From ArchWiki
Revision as of 16:57, 1 August 2012 by Alexfikl (talk | contribs) (added some info about plotting backends and GUIs and more about Octave-Forge. rewrote a lot of things.)
Jump to: navigation, search

Octave is a high-level language, primarily intended for numerical computations. From the official website:

GNU Octave is a high-level interpreted language, primarily intended for numerical computations. It provides capabilities for the numerical solution of linear and nonlinear problems, and for performing other numerical experiments. It also provides extensive graphics capabilities for data visualization and manipulation. Octave is normally used through its interactive command line interface, but it can also be used to write non-interactive programs. The Octave language is quite similar to Matlab so that most programs are easily portable.


Octave can be installed with the package octave, available in the official repositories.


Extra packages for GNU Octave can be found in Octave-Forge. They are comparable to toolboxes in Matlab. Some may be found in AUR (a non-exhaustive list):

  • Image — Provides functions for processing images.
http://octave.sourceforge.net/image/index.html || octave-imageAUR
  • Statistics — Provides additional statistical functions.
http://octave.sourceforge.net/statistics/index.html || octave-statisticsAUR
  • IO — Provides input and output in external formats such as XML, CSV or XLS.
http://octave.sourceforge.net/io/index.html || octave-ioAUR
  • Signal — Provides functions for processing images
http://octave.sourceforge.net/image/index.html || octave-signalAUR
  • Multicore — Provides functions for parallel processing on multiple cores.
http://octave.sourceforge.net/multicore/index.html || octave-multicoreAUR

A complete list can be found on Octave-Forge. For packages that do not exist on the AUR the following steps need to be followed :

  1. Download the .tar.gz package from Octave-Forge.
  2. Install it from Octave as user by executing:
octave:1> pkg install package_file_name.tar.gz

Packages have dependencies which you have to install before installing your package, if you want to bypass this use:

octave:1> pkg install -no-deps package_file_name.tar.gz

To uninstall a package:

octave:1> pkg uninstall package_file_name.tar.gz

Some packages get loaded automatically by Octave, for those which do not:

octave:1> pkg load package_name


octave:1> pkg load all

To see which packages have been loaded use pkg list, the packages with an asterisk are the ones that are already loaded.


Octave has two official plotting backends:

  • GnuPlot: The classic plotting utility.
  • FLTK: A new experimental OpenGL backend based on the fltk GUI toolkit.

GnuPlot is still the default, but it can be changed by:

octave:1> graphics_toolkit("fltk");

To make this change permanent add it to your ~/.octaverc file.

Graphical interfaces

There are very few graphical interfaces for Octave and none of them are official:

  • Cantor — A graphical user interface that delegates its mathematical operations to one of several back ends (Scilab, Maxima, Octave and others)
http://edu.kde.org/cantor/ || kdeedu-cantor
  • QtOctave — A Qt frontend for Octave
https://forja.rediris.es/projects/csl-qtoctave/ || qtoctaveAUR

Reading Microsoft Excel Spreadsheets

There are several ways to read Microsoft Excel files with Octave.

Converting to an open format

The easiest way to use .xls files in Octave would be to convert them to .csv or .ods using Calc (limited to 1024 columns) from Libreoffice or Sheets(limited to 32768 columns) from the the Calligra Suite.

After the conversion is complete you can use the build-in Octave function csvread for .csv files:

octave:1> csvread('myfile.csv');

For .ods files the octave-ioAUR package is necessary which contains the odsread function:

octave:1> odsread('myfile.ods');

For .xlsx files you can use the xlsx2csvAUR package from AUR:

 xlsx2csv -t /path/to/save/location -x /path/to/myfile.xlsx 

Reading xls files directly from Octave

If converting to .csv or .ods is not an option or does not work for whatever reason, Octave is able to read .xls files using octave-ioAUR package which contains the function xlsread.

The xlsread requires some Java libraries as the support for XLS files is not built-in. The steps necessary to make it work are:

1. Install jdk7-openjdk, available in the official repositories.
A common problem is that Octave cannot find the Java libraries properly. To fix this execute the following commands in your shell:
$ export JAVA_HOME=/usr/lib/jvm/java-7-openjdk
2. Install a Java XLS library for xlsread. A comparison of the available interfaces can be found at this link, with the recommended variant apache-poiAUR that can be found in the AUR.
3. Finally, install the octave-javaAUR package from AUR.

To check if Java is working correctly in Octave, see the output of:

 octave:1> javaclasspath 
     - empty -
     - empty -

To load the selected library in Octave, check if it is in the Java path. If not:

 octave:1> javaaddpath path/to/file.jar

(should be /usr/share/java/apache-poi/poi-3.x.jar and /usr/share/java/apache-poi/poi-ooxml-3.x.jar).

To check if it works

 octave:1> chk_spreadsheet_support 

The output should be > 0:

                   0 No spreadsheet I/O support found
                 ---------- XLS (Excel) interfaces: ----------
                   1 = COM (ActiveX / Excel)
                   2 = POI (Java / Apache POI)
                   4 = POI+OOXML (Java / Apache POI)
                   8 = JXL (Java / JExcelAPI)
                  16 = OXS (Java / OpenXLS)
                 --- ODS (OpenOffice.org Calc) interfaces ----
                  32 = OTK (Java/ ODF Toolkit)
                  64 = JOD (Java / jOpenDocument)
                 ----------------- XLS & ODS: ----------------
                 128 = UNO (Java / UNO bridge - OpenOffice.org)

To make this permanent add the javaaddpath commands to your ~/.octaverc file.