- Getting Started
- Hardware
- Software
- Brick Daemon (brickd)
- Brick Viewer (brickv)
- Brick Logger
- API Bindings
- ESP32 Firmware
- APT Repository
- Device Identifier
- Source Code and Bug Tracking
- Programming Interface

- Kits
- Embedded Boards
- Specifications

The MATLAB/Octave bindings allow you to control Bricks and Bricklets from your MATLAB/Octave scripts. The ZIP file for the bindings contains:

`matlab/Tinkerforge.jar`

, a precompiled Java library for MATLAB- in
`matlab/source/`

the source code of`Tinkerforge.jar`

for MATLAB - in
`matlab/examples/`

the MATLAB examples for every Brick and Bricklet `octave/Tinkerforge.jar`

, a precompiled Java library for Octave- in
`octave/source/`

the source code of`Tinkerforge.jar`

for Octave - in
`octave/examples/`

the Octave examples for every Brick and Bricklet

The MATLAB/Octave bindings are based on the Java bindings.

- MATLAB or Octave 3.6 or newer with Java 8 support

Before you and use the bindings with MATLAB or Octave you have to install them.

The Java support in MATLAB is typically enabled by default. You can test this with the following command in MATLAB:

```
version -java
```

If this command doesn't show Java support then see the MATLAB documentation about how to configure Java for MATLAB.

To use the bindings MATLAB needs to know where to find the `Tinkerforge.jar`

file. There are several ways to archive this, see the MATLAB documentation
for more details on all of them.
The recommended way is to the add the bindings to the preferences folder.

Start MATLAB and run the following command to get the path of the preferences folder:

```
prefdir(1)
```

Preferences folder path examples:

- Windows:
`C:\Users\<user>\AppData\local\MathWorks\MATLAB\R2016a`

- Linux:
`/home/<user>/.matlab/R2016a`

- macOS:
`/Users/<user>/.matlab/R2016a`

Copy the `Tinkerforge.jar`

file from the `matlab/`

folder to the preferences
folder. Then the `Tinkerforge.jar`

file has to be added to MATLAB's class path.
Edit or create a file named `javaclasspath.txt`

in the preferences folder
and add the absolute path to the `Tinkerforge.jar`

file as a new line to it.
For example:

- Windows:
`C:\Users\<user>\AppData\local\MathWorks\MATLAB\R2016a\Tinkerforge.jar`

- Linux:
`/home/<user>/.matlab/R2016a/Tinkerforge.jar`

- macOS:
`/Users/<user>/.matlab/R2016a/Tinkerforge.jar`

Restart MATLAB and run the following command, it should list the
`Tinkerforge.jar`

file:

```
javaclasspath
```

The bindings are now ready to use.

If Java support is available in Octave depends on the Octave version. Until version 3.6 (inclusive) Java support was in a separate module. Since version 3.8 it is available by default.

On Linux you have to install an extra package for the Java support in Octave 3.6:

```
sudo apt-get install octave octave-java
```

For Windows we recommend the MinGW build of Octave, because it comes with Java support by default. The Octave Wiki has a guide about how to set up Octave for Windows.

You can test if Java support is available with the following command in Octave:

```
octave_config_info("features").JAVA
```

The bindings are available in our APT repository for Debian based Linux distributions. Follow the setup guide then install the bindings:

```
sudo apt install octave-tinkerforge
```

The bindings JAR file is installed to this location:

```
/usr/share/octave/packages/tinkerforge/tinkerforge.jar
```

To make the bindings available in Octave the bindings JAR file has to be added to Octave's class path. This can be done with the following Octave command on Debian based Linux distributions:

```
javaaddpath("/usr/share/octave/packages/tinkerforge/tinkerforge.jar");
```

Alternatively, the `Tinkerforge.jar`

from the `octave/`

folder in the ZIP
file has to beadded to Octave's class path. This can be done with the following
Octave command on Windows:

```
javaaddpath("C:\\Absolute\\path\\to\\Octave\\Tinkerforge.jar");
```

Or by this Octave command on Linux:

```
javaaddpath("/Absolute/path/to/Octave/Tinkerforge.jar");
```

To make this change persistent you can add the command to the following file on Linux:

```
~/.octaverc
```

If this file didn't exist yet you can just create it. Octave has to be restarted after changing this file.

To test a MATLAB/Octave example Brick Daemon and Brick Viewer have to be installed first. Brick Daemon acts as a proxy between the USB interface of the Bricks and the API bindings. Brick Viewer connects to Brick Daemon and helps to figure out basic information about the connected Bricks and Bricklets.

As an example we will run the Stepper Brick configuration example. To do this
open the `matlab_example_configuration.m`

file from the
`matlab/examples/brick/stepper/`

folder in MATLAB.

In the example `HOST`

and `PORT`

specify at which network address the
Stepper Brick can be found. If it is connected locally to USB then `localhost`

and 4223 is correct. The `UID`

value has to be changed to the UID of the
connected Stepper Brick, which you can figure out using Brick Viewer:

```
HOST = 'localhost';
PORT = 4223;
UID = 'XXYYZZ'; % Change XXYYZZ to the UID of your Stepper Brick
```

Now you're ready to test an example.

As an example we will run the Stepper Brick configuration example. To do this
open the `octave_example_configuration.m`

file from the
`octave/examples/brick/stepper/`

folder in Octave.

In the example `HOST`

and `PORT`

specify at which network address the
Stepper Brick can be found. If it is connected locally to USB then `localhost`

and 4223 is correct. The `UID`

value has to be changed to the UID of the
connected Stepper Brick, which you can figure out using Brick Viewer:

```
HOST = "localhost";
PORT = 4223;
UID = "XXYYZZ"; % Change XXYYZZ to the UID of your Stepper Brick
```

Now you're ready to test an example.

The Octave examples are function files. To execute them directly from a terminal, they have to be extended to script files. Just add a call to the example function at the end of the example:

```
function octave_example_configuration()
% ...
end
octave_example_configuration(); % Add this line
```

**Callbacks don't work with Octave 3.8 or newer** (solved)

In bindings version 2.0.13 or older in Octave 3.8 the Invoke function throws
an `java.lang.UnsatisfiedLinkError`

exception. The Invoke function is required
to call Octave functions from Java. The bindings use this for callbacks.
This means that you cannot use callbacks in Octave 3.8. A discussion
on the Octave mailing list didn't come to a conclusion of fix for this.

This issue is fixed since bindings version 2.0.14.

Links to the API reference for the IP Connection, Bricks and Bricklets as well as the examples from the ZIP file of the bindings are listed in the following table. Further project descriptions can be found in the Kits section.