Re: [OctDev] fixed fails to build on linux

2008-03-28 Thread dbateman
Tatsuro MATSUOKA-2 wrote: > > Main problem is that new Makefile by you did not give a > liboctave_fixed.dll.a > so that the link seeme to be failed. > The idea was to try and link against the fixed.oct file rather than a separate library. That is what Michael did under MSVC, so I suspect th

Re: [OctDev] fixed fails to build on linux

2008-03-28 Thread Tatsuro MATSUOKA
Hello I felt bad in the morning so that I canceled today's work. In the afternoon my situaion rather became better so I tried the be below. I have down loaded the SVN today by langus (an alias comand by me to set locale to US for avoiding character traslation mistakes) svn co https://octave.svn

[OctDev] [Changeset]: fix for plot(zeros(1, 0), zeros(1, 0) (Was: Octave-Forge: boxplot)

2008-03-28 Thread John W. Eaton
On 28-Mar-2008, David Bateman wrote: | I'd hesitate to make that change since as you say the bug is in Octave | itself.. We shouldn't add additional cruft to work around bugs when we | can fix the issue at the source.. | | This reason this issue is happening is that "isvector([])" returns false |

Re: [OctDev] [Changeset]: fix for plot(zeros(1, 0), zeros(1, 0) (Was: Octave-Forge: boxplot)

2008-03-28 Thread Shaun Jackman
Hi David, You're patch works for me. Thanks! Cheers, Shaun On 28-Mar-08, at 2:25 AM, David Bateman wrote: > I'd hesitate to make that change since as you say the bug is in Octave > itself.. We shouldn't add additional cruft to work around bugs when we > can fix the issue at the source.. > > This

Re: [OctDev] Suggested patch to imwrite for Windows

2008-03-28 Thread David Bateman
Michal Fita wrote: > Hi Developers! > > I would like to thank you for your great effort put into Octave development > - it is realy useful to have such tool really open and widely available for > scientific and engineering work. > > I attach modified imwrite.m which now works under Windows with Ima

Re: [OctDev] window@(tukeywin,n)

2008-03-28 Thread David Bateman
Laufer, Ralph wrote: > I am a complete newbie and was wondering if Octave can interpret a > Matlab window function of the type: > > w = window(@tukeywin,n) > > and if so, are there any special add-ins needed? > > Thanks > Ralph > > The reason for the "sick" syntax in windows is that "w =

[OctDev] [Changeset]: fix for plot(zeros(1, 0), zeros(1, 0) (Was: Octave-Forge: boxplot)

2008-03-28 Thread David Bateman
I'd hesitate to make that change since as you say the bug is in Octave itself.. We shouldn't add additional cruft to work around bugs when we can fix the issue at the source.. This reason this issue is happening is that "isvector([])" returns false whereas "isvector(zeros(1,0))" returns true which