octave-dev — GNU Octave developers' list

Re: [OctDev] signal package: possible bug on fir2 with patch (please review)

[Carnë D. <car...@gm...>]

On 9 August 2012 03:11, Mike Miller <mtm...@ie...> wrote:
> On Fri, Jul 27, 2012 at 10:14 AM, Carnë Draug wrote:
>> On 6 July 2012 03:11, Mike Miller <mtm...@ie...> wrote:
>>>
>>> The only thing that is not clear to me is whether this matches ML
>>> behavior, and whether we care, e.g. if the caller passes in a 4th arg
>>> too small, does ML generate an error or silently make it larger like
>>> this line 29 does?
>>
>> MatLab gives an error in somes cases and silently makes it larger in others.
>>
>>>> f = [0 0.6 0.6 1]; m = [1 1 0 0];
>>>> b5 = fir2(30,f,m, 5);
>> Error using fir2 (line 135)
>> The number of grid points npt must be greater than or equal to 16 for
>> filter order 30.
>>
>>>> b8 = fir2(30,f,m, 8);
>> Error using fir2 (line 135)
>> The number of grid points npt must be greater than or equal to 16 for
>> filter order 30.
>>
>>>> b9 = fir2(30,f,m, 9);
>>>> b15 = fir2(30,f,m, 15);
>>>> b16 = fir2(30,f,m, 16);
>>>> b17 = fir2(30,f,m, 17);
>>>> isequal (b9, b15)
>>
>> ans =
>>
>>      1
>>
>>>> isequal (b15, b16)
>>
>> ans =
>>
>>      1
>>
>>>> isequal (b16, b17)
>>
>> ans =
>>
>>      0
>>
>>
>> Seems that MatLab behaviour is to adjust number of points (to the
>> minimum), if the value is more than half of it and error otherwise.
>>
>> Not sure what's the best behavior but I'd guess at least a warning
>> should be given.
>
> Finally got back to this, it's now closer to what you showed here.
> It's clear that Matlab is forcing the 4th argument to always be a
> power of 2, even though that's not in the help text. I implemented
> that, added an error message for the case of the argument being too
> small (after conversion to a power of 2), and added a test case based
> on your examples above. Thanks for those!

Hi Mike

I have just noticed that this introduced another bug (which I think
was already there before, this just made it more apparent. For
example, if the minimum value for 4th argument is more than the
default, it also automatically (and silently) increases though I'm not
sure for how much. For example:

>> f = [0 0.1 0.1 0.2 0.2 1]; m = [0 0 1 1 0 0];
>> bdef = fir2(1024, f, m);

should make the 4th argument default to 512 which would be too low and
return an error (current behaviour with Octave). However, this does
not happen in Matlab. But if the default value is specified, then it
gives an error. And it doesn't default to the next possible value
(1024) as I first expected, it defaults to 2048.

>> b512 = fir2(1024, f, m, 512);
Error using fir2 (line 135)
The number of grid points npt must be greater than or equal to 513 for
filter order 1024.

>> b1024 = fir2(1024, f, m, 1024);
>> b2048 = fir2(1024, f, m, 2048);
>> isequal (bdef, b1024)

ans =

     0

>> isequal (bdef, b2048)

ans =

     1

I don't know exactly what is used to calculate the defaults but seems
to be the next above it. For example:
>> bdef = fir2(2048, f, m);
>> b2048 = fir2(2048, f, m, 2048);
>> b4096 = fir2(2048, f, m, 4096);
>> isequal (bdef, b2048)

ans =

     0

>> isequal (bdef, b4096)

ans =

     1


Let me know if you need more tests.

Carnë

[OctDev] Referenced function that does not exist in documentation of geometry package

[Coelho <coe...@gm...>]

Hi,

There is a function called polygonCentroid that is referenced in
centroid<http://octave.sourceforge.net/geometry/function/centroid.html>function
description and does not exists in documentation and neither in
geometry package.

Thanks.

-- 
Thiago Coelho Prado

[OctDev] openmpi_ext

[c. <car...@gm...>]

Hi,

It seems from a few recent posts on the list that openmpi_ext is getting some
attention lately, so I decided to have a look at its current status.

It seems that although it received very little maintainace in the last three years
it is still in very good shape and is still useful.

I went through the code and did a little cleanup, fixed some trivial bugs and removed
a lot of code duplication making use of templates and macros where possible.

Does anyone object if I apply the attached patch and prepare a new release?

c.

Re: [OctDev] Current MPI-2 support in Octave?

[c. <car...@gm...>]

On 13 Aug 2012, at 14:18, Dmitry S. Kravtsov wrote:

> Hello everybody,
> 
> I was asked to do some research with MPI tools for Octave and current situation seems pretty strange to me.
> Am I right, that best solution now is "openmpi-ext" package from Forge? The alternative solution, as google suggests,
> is "mpitb for octave", but it looks dead for 4 years already. "openmpi-ext" looks not only to be OpenMPI specific, but also unmature (according to API reference).
> Is there any better solutions for Octave, or solutions, that aren't tied directly with OpenMPI (say if I want to use MPICH2)?
> 
> -- 
> Dmitry S. Kravtsov

I just had a look at the openmpi_ext, the code looks quite simple and I don't think there is much to be modified there to make it work with mpich ...
c. 

[OctDev] Current MPI-2 support in Octave?

[Dmitry S. K. <idk...@gm...>]

Hello everybody,

I was asked to do some research with MPI tools for Octave and current
situation seems pretty strange to me.
Am I right, that best solution now is "openmpi-ext" package from Forge? The
alternative solution, as google suggests,
is "mpitb for octave", but it looks dead for 4 years already. "openmpi-ext"
looks not only to be OpenMPI specific, but also unmature (according to API
reference).
Is there any better solutions for Octave, or solutions, that aren't tied
directly with OpenMPI (say if I want to use MPICH2)?

-- 
Dmitry S. Kravtsov

[OctDev] Fem PLotting: fpl 1.3.1 released

[Carnë D. <car...@gm...>]

Hi everyone

a new release of fpl package is out, version 1.3.1, by Carlo de Falco.
Enjoy Octave responsibly.

Carnë

Re: [OctDev] Alpha release of Serial I/O package (v0.1)

[Juan P. C. <car...@if...>]

On Thu, Aug 9, 2012 at 4:02 AM, Andrius Sutas <and...@gm...> wrote:
> Bug with reopening closed interface fixed in r10845. Could someone please
> test package in Unix/Mac environment?
>
> On Tue, Aug 7, 2012 at 9:37 AM, Juan Pablo Carbajal <car...@if...>
> wrote:
>>
>> On Tue, Aug 7, 2012 at 4:37 AM, Andrius Sutas <and...@gm...>
>> wrote:
>> > Hey,
>> >
>> > please see alpha release of Serial I/O package.
>> >
>> > I am looking for any comments on code style, usability, naming etc.
>> > Also, it
>> > would be great if people could reply on which platforms they
>> > successfully
>> > (and not, which is more important) tested it. Sorry but Windows is not
>> > yet
>> > supported.
>> >
>> > Currently open flags are hardcoded to: O_RDWR | O_NOCTTY | O_SYNC, as I
>> > yet
>> > to decide what is the best way to maintain cross-platformability and
>> > ease of
>> > use (e.g. using string for parity setting and then doing switch
>> > according to
>> > platform being used instead of making the same flags from different
>> > platforms global to Octave), comments are welcome.
>> >
>> > Also, a simple example to get started (for testing adapter's RX and TX
>> > lines
>> > are connected together):
>> >
>> > s0 = serial() # Opens default serial port ttyUSB0 in default
>> > configuration
>> > of 115200, 8-N-1
>> >
>> > s1 = serial("/dev/ttyUSB1", 115200) # Opens serial port ttyUSB1 with
>> > baudrate of 115200 (config defaults to 8-N-1)
>> > srl_flush(s1); # Flush input and output buffers
>> > srl_write(s1, "Hello world!") # Blocking write call, currently only
>> > accepts
>> > strings
>> > data = srl_read(s1, 11)  # Blocking read call, returns uint8 array of
>> > exactly 11 bytes read (data = [72  101  108  108  111   32  119  111
>> > 114
>> > 108  100])
>> >
>> > char(data) # Converts uint8 array to string, (ans = "Hello world", note:
>> > no
>> > exclamation mark)
>> >
>> > srl_baudrate(s1, 9600) # Change baudrate
>> > srl_bytesize(s1, 5) # Change byte size (config becomes 5-N-1)
>> > srl_parity(s1, "E") # Changes parity checking (config becomes 5-E-1),
>> > possible values [E]ven, [O]dd, [N]one.
>> > srl_stopbits(s1, 2) # Changes stop bits (config becomes 5-E-2), possible
>> > values 1, 2.
>> >
>> > s2 = serial("/dev/ttyS0", 9600, 6, "odd", 2) # Opens serial port ttyS0
>> > in
>> > 9600, 6-O-2 configuration
>> >
>> > srl_close(s0) # Closes and releases file descriptor
>> >
>> >
>> > ------------------------------------------------------------------------------
>> > Live Security Virtual Conference
>> > Exclusive live event will cover all the ways today's security and
>> > threat landscape has changed and how IT managers can respond.
>> > Discussions
>> > will include endpoint security, mobile security and the latest in
>> > malware
>> > threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
>> > _______________________________________________
>> > Octave-dev mailing list
>> > Oct...@li...
>> > https://lists.sourceforge.net/lists/listinfo/octave-dev
>> >
>>
>> Hi Andrius,
>>
>> Thanks for the first release!
>>
>> When installing on Ubuntu 10.10, gcc version 4.4.5 (Ubuntu/Linaro
>> 4.4.4-14ubuntu5.1) I get the following deprecation warning
>>
>> srl_read.cc: In function ‘octave_value_list Fsrl_read(const
>> octave_value_list&, int)’:
>> srl_read.cc:92: warning: ‘void Array<T>::resize(octave_idx_type) [with
>> T = octave_int<unsigned char>]’ is deprecated (declared at
>> /usr/local/include/octave-3.6.2/octave/../octave/Array.h:459)
>>
>> but I guess this is something in core.
>>
>> I have no usb-to-serial at hand (I am away from my lab) but I tested
>> using virtual loopback using pseudo terminals.
>> (for the ones who wan to try: In linux install "socat" and then run
>> "socat -d -d PTY: PTY:")
>> The following lines work as expected
>>
>> pkg load serial
>> s1 = serial("/dev/ptmx", 115200);
>> srl_flush(s1);
>> srl_write(s1, "Hello world!")
>> data = srl_read(s1,12);
>> char(data)
>> srl_close(s1)
>>
>> However when run again I get
>>
>> s1 = serial("/dev/ptmx", 115200);
>> srl_flush(s1);
>> srl_write(s1, "Hello world!");
>> data = srl_read(s1,12);
>> error: srl_read: Invalid resizing operation or ambiguous assignment to
>> an out-of-bounds array element
>>
>> This may be due to the virtual loopback. I will try again with
>> hardware later, but it would be nice to see why there is a problem (if
>> there is!) with virtual loopbacks.
>>
>>
>> --
>> M. Sc. Juan Pablo Carbajal
>> -----
>> PhD Student
>> University of Zürich
>> http://ailab.ifi.uzh.ch/carbajal/
>
>

Hi,
Tested on Ubuntu 10.04 and it works fine. Several calls to the test do not break

Request: srl_read is blocking that is ok, however if by error we
called srl_read and there is nothing written on the port Octave hangs
and not even ctrl-C releases the block. the only solution is to kill
Octave. Can that be fixed?


-- 
M. Sc. Juan Pablo Carbajal
-----
PhD Student
University of Zürich
http://ailab.ifi.uzh.ch/carbajal/

Re: [OctDev] Fem PLotting: fpl 1.3.0 released

[Mike M. <mtm...@ie...>]

On Fri, Aug 10, 2012 at 2:07 AM, c. wrote:
>
> It seems the problem is that on my system I have multiple instances of Octave
> installed but which ever I run, when "mkoctfile" is invoked in the Makefile,
> what is running is the "mkoctfile" shipping whith 3.7.0+.

That should be fixed with

MKOCTFILE ?= mkoctfile

see main/signal/src/Makefile for example.

-- 
mike

Re: [OctDev] Windows dicom pkg?

[Andy B. <and...@gm...>]

On 9 August 2012 15:46, <a.b...@gm...> wrote:
>
> Hi Andy,
>
> I'm new here. I hope this is the right way to find this info. I apologize if it's not or you prefer not to be contacted directly this way.

It's better to use the list (see CC), then the conversation is public,
so others can benefit.

> I was looking for some info on Windows support of the dicom package. I found the following related threads: http://octave.1599824.n4.nabble.com/Dicom-files-td3620357.html, http://octave.1599824.n4.nabble.com/dicom-support-td2122699.html.

You missed the really good news, which is here
http://sourceforge.net/mailarchive/message.php?msg_id=29403428

> Unfortunately, I'm pretty new to Octave and couldn't make perfect sense of them. I think in the second thread you mention working on dicominfo with some success (the only function I actually need from the dicom package). Did you figure out how to support that function for Octave? Can you point me to a place that explains how I can use that function?

http://octave.sourceforge.net/dicom/function/dicominfo.html

I think Matlab is similar

http://www.mathworks.co.uk/help/toolbox/images/ref/dicominfo.html

The structure that you get back can be heavily nested.

rp=dicominfo('.\new\RP.1.2.246.352.71.5.194670609.279985.20120522152124.dcm');
printf("%30s: %s\n","name",rp.PatientName);
printf("%30s: %.1f\n","Gantry angle of first
beam",rp.BeamSequence.Item_1.ControlPointSequence.Item_1.GantryAngle);

> I currently run Octave 3.2.4, but can DL a different version if that would help.

Use the MSVC build of 3.6.2 and get the dicom package for it.

http://sourceforge.net/projects/octave/files/Octave%20Windows%20binaries/Octave%203.6.2%20for%20Windows%20Microsoft%20Visual%20Studio/

octave-3.6.2-vs2010-setup.exe
octave-3.6.2-dicom-0.1.1-vs2010-setup.exe

after octave starts you will need to do

pkg load dicom
addpath('C:\Program Files\Octave-3.6.2\share\octave\packages\dicom-0.1.1')

The second line tells it where to find the dicom dictionary (assuming
you go with the default path).

--
/* andy buckle */

Re: [OctDev] Fem PLotting: fpl 1.3.0 released

[c. <car...@gm...>]

On 9 Aug 2012, at 19:30, Alexander Hansen wrote:

> The fpl-1.3.0 release.
> 
> Here is the error message.  I'm building it with clang on Mac OS 10.8
> against a mostly-as-released Octave-3.2.4:
> 
> array_to_uint8.cc:65:11: error: no member named 'resize1' in
>      'intNDArray<octave_int<unsigned char>>'; did you mean 'resize'?
>          MAKE_INT_BRANCH(int8)
>          ^
> array_to_uint8.cc:57:21: note: expanded from macro 'MAKE_INT_BRANCH'
>                out.resize1 (len);                                      \
>                    ^
> /sw/include/octave-3.2.4/octave/Array.h:471:8: note: 'resize' declared here
>  void resize (octave_idx_type n)
>       ^
> 
> I don't see any instances of "resize1" in my installed Octave-3.2.4, or,
> for that matter, in the Octave-3.2.4 source.

Hi,

Thnks for providing more info.

It seems the problem is that on my system I have multiple instances of Octave
installed but which ever I run, when "mkoctfile" is invoked in the Makefile,
what is running is the "mkoctfile" shipping whith 3.7.0+.

Backward compatibility could be easily recovered by substituting "out.resize1" by 
"out.resize ( ... )" but I don't think it makes much sense to spend any work to 
support releases older than 3.4, so I'll just update the "depends:" field in "DESCRIPTION"
to specify 3.4 is needed.

Thanks,
c.

[OctDev] libcruft/amos - nonfree Bessel code in Octave?

[Steven G. J. <st...@al...>]

Dear Octave developers,

I was looking for a complex-argument Bessel-function code to use in our 
own free-software package, and I noticed that Octave is using the "amos" 
software (by Don. Amos).

However, I am concerned that the amos software may be non-free.  This 
software can also be found on Netlib (http://netlib.org/amos/), which 
says that they are derived from TOMS Algorithm 644:

	http://dl.acm.org/citation.cfm?id=214331

Unfortunately, all of the software in TOMS (ACM Transactions on 
Mathematical Software) is by default non-free, since the ACM license 
allows only noncommercial use:
	http://www.acm.org/publications/policies/softwarecrnotice
(Note that ACM also claims copyright ownership of all TOMS code, since 
authors are required to make a copyright assignment to ACM prior to 
publication.)

Have the Octave authors or the FSF investigated this and found a way to 
distribute the amos software legally under the GPL?   (In some 
circumstances, the ACM has granted waivers of its copyright policy, see 
e.g. http://projects.scipy.org/scipy/changeset/6120 for a different TOMS 
package.)

Regards,
Steven G. Johnson

[OctDev] /octave/about.html license?

[Ivan S. <one...@gm...>]

	The “about” Web page for GNU Octave currently bears a strict “no
	derivatives” license:

--cut: http://gnu.org/s/octave/about.html --
    Copyright © 1998-2012 John W. Eaton.  Verbatim copying and
    distribution is permitted in any medium, provided this notice is
    preserved.
--cut: http://gnu.org/s/octave/about.html --

	Unfortunately, such a wording (AIUI) prohibits the translation
	of this page into other languages, as well as the making of
	excerpts (e. g., to incorporate into other works.)

	Could this license be changed to a “freer” one, please?

	TIA.

	The backstory is that we're preparing for Software Freedom Day,
	and have decided to put short notes on prominent free software
	packages on the reverse of our “Join us on SFD!” flyers.  The
	author of the note describing GNU Octave was eager to use a
	couple of paragraphs from the aforementioned about.html as the
	basis for his work, only to discover later that that there's no
	permission granted for such a use.

	Are there anything Octave-related (posters, videos, etc.) we can
	show at SFD, BTW?

-- 
FSF associate member #7257	http://sf-day.org/

Re: [OctDev] Fem PLotting: fpl 1.3.0 released

[Alexander H. <ale...@gm...>]

On 8/9/12 10:15 AM, c. wrote:
> 
> On 9 Aug 2012, at 18:05, Alexander Hansen wrote:
> 
>> One minor note:  http://octave.sourceforge.net/fpl/index.html says that
>> the current version works with Octave >= 3.0.0, but I was only able to
>> get it to build with 3.4.x and later.
> 
> what version did you try? it seems to build correctly for me on 3.2.4.
> c.
> 

The fpl-1.3.0 release.

Here is the error message.  I'm building it with clang on Mac OS 10.8
against a mostly-as-released Octave-3.2.4:

array_to_uint8.cc:65:11: error: no member named 'resize1' in
      'intNDArray<octave_int<unsigned char>>'; did you mean 'resize'?
          MAKE_INT_BRANCH(int8)
          ^
array_to_uint8.cc:57:21: note: expanded from macro 'MAKE_INT_BRANCH'
                out.resize1 (len);                                      \
                    ^
/sw/include/octave-3.2.4/octave/Array.h:471:8: note: 'resize' declared here
  void resize (octave_idx_type n)
       ^

I don't see any instances of "resize1" in my installed Octave-3.2.4, or,
for that matter, in the Octave-3.2.4 source.
-- 
Alexander Hansen, Ph.D.
Fink User Liaison
My package updates: http://finkakh.wordpress.com/

Re: [OctDev] Fem PLotting: fpl 1.3.0 released

[c. <car...@gm...>]

On 9 Aug 2012, at 18:05, Alexander Hansen wrote:

> One minor note:  http://octave.sourceforge.net/fpl/index.html says that
> the current version works with Octave >= 3.0.0, but I was only able to
> get it to build with 3.4.x and later.

what version did you try? it seems to build correctly for me on 3.2.4.
c.

Re: [OctDev] Fem PLotting: fpl 1.3.0 released

[Alexander H. <ale...@gm...>]

On 8/8/12 7:17 AM, Carnë Draug wrote:
> Hi everyone
> 
> a new release of fpl package is out, version 1.3.0, by Carlo de Falco.
> Enjoy octave responsibly.
> 
> Carnë
> _______________________________________________

One minor note:  http://octave.sourceforge.net/fpl/index.html says that
the current version works with Octave >= 3.0.0, but I was only able to
get it to build with 3.4.x and later.

-- 
Alexander Hansen, Ph.D.
Fink User Liaison
My package updates: http://finkakh.wordpress.com/

Re: [OctDev] signal package: possible bug on fir2 with patch (please review)

[Mike M. <mtm...@ie...>]

On Fri, Jul 27, 2012 at 10:14 AM, Carnë Draug wrote:
> On 6 July 2012 03:11, Mike Miller <mtm...@ie...> wrote:
>>
>> The only thing that is not clear to me is whether this matches ML
>> behavior, and whether we care, e.g. if the caller passes in a 4th arg
>> too small, does ML generate an error or silently make it larger like
>> this line 29 does?
>
> MatLab gives an error in somes cases and silently makes it larger in others.
>
>>> f = [0 0.6 0.6 1]; m = [1 1 0 0];
>>> b5 = fir2(30,f,m, 5);
> Error using fir2 (line 135)
> The number of grid points npt must be greater than or equal to 16 for
> filter order 30.
>
>>> b8 = fir2(30,f,m, 8);
> Error using fir2 (line 135)
> The number of grid points npt must be greater than or equal to 16 for
> filter order 30.
>
>>> b9 = fir2(30,f,m, 9);
>>> b15 = fir2(30,f,m, 15);
>>> b16 = fir2(30,f,m, 16);
>>> b17 = fir2(30,f,m, 17);
>>> isequal (b9, b15)
>
> ans =
>
>      1
>
>>> isequal (b15, b16)
>
> ans =
>
>      1
>
>>> isequal (b16, b17)
>
> ans =
>
>      0
>
>
> Seems that MatLab behaviour is to adjust number of points (to the
> minimum), if the value is more than half of it and error otherwise.
>
> Not sure what's the best behavior but I'd guess at least a warning
> should be given.

Finally got back to this, it's now closer to what you showed here.
It's clear that Matlab is forcing the 4th argument to always be a
power of 2, even though that's not in the help text. I implemented
that, added an error message for the case of the argument being too
small (after conversion to a power of 2), and added a test case based
on your examples above. Thanks for those!

-- 
mike

Re: [OctDev] Alpha release of Serial I/O package (v0.1)

[Andrius S. <and...@gm...>]

Bug with reopening closed interface fixed in r10845. Could
someone please test package in Unix/Mac environment?

On Tue, Aug 7, 2012 at 9:37 AM, Juan Pablo Carbajal <car...@if...>wrote:

> On Tue, Aug 7, 2012 at 4:37 AM, Andrius Sutas <and...@gm...>
> wrote:
> > Hey,
> >
> > please see alpha release of Serial I/O package.
> >
> > I am looking for any comments on code style, usability, naming etc.
> Also, it
> > would be great if people could reply on which platforms they successfully
> > (and not, which is more important) tested it. Sorry but Windows is not
> yet
> > supported.
> >
> > Currently open flags are hardcoded to: O_RDWR | O_NOCTTY | O_SYNC, as I
> yet
> > to decide what is the best way to maintain cross-platformability and
> ease of
> > use (e.g. using string for parity setting and then doing switch
> according to
> > platform being used instead of making the same flags from different
> > platforms global to Octave), comments are welcome.
> >
> > Also, a simple example to get started (for testing adapter's RX and TX
> lines
> > are connected together):
> >
> > s0 = serial() # Opens default serial port ttyUSB0 in default
> configuration
> > of 115200, 8-N-1
> >
> > s1 = serial("/dev/ttyUSB1", 115200) # Opens serial port ttyUSB1 with
> > baudrate of 115200 (config defaults to 8-N-1)
> > srl_flush(s1); # Flush input and output buffers
> > srl_write(s1, "Hello world!") # Blocking write call, currently only
> accepts
> > strings
> > data = srl_read(s1, 11)  # Blocking read call, returns uint8 array of
> > exactly 11 bytes read (data = [72  101  108  108  111   32  119  111  114
> > 108  100])
> >
> > char(data) # Converts uint8 array to string, (ans = "Hello world", note:
> no
> > exclamation mark)
> >
> > srl_baudrate(s1, 9600) # Change baudrate
> > srl_bytesize(s1, 5) # Change byte size (config becomes 5-N-1)
> > srl_parity(s1, "E") # Changes parity checking (config becomes 5-E-1),
> > possible values [E]ven, [O]dd, [N]one.
> > srl_stopbits(s1, 2) # Changes stop bits (config becomes 5-E-2), possible
> > values 1, 2.
> >
> > s2 = serial("/dev/ttyS0", 9600, 6, "odd", 2) # Opens serial port ttyS0 in
> > 9600, 6-O-2 configuration
> >
> > srl_close(s0) # Closes and releases file descriptor
> >
> >
> ------------------------------------------------------------------------------
> > Live Security Virtual Conference
> > Exclusive live event will cover all the ways today's security and
> > threat landscape has changed and how IT managers can respond. Discussions
> > will include endpoint security, mobile security and the latest in malware
> > threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> > _______________________________________________
> > Octave-dev mailing list
> > Oct...@li...
> > https://lists.sourceforge.net/lists/listinfo/octave-dev
> >
>
> Hi Andrius,
>
> Thanks for the first release!
>
> When installing on Ubuntu 10.10, gcc version 4.4.5 (Ubuntu/Linaro
> 4.4.4-14ubuntu5.1) I get the following deprecation warning
>
> srl_read.cc: In function ‘octave_value_list Fsrl_read(const
> octave_value_list&, int)’:
> srl_read.cc:92: warning: ‘void Array<T>::resize(octave_idx_type) [with
> T = octave_int<unsigned char>]’ is deprecated (declared at
> /usr/local/include/octave-3.6.2/octave/../octave/Array.h:459)
>
> but I guess this is something in core.
>
> I have no usb-to-serial at hand (I am away from my lab) but I tested
> using virtual loopback using pseudo terminals.
> (for the ones who wan to try: In linux install "socat" and then run
> "socat -d -d PTY: PTY:")
> The following lines work as expected
>
> pkg load serial
> s1 = serial("/dev/ptmx", 115200);
> srl_flush(s1);
> srl_write(s1, "Hello world!")
> data = srl_read(s1,12);
> char(data)
> srl_close(s1)
>
> However when run again I get
>
> s1 = serial("/dev/ptmx", 115200);
> srl_flush(s1);
> srl_write(s1, "Hello world!");
> data = srl_read(s1,12);
> error: srl_read: Invalid resizing operation or ambiguous assignment to
> an out-of-bounds array element
>
> This may be due to the virtual loopback. I will try again with
> hardware later, but it would be nice to see why there is a problem (if
> there is!) with virtual loopbacks.
>
>
> --
> M. Sc. Juan Pablo Carbajal
> -----
> PhD Student
> University of Zürich
> http://ailab.ifi.uzh.ch/carbajal/
>

Re: [OctDev] csv2cell stalls when input has no eol

[Carnë D. <car...@gm...>]

On 8 August 2012 09:50, Muhali <jua...@gm...> wrote:
> with an input file that has no end of line ('\n'), csv2cell stalls.

Hi

could submit a patch to fix this? If not, could you please report it
on the bug tracker at
https://sourceforge.net/tracker/?group_id=2888&atid=102888 ? I'm also
CC'in Laurent Mazet, the author of this function.

Thanks in advance,
Carnë

[OctDev] Fem PLotting: fpl 1.3.0 released

[Carnë D. <car...@gm...>]

Hi everyone

a new release of fpl package is out, version 1.3.0, by Carlo de Falco.
Enjoy octave responsibly.

Carnë

[OctDev] csv2cell stalls when input has no eol

[Muhali <jua...@gm...>]

with an input file that has no end of line ('\n'), csv2cell stalls.

M.



--
View this message in context: http://gnu-octave-repository.2306053.n4.nabble.com/csv2cell-stalls-when-input-has-no-eol-tp4655568.html
Sent from the octave-dev mailing list archive at Nabble.com.

Re: [OctDev] Alpha release of Serial I/O package (v0.1)

[Carnë D. <car...@gm...>]

On 7 August 2012 12:08, Juan Pablo Carbajal <car...@if...> wrote:
> Please commit your code to the OF svn. (maybe in main/serial?)
>
> @Carnë: What do you think?

Sounds good to me. Please go ahead and make the commit, you already
have write permissions to the repository.

Carnë

Re: [OctDev] Alpha release of Serial I/O package (v0.1)

[Juan P. C. <car...@if...>]

On Tue, Aug 7, 2012 at 12:18 PM, Andrius Sutas <and...@gm...> wrote:
> Well thanks for the first bug!
>
> It looks like I have problems with interface closing and then reopening. If
> interface is not closed it works as expected. Will look into it tonight.
>
> s1 = serial("/dev/ptmx", 115200)
> srl_flush(s1);
> srl_write(s1, "Hello world!")
> data = srl_read(s1,12);
> char(data)
>
> srl_write(s1, "World!")
> data = srl_read(s1, 6);
> char(data)
>
> srl_close(s1)
>
> On Tue, Aug 7, 2012 at 9:37 AM, Juan Pablo Carbajal <car...@if...>
> wrote:
>>
>> On Tue, Aug 7, 2012 at 4:37 AM, Andrius Sutas <and...@gm...>
>> wrote:
>> > Hey,
>> >
>> > please see alpha release of Serial I/O package.
>> >
>> > I am looking for any comments on code style, usability, naming etc.
>> > Also, it
>> > would be great if people could reply on which platforms they
>> > successfully
>> > (and not, which is more important) tested it. Sorry but Windows is not
>> > yet
>> > supported.
>> >
>> > Currently open flags are hardcoded to: O_RDWR | O_NOCTTY | O_SYNC, as I
>> > yet
>> > to decide what is the best way to maintain cross-platformability and
>> > ease of
>> > use (e.g. using string for parity setting and then doing switch
>> > according to
>> > platform being used instead of making the same flags from different
>> > platforms global to Octave), comments are welcome.
>> >
>> > Also, a simple example to get started (for testing adapter's RX and TX
>> > lines
>> > are connected together):
>> >
>> > s0 = serial() # Opens default serial port ttyUSB0 in default
>> > configuration
>> > of 115200, 8-N-1
>> >
>> > s1 = serial("/dev/ttyUSB1", 115200) # Opens serial port ttyUSB1 with
>> > baudrate of 115200 (config defaults to 8-N-1)
>> > srl_flush(s1); # Flush input and output buffers
>> > srl_write(s1, "Hello world!") # Blocking write call, currently only
>> > accepts
>> > strings
>> > data = srl_read(s1, 11)  # Blocking read call, returns uint8 array of
>> > exactly 11 bytes read (data = [72  101  108  108  111   32  119  111
>> > 114
>> > 108  100])
>> >
>> > char(data) # Converts uint8 array to string, (ans = "Hello world", note:
>> > no
>> > exclamation mark)
>> >
>> > srl_baudrate(s1, 9600) # Change baudrate
>> > srl_bytesize(s1, 5) # Change byte size (config becomes 5-N-1)
>> > srl_parity(s1, "E") # Changes parity checking (config becomes 5-E-1),
>> > possible values [E]ven, [O]dd, [N]one.
>> > srl_stopbits(s1, 2) # Changes stop bits (config becomes 5-E-2), possible
>> > values 1, 2.
>> >
>> > s2 = serial("/dev/ttyS0", 9600, 6, "odd", 2) # Opens serial port ttyS0
>> > in
>> > 9600, 6-O-2 configuration
>> >
>> > srl_close(s0) # Closes and releases file descriptor
>> >
>> >
>> > ------------------------------------------------------------------------------
>> > Live Security Virtual Conference
>> > Exclusive live event will cover all the ways today's security and
>> > threat landscape has changed and how IT managers can respond.
>> > Discussions
>> > will include endpoint security, mobile security and the latest in
>> > malware
>> > threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
>> > _______________________________________________
>> > Octave-dev mailing list
>> > Oct...@li...
>> > https://lists.sourceforge.net/lists/listinfo/octave-dev
>> >
>>
>> Hi Andrius,
>>
>> Thanks for the first release!
>>
>> When installing on Ubuntu 10.10, gcc version 4.4.5 (Ubuntu/Linaro
>> 4.4.4-14ubuntu5.1) I get the following deprecation warning
>>
>> srl_read.cc: In function 'octave_value_list Fsrl_read(const
>> octave_value_list&, int)':
>> srl_read.cc:92: warning: 'void Array<T>::resize(octave_idx_type) [with
>> T = octave_int<unsigned char>]' is deprecated (declared at
>> /usr/local/include/octave-3.6.2/octave/../octave/Array.h:459)
>>
>> but I guess this is something in core.
>>
>> I have no usb-to-serial at hand (I am away from my lab) but I tested
>> using virtual loopback using pseudo terminals.
>> (for the ones who wan to try: In linux install "socat" and then run
>> "socat -d -d PTY: PTY:")
>> The following lines work as expected
>>
>> pkg load serial
>> s1 = serial("/dev/ptmx", 115200);
>> srl_flush(s1);
>> srl_write(s1, "Hello world!")
>> data = srl_read(s1,12);
>> char(data)
>> srl_close(s1)
>>
>> However when run again I get
>>
>> s1 = serial("/dev/ptmx", 115200);
>> srl_flush(s1);
>> srl_write(s1, "Hello world!");
>> data = srl_read(s1,12);
>> error: srl_read: Invalid resizing operation or ambiguous assignment to
>> an out-of-bounds array element
>>
>> This may be due to the virtual loopback. I will try again with
>> hardware later, but it would be nice to see why there is a problem (if
>> there is!) with virtual loopbacks.
>>
>>
>> --
>> M. Sc. Juan Pablo Carbajal
>> -----
>> PhD Student
>> University of Zürich
>> http://ailab.ifi.uzh.ch/carbajal/
>
>
>
>
> --
> Andrius Šutas
>
> Skype: andrius.sutas
> Mob. Tel.: +37063443035
> GTalk: and...@gm...

Ok, happy to see the problem wasn't with the virtual loopback.

Please commit your code to the OF svn. (maybe in main/serial?)

@Carnë: What do you think?


-- 
M. Sc. Juan Pablo Carbajal
-----
PhD Student
University of Zürich
http://ailab.ifi.uzh.ch/carbajal/

Re: [OctDev] Alpha release of Serial I/O package (v0.1)

[Andrius S. <and...@gm...>]

Well thanks for the first bug!

It looks like I have problems with interface closing and then reopening. If
interface is not closed it works as expected. Will look into it tonight.

s1 = serial("/dev/ptmx", 115200)
srl_flush(s1);
srl_write(s1, "Hello world!")
data = srl_read(s1,12);
char(data)

srl_write(s1, "World!")
data = srl_read(s1, 6);
char(data)

srl_close(s1)

On Tue, Aug 7, 2012 at 9:37 AM, Juan Pablo Carbajal <car...@if...>wrote:

> On Tue, Aug 7, 2012 at 4:37 AM, Andrius Sutas <and...@gm...>
> wrote:
> > Hey,
> >
> > please see alpha release of Serial I/O package.
> >
> > I am looking for any comments on code style, usability, naming etc.
> Also, it
> > would be great if people could reply on which platforms they successfully
> > (and not, which is more important) tested it. Sorry but Windows is not
> yet
> > supported.
> >
> > Currently open flags are hardcoded to: O_RDWR | O_NOCTTY | O_SYNC, as I
> yet
> > to decide what is the best way to maintain cross-platformability and
> ease of
> > use (e.g. using string for parity setting and then doing switch
> according to
> > platform being used instead of making the same flags from different
> > platforms global to Octave), comments are welcome.
> >
> > Also, a simple example to get started (for testing adapter's RX and TX
> lines
> > are connected together):
> >
> > s0 = serial() # Opens default serial port ttyUSB0 in default
> configuration
> > of 115200, 8-N-1
> >
> > s1 = serial("/dev/ttyUSB1", 115200) # Opens serial port ttyUSB1 with
> > baudrate of 115200 (config defaults to 8-N-1)
> > srl_flush(s1); # Flush input and output buffers
> > srl_write(s1, "Hello world!") # Blocking write call, currently only
> accepts
> > strings
> > data = srl_read(s1, 11)  # Blocking read call, returns uint8 array of
> > exactly 11 bytes read (data = [72  101  108  108  111   32  119  111  114
> > 108  100])
> >
> > char(data) # Converts uint8 array to string, (ans = "Hello world", note:
> no
> > exclamation mark)
> >
> > srl_baudrate(s1, 9600) # Change baudrate
> > srl_bytesize(s1, 5) # Change byte size (config becomes 5-N-1)
> > srl_parity(s1, "E") # Changes parity checking (config becomes 5-E-1),
> > possible values [E]ven, [O]dd, [N]one.
> > srl_stopbits(s1, 2) # Changes stop bits (config becomes 5-E-2), possible
> > values 1, 2.
> >
> > s2 = serial("/dev/ttyS0", 9600, 6, "odd", 2) # Opens serial port ttyS0 in
> > 9600, 6-O-2 configuration
> >
> > srl_close(s0) # Closes and releases file descriptor
> >
> >
> ------------------------------------------------------------------------------
> > Live Security Virtual Conference
> > Exclusive live event will cover all the ways today's security and
> > threat landscape has changed and how IT managers can respond. Discussions
> > will include endpoint security, mobile security and the latest in malware
> > threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> > _______________________________________________
> > Octave-dev mailing list
> > Oct...@li...
> > https://lists.sourceforge.net/lists/listinfo/octave-dev
> >
>
> Hi Andrius,
>
> Thanks for the first release!
>
> When installing on Ubuntu 10.10, gcc version 4.4.5 (Ubuntu/Linaro
> 4.4.4-14ubuntu5.1) I get the following deprecation warning
>
> srl_read.cc: In function ‘octave_value_list Fsrl_read(const
> octave_value_list&, int)’:
> srl_read.cc:92: warning: ‘void Array<T>::resize(octave_idx_type) [with
> T = octave_int<unsigned char>]’ is deprecated (declared at
> /usr/local/include/octave-3.6.2/octave/../octave/Array.h:459)
>
> but I guess this is something in core.
>
> I have no usb-to-serial at hand (I am away from my lab) but I tested
> using virtual loopback using pseudo terminals.
> (for the ones who wan to try: In linux install "socat" and then run
> "socat -d -d PTY: PTY:")
> The following lines work as expected
>
> pkg load serial
> s1 = serial("/dev/ptmx", 115200);
> srl_flush(s1);
> srl_write(s1, "Hello world!")
> data = srl_read(s1,12);
> char(data)
> srl_close(s1)
>
> However when run again I get
>
> s1 = serial("/dev/ptmx", 115200);
> srl_flush(s1);
> srl_write(s1, "Hello world!");
> data = srl_read(s1,12);
> error: srl_read: Invalid resizing operation or ambiguous assignment to
> an out-of-bounds array element
>
> This may be due to the virtual loopback. I will try again with
> hardware later, but it would be nice to see why there is a problem (if
> there is!) with virtual loopbacks.
>
>
> --
> M. Sc. Juan Pablo Carbajal
> -----
> PhD Student
> University of Zürich
> http://ailab.ifi.uzh.ch/carbajal/
>



-- 
Andrius Šutas

Skype: andrius.sutas
Mob. Tel.: +37063443035
GTalk: and...@gm...

Re: [OctDev] Alpha release of Serial I/O package (v0.1)

[Juan P. C. <car...@if...>]

On Tue, Aug 7, 2012 at 4:37 AM, Andrius Sutas <and...@gm...> wrote:
> Hey,
>
> please see alpha release of Serial I/O package.
>
> I am looking for any comments on code style, usability, naming etc. Also, it
> would be great if people could reply on which platforms they successfully
> (and not, which is more important) tested it. Sorry but Windows is not yet
> supported.
>
> Currently open flags are hardcoded to: O_RDWR | O_NOCTTY | O_SYNC, as I yet
> to decide what is the best way to maintain cross-platformability and ease of
> use (e.g. using string for parity setting and then doing switch according to
> platform being used instead of making the same flags from different
> platforms global to Octave), comments are welcome.
>
> Also, a simple example to get started (for testing adapter's RX and TX lines
> are connected together):
>
> s0 = serial() # Opens default serial port ttyUSB0 in default configuration
> of 115200, 8-N-1
>
> s1 = serial("/dev/ttyUSB1", 115200) # Opens serial port ttyUSB1 with
> baudrate of 115200 (config defaults to 8-N-1)
> srl_flush(s1); # Flush input and output buffers
> srl_write(s1, "Hello world!") # Blocking write call, currently only accepts
> strings
> data = srl_read(s1, 11)  # Blocking read call, returns uint8 array of
> exactly 11 bytes read (data = [72  101  108  108  111   32  119  111  114
> 108  100])
>
> char(data) # Converts uint8 array to string, (ans = "Hello world", note: no
> exclamation mark)
>
> srl_baudrate(s1, 9600) # Change baudrate
> srl_bytesize(s1, 5) # Change byte size (config becomes 5-N-1)
> srl_parity(s1, "E") # Changes parity checking (config becomes 5-E-1),
> possible values [E]ven, [O]dd, [N]one.
> srl_stopbits(s1, 2) # Changes stop bits (config becomes 5-E-2), possible
> values 1, 2.
>
> s2 = serial("/dev/ttyS0", 9600, 6, "odd", 2) # Opens serial port ttyS0 in
> 9600, 6-O-2 configuration
>
> srl_close(s0) # Closes and releases file descriptor
>
> ------------------------------------------------------------------------------
> Live Security Virtual Conference
> Exclusive live event will cover all the ways today's security and
> threat landscape has changed and how IT managers can respond. Discussions
> will include endpoint security, mobile security and the latest in malware
> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> _______________________________________________
> Octave-dev mailing list
> Oct...@li...
> https://lists.sourceforge.net/lists/listinfo/octave-dev
>

Hi Andrius,

Thanks for the first release!

When installing on Ubuntu 10.10, gcc version 4.4.5 (Ubuntu/Linaro
4.4.4-14ubuntu5.1) I get the following deprecation warning

srl_read.cc: In function ‘octave_value_list Fsrl_read(const
octave_value_list&, int)’:
srl_read.cc:92: warning: ‘void Array<T>::resize(octave_idx_type) [with
T = octave_int<unsigned char>]’ is deprecated (declared at
/usr/local/include/octave-3.6.2/octave/../octave/Array.h:459)

but I guess this is something in core.

I have no usb-to-serial at hand (I am away from my lab) but I tested
using virtual loopback using pseudo terminals.
(for the ones who wan to try: In linux install "socat" and then run
"socat -d -d PTY: PTY:")
The following lines work as expected

pkg load serial
s1 = serial("/dev/ptmx", 115200);
srl_flush(s1);
srl_write(s1, "Hello world!")
data = srl_read(s1,12);
char(data)
srl_close(s1)

However when run again I get

s1 = serial("/dev/ptmx", 115200);
srl_flush(s1);
srl_write(s1, "Hello world!");
data = srl_read(s1,12);
error: srl_read: Invalid resizing operation or ambiguous assignment to
an out-of-bounds array element

This may be due to the virtual loopback. I will try again with
hardware later, but it would be nice to see why there is a problem (if
there is!) with virtual loopbacks.


-- 
M. Sc. Juan Pablo Carbajal
-----
PhD Student
University of Zürich
http://ailab.ifi.uzh.ch/carbajal/

Re: [OctDev] pei_tseng_notch function failing unit test

[Dr. A. K. <Ale...@ma...>]

Am 06.08.2012 um 01:31 schrieb Mike Miller:

> Hi Alex, I see that you are the author of pei_tseng_notch, would you
> mind taking a look at the test failure? I get the same result as that
> shown in the report (reproduced below).

> assert (damp_db,[-3, -251.9, -3],0.1) expected
>    -3.0000  -251.9000    -3.0000
> but got
>    -3.0103  -250.7217    -3.0935
> maximum absolute error 1.17825 exceeds tolerance 0.1


Mike,

the error is much too small to have any significant impact in applications, and I should have chosen relative tolerances instead of absolute ones right away.

I'll fix that during the week.

	Alex

-- 
Dr. Alexander Klein, Diplom-Mathematiker
Physiologisches Institut der JLU-Gießen
Aulweg 129
35392 Gießen

http://www.med.uni-giessen.de/physio/