This directory contains a proposal (still a work in progress) for a domain specific language serialization for the VODML called VODSL. The aims of the serialization are
- have something easier than XML to edit
- more constrained and specific than UML
- focused editors via eclipse Xtext with simultaneous graphical visualization via FXDiagram
Some background information and demonstration of how to use the editor can be found in this presentation.
and there is a screencast showing installation and simple use of the editor. (note that the installation has now changed - see below)
In addition to be able to use some of the features of the graphical visualization via FXDiagram it is worth looking at the introductory video on that site.
The language is discussed in this note
There is an example that uses most of the language constructs.
The VODSLEditor.setup file can be read into the eclipse installer in advanced mode as a user product to create a minimal customized eclipse installation that just contains the vodsl editor.
- ensure that you have a Java 11 or later installed on your machine as well as graphviz
- download the eclipse installer https://www.eclipse.org/downloads/
- Run the eclipse installer and select "advanced mode" from the menu at top right.
- use the green arrow at the top right to add a new user product with the following url https://raw.githubusercontent.com/pahjbo/vodsl/master/VODSLEditor.setup
- select the "VODSL" user product and just click next through the dialogs until you have a running editor.
- create a new "general" project and then create a file with extension
.vodsl
- eclipse will prompt whether to convert the project to "XText" - say yes.
Once you have an editor then you can also explore this GitHub project which contains some example/standard models expressed in VODSL.
It is possible to use the parser machinery in a stand-alone fashion (i.e. without
having to work in eclipse) by using the jar file that is produced in the vodsl.standalone
sub-directory.
This jar (only) is published to maven central so you do not to have to build it.
Use the following command to run the parser
java -jar vodslparser-0.4.9-standalone.jar model.vodsl
which will produce a file model.vo-dml.xml
of the equivalent VO-DML.
The stand-alone parser is built using maven. All that is necessary (after building the editor plugins in the top level directory with mvn install) is to run
mvn install
in the vodsl.standalone/
directory and a jar file will be produced in target/
Note the standalone parser is now integrated in the standard VO-DML tooling which probably provides a more comprehensive set of tools to work with VO-DML/VODSL models.
If you want to develop vodsl itself then see the developer instructions.