using Tomcat but need to do more? Need to support web services, =20
> security?
> Get stuff done quickly with pre-integrated technology to make your job =
=20
> easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache =20
> Geronimo
> http://sel.as-us.falkag.net/sel?=20
> cmd=3Dlnk&kid=3D120709&bid=3D263057&dat=3D121642
> _______________________________________________
> Octave-dev mailing list
> Oct...@li...
> https://lists.sourceforge.net/lists/listinfo/octave-dev

Re: [OctDev] Using the package system

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

On Aug 7, 2006, at 1:34 PM, S=F8ren Hauberg wrote:

> s=F8n, 06 08 2006 kl. 07:27 -0400, skrev Paul Kienzle:
>> If you needed to change the package manager in order to make it work=20=

>> in
>> octave 2.9.7 then we will have to wait for the 2.9.8 release.
> Wait before we make an octave-forge release, or wait before we check
> changes into cvs?

If the current octave-forge works with 2.9.7 then you can start checking
in changes now.  Anybody tried it?

Otherwise wait for a new release.

>
>> The octave-forge build/release infrastructure will need to be changed
>> to make the packages available individually and as one large =
download.
> When you say one large download, do you mean a large package (to be=20
> used
> with the package manager) or a download that is installed in the=20
> regular
>   ./configure; make; make install
> way?

I mean some way to download and install a bundle of packages all at=20
once.
Presumably this would be a simple script that walks through every=20
package
in the tree and asked it to confirm its prerequisites, build, test, and
perhaps install while collecting a summary of the results to present to
the user when the process is complete.

>
>> A convenience installer allowing the user to select the list of=20
>> packages
>> to install would be nice.
> Agreed, but that's a not a problem that needs solving right now.
>
>> The extra and nonfree directories will need
>> to be packaged as well.
> This should be trivial (1-2 hours of work) and I'll do it soon.
>
>> Any volunteers?
> I'd love to help out, but I really don't know anything about =
autotools,
> and I have a feeling that such skills will be needed.

There shouldn't be any global autotools.  It should only be a matter of
asking each package to detect whether it's dependencies are met.  This
is currently encoded in the configure.add file.  There are some global
checks done in the top level configure which hopefully are no longer
needed.

- Paul=

[OctDev] Becoming a developer?

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

Hi,
  How do I get developers (write) access to the octave-forge cvs
repository?

S=C3=B8ren

Re: [OctDev] Using the package system

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

s=C3=B8n, 06 08 2006 kl. 07:27 -0400, skrev Paul Kienzle:
> If you needed to change the package manager in order to make it work in
> octave 2.9.7 then we will have to wait for the 2.9.8 release.
Wait before we make an octave-forge release, or wait before we check
changes into cvs?

> The octave-forge build/release infrastructure will need to be changed
> to make the packages available individually and as one large download.
When you say one large download, do you mean a large package (to be used
with the package manager) or a download that is installed in the regular
  ./configure; make; make install
way?

> A convenience installer allowing the user to select the list of package=
s
> to install would be nice.=20
Agreed, but that's a not a problem that needs solving right now.

> The extra and nonfree directories will need
> to be packaged as well.
This should be trivial (1-2 hours of work) and I'll do it soon.

> Any volunteers?
I'd love to help out, but I really don't know anything about autotools,
and I have a feeling that such skills will be needed.

S=C3=B8ren

Re: [OctDev] Using the package system

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

On Aug 6, 2006, at 5:04 AM, S=F8ren Hauberg wrote:
> So moving to the package system should be a trivial task, and we =
simply
> need to decide when we do this.

If you needed to change the package manager in order to make it work in
octave 2.9.7 then we will have to wait for the 2.9.8 release.

The octave-forge build/release infrastructure will need to be changed
to make the packages available individually and as one large download.
A convenience installer allowing the user to select the list of packages
to install would be nice. The extra and nonfree directories will need
to be packaged as well.

Any volunteers?

	- Paul

Re: [OctDev] Using the package system

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

I did a CVS checkout this friday, and I have now moved main to the
package system. It was a fairly easy task that just required some manual
work. I'm having (configure) problems with comm, fixed, and
linear-algebra, but I also saw these errors when building the entire
octave-forge.

I would love to make the packages available to the rest of you, but my
webserver is giving me a hard time so I don't have a place to publish
the packages. Does anybody have a place where I can upload the packages?

So moving to the package system should be a trivial task, and we simply
need to decide when we do this.

S=C3=B8ren

ons, 02 08 2006 kl. 21:53 +0200, skrev S=C3=B8ren Hauberg:
> Hi,
>   Hopefully I'll get some spare time in a couple of days, so I'd like t=
o
> start converting Octave Forge to the package system which is included i=
n
> Octave 2.9.6 and later.
>   Is anything happening in Octave Forge currently that I should have in
> mind, when doing this transition?=20
>   Should I just take a CVS snapshot and use that as my basis, or should
> I do something else?
>=20
> S=C3=B8ren
>=20
>=20
>=20
> -----------------------------------------------------------------------=
--
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to share=
 your
> opinions on IT & business topics through brief surveys -- and earn cash
> http://www.techsay.com/default.php?page=3Djoin.php&p=3Dsourceforge&CID=3D=
DEVDEV
> _______________________________________________
> Octave-dev mailing list
> Oct...@li...
> https://lists.sourceforge.net/lists/listinfo/octave-dev

Re: [OctDev] Using the package system

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

On Aug 2, 2006, at 3:53 PM, S=F8ren Hauberg wrote:

> Hi,
>   Hopefully I'll get some spare time in a couple of days, so I'd like=20=

> to
> start converting Octave Forge to the package system which is included=20=

> in
> Octave 2.9.6 and later.
>   Is anything happening in Octave Forge currently that I should have =
in
> mind, when doing this transition?
>   Should I just take a CVS snapshot and use that as my basis, or =
should
> I do something else?

It's all clear from my end.

Don't bother with extra/mex --- it is going away.

- Paul

[OctDev] Using the package system

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

Hi,
  Hopefully I'll get some spare time in a couple of days, so I'd like to
start converting Octave Forge to the package system which is included in
Octave 2.9.6 and later.
  Is anything happening in Octave Forge currently that I should have in
mind, when doing this transition?=20
  Should I just take a CVS snapshot and use that as my basis, or should
I do something else?

S=C3=B8ren

Re: [OctDev] Multi-threaded optimal interpolation module for octave

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

on 7/10/06 7:46 AM, Alexander Barth at ab...@ma... wrote:

> Hi all,
> 
> I wrote a Fortran 90 module performing a n-dimensional local optimal
> interpolation. The optimal interpolation allows to interpolate
> arbitrarily located observations to a regular grid using a background
> field as first guess. It can be used as an alternative to griddata. The
> optimal interpolated module can be called from Octave using a C++ wrapper.
> 
> I was able to create a working oct-file with gfortran, g95, Intel
> Fortran Compiler (free as in beer) and PGI Compiler (commercial) .
> Unfortunately, gfortran version 4.0 doesn't work but version 4.1 works fine.
> 
> It is also possible to use multiple CPUs during the optimal
> interpolation in octave if the Fortran code is compiled with OpenMP
> support enabled. This works with the Intel and PGI compilers. OpenMP
> support for GCC is planned for version 4.2.
> 
> Instructions and the source code can be found at:
> http://ocgmod1.marine.usf.edu/OI
> 
> Suggestions are very welcome.
> 
> Cheers
> Alex

Alex,

I downloaded your OI-0.1 package to give it a spin. My octave-2.1.73 and
octave-forge are both built with g95, so I thought I was ready for a
straight mkoctfile optiminterp.cc. This was on my Mac G5 with OS X 10.3.9.
To be sure, I rebuilt octave and octave-forge with the most recent (7/10/06)
g95 for the Mac. (Andy Vaught recently modified his name mangling approach
for g95, so I thought it prudent to rebuild octave.)

When I run mkoctfile, I got the following:

Joe-Koskis-Computer:/Tools/OI-0.1 jakoski$ mkoctfile optiminterp.cc
ld: warning multiple definitions of symbol _xerbla_
/usr/local/lib/octave-2.1.73/liboctinterp.dylib(single module) definition of
_xerbla_
/usr/local/lib/octave-2.1.73/libcruft.dylib(single module) definition of
_xerbla_
/System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib
.framework/Versions/A/libBLAS.dylib(single module) definition of _xerbla_
ld: Undefined symbols:
_optiminterp_f77_

The multiple definitions of xerbla are normal, but why do I need
_optiminterp_f77_ with a Fortran 95 compiler? Are we having problems with .F
vs. .f as an extension? Macs are not as case sensitive as Linux, although I
tried renaming optimal_interpolation.F to optimal_interpolation.f90, and
optiminterp_f77.f to a dummy without success.

In my OI-0.1 directory I have (after running mkoctfile)

Joe-Koskis-Computer:/Tools/OI-0.1 jakoski$ ls
AUTHORS                         optiminterp.cc
COPYING                         optiminterp.o
README.html                     optiminterp1.m
example_optiminterp.f           optiminterp2.m
example_optiminterp.m           optiminterp3.m
ndgrid.m                        optiminterp_f77.f_unused
optimal_interpolation.F         test_optiminterp.m
optimal_interpolation.f90

Any ideas?

Joe

[OctDev] Multi-threaded optimal interpolation module for octave

[Alexander B. <ab...@ma...>]

Hi all,

I wrote a Fortran 90 module performing a n-dimensional local optimal 
interpolation. The optimal interpolation allows to interpolate 
arbitrarily located observations to a regular grid using a background 
field as first guess. It can be used as an alternative to griddata. The 
optimal interpolated module can be called from Octave using a C++ wrapper.

I was able to create a working oct-file with gfortran, g95, Intel 
Fortran Compiler (free as in beer) and PGI Compiler (commercial) . 
Unfortunately, gfortran version 4.0 doesn't work but version 4.1 works fine.

It is also possible to use multiple CPUs during the optimal 
interpolation in octave if the Fortran code is compiled with OpenMP 
support enabled. This works with the Intel and PGI compilers. OpenMP 
support for GCC is planned for version 4.2.

Instructions and the source code can be found at:
http://ocgmod1.marine.usf.edu/OI

Suggestions are very welcome.

Cheers
Alex

-- 
_______________________________________________________________

  Alexander Barth

  Ocean Circulation Group
  University of South Florida
  College of Marine Science
  140 Seventh Avenue South 
  St. Petersburg, Florida  33701
  USA

  Phone: +1-727-553-3508     FAX:   +1-727-553-1189

_______________________________________________________________

[OctDev] test release of octave-forge for 2.9.6

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

Hi all,

On the basis of reports of successful builds of the CVS version
of octave-forge, I created a new release.  It may be several
hours before it is available for download.  Please check that
it works on your platform.  If I don't hear anything bad in the
next couple of days I will make an announcement on he...@oc....

- Paul

Re: [OctDev] password (done)

[Etienne G. <et...@cs...>]

  Done

On Mon, Jul 03, 2006 at 03:43:41PM -0500, Jacks, Adam wrote:
# Hi there,
# 
#  
# 
# I'd like to have access to the octave wiki discussion group.  Thanks!
# 
#  
# 
# Adam
# 
#  
# 
# ____________________________________________________
# 
#  
# 
# Adam Jacks, Ph.D., CCC-SLP
# 
# Postdoctoral fellow
# 
# Research Imaging Center
# 
# The University of Texas Health Science Center at San Antonio
# 
# ja...@ut...
# 
# (210) 567-8132
# 
# ____________________________________________________
# 
#  
# 

# Using Tomcat but need to do more? Need to support web services, security?
# Get stuff done quickly with pre-integrated technology to make your job easier
# Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
# http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
# _______________________________________________
# Octave-dev mailing list
# Oct...@li...
# https://lists.sourceforge.net/lists/listinfo/octave-dev


-- 
Etienne Grossmann ------ http://www.cs.uky.edu/~etienne

Re: [OctDev] password

[Jacks, A. <ja...@ut...>]

Hi there,

=20

I'd like to have access to the octave wiki discussion group.  Thanks!

=20

Adam

=20

____________________________________________________

=20

Adam Jacks, Ph.D., CCC-SLP

Postdoctoral fellow

Research Imaging Center

The University of Texas Health Science Center at San Antonio

ja...@ut...

(210) 567-8132

____________________________________________________

=20

Re: [OctDev] Build errors [was: Re: New release?]

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

Hi,
  Sorry about the late reply. So I've been experiencing some problems
build octave-forge, but I've been able to hack around them.

My problems with building fixed, and linear-algebra comes from the
following lines in Makeconf:

%.o: %.c ; $(MKOCTFILE) -c $<
%.o: %.f ; $(MKOCTFILE) -c $<
%.o: %.cc ; $(MKOCTFILE) -c $<
%.oct: %.cc ; $(MKOCTFILE) $<

These appear to overwrite the corresponding rules in the Makefiles in
fixed and linear-algebra. If I remove them I can build fixed, but then I
can't build extra/pdb, signal, image, io, linear-algebra, optim,
specfun, and symbolic. So I just added the lines I removed from Makeconf
to the (bottom of the) Makefiles of the packages that wouldn't build,
and then things got build (and all test passed on Ubuntu).

I'm not an autotools expert so I don't know what the solution is to the
problem.

S=C3=B8ren

Re: [OctDev] Mac tests? (Re: New release?)

[David B. <Dav...@mo...>]

Paul Kienzle wrote:

>On Jun 29, 2006, at 5:20 PM, Quentin Spencer wrote:
>
>  
>
>>Thomas Weber wrote:
>>    
>>
>>>Hi,
>>>
>>>Am Donnerstag, den 22.06.2006, 09:02 +0200 schrieb Thomas Weber:
>>>
>>>      
>>>
>>>>Sorry, there are (Debian unstable, gcc-4.1):
>>>>
>>>>        
>>>>
>>>The latest CVS version passes all tests (thanks David :) ). This is 
>>>gcc
>>>4.1, Octave 2.9.6 on Debian AMD64. I will try a build on i386 
>>>tomorrow.
>>>
>>>      
>>>
>>I can confirm that it passes all tests on Fedora Core 5, octave 2.9.6,
>>gcc 4.1.1 on i386. As soon as we have a new octave-forge release, I 
>>will
>>release 2.9.6 and corresponding o-f for Fedora.
>>    
>>
>
>Has anyone confirmed that the latest release will build and pass tests
>on OS X?
>
>Thanks,
>
>	- Paul
>  
>
There is a seg-fault issue that Marius Schamschula  reported. I'm having
trouble tracking down exactly which part of the sparse code in octave is
causing the seg-fault, and at the moment have given him a new version of
test.m with fflush'es before every fprintf to ensure all messages are
flushed prior to the test. I think a similar change should be made in
the CVS..

However, this appears to be independent of octave-forge and so I'm not
sure it should delay an octave-forge release.

D.


-- 
David Bateman                                Dav...@mo...
Motorola Labs - Paris                        +33 1 69 35 48 04 (Ph) 
Parc Les Algorithmes, Commune de St Aubin    +33 6 72 01 06 33 (Mob) 
91193 Gif-Sur-Yvette FRANCE                  +33 1 69 35 77 01 (Fax) 

The information contained in this communication has been classified as: 

[x] General Business Information 
[ ] Motorola Internal Use Only 
[ ] Motorola Confidential Proprietary

[OctDev] Mac tests? (Re: New release?)

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

On Jun 29, 2006, at 5:20 PM, Quentin Spencer wrote:

> Thomas Weber wrote:
>> Hi,
>>
>> Am Donnerstag, den 22.06.2006, 09:02 +0200 schrieb Thomas Weber:
>>
>>> Sorry, there are (Debian unstable, gcc-4.1):
>>>
>>
>> The latest CVS version passes all tests (thanks David :) ). This is 
>> gcc
>> 4.1, Octave 2.9.6 on Debian AMD64. I will try a build on i386 
>> tomorrow.
>>
>
> I can confirm that it passes all tests on Fedora Core 5, octave 2.9.6,
> gcc 4.1.1 on i386. As soon as we have a new octave-forge release, I 
> will
> release 2.9.6 and corresponding o-f for Fedora.

Has anyone confirmed that the latest release will build and pass tests
on OS X?

Thanks,

	- Paul

Re: [OctDev] New release?

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

Thomas Weber wrote:
> Hi, 
>
> Am Donnerstag, den 22.06.2006, 09:02 +0200 schrieb Thomas Weber:
>   
>> Sorry, there are (Debian unstable, gcc-4.1):
>>     
>
> The latest CVS version passes all tests (thanks David :) ). This is gcc
> 4.1, Octave 2.9.6 on Debian AMD64. I will try a build on i386 tomorrow.
>   

I can confirm that it passes all tests on Fedora Core 5, octave 2.9.6, 
gcc 4.1.1 on i386. As soon as we have a new octave-forge release, I will 
release 2.9.6 and corresponding o-f for Fedora.

Quentin

Re: [OctDev] New release?

[Thomas W. <tho...@gm...>]

Hi, 

Am Donnerstag, den 22.06.2006, 09:02 +0200 schrieb Thomas Weber:
> Sorry, there are (Debian unstable, gcc-4.1):

The latest CVS version passes all tests (thanks David :) ). This is gcc
4.1, Octave 2.9.6 on Debian AMD64. I will try a build on i386 tomorrow.


Thanks
	Thomas