OceanDSL Python-based DSL Implementation

Name

oceandsl-languages

Description

This project contains multiple DSLs for ocean and earth system models.

• CP-DSL including Declaration-DSL, Configuration-DSL and Template-DSL
• TDD-DSL testing DSL for test-driven development
• BGC-DSL language to specify biogeochemical models

Documentation including the languages and architecture can be found in the doc.

License

Apache 2.0

Install

The OceanDSL DSLs all are based on Python and the provided instructions in the readmes uses VSCode. We will add instructions for Emacs and Vi at a later date.

In case you do not want to use VSCode, you can skip npm, node and VSCode installation instructions.

Install Python

The OceanDSL languages are based on python3.

Install Python in Ubuntu with:

sudo apt install python3 python3-pip

In some distributions python3 uses the executable name python3 leaving python for python2. As the following instructions use python, you have either to substitute the names accordingly or in Ubuntu install the following package:

sudo apt install python-is-python3

Install the virtual environment. This helps to keep your distribution Python setup and the setup for the DSLs separate.

sudo apt install python3-virtualenv

Alternatively, you can install the packages via pip. Howver, this does not work in Ubuntu, as it manages its packages via apt/dpkg.

python -m pip install --upgrade pip setuptools wheel

Check whether virtualenv is installed:

python -m pip show virtualenv

This should result in Name: virtualenv Version: 20.24.1+ds Summary: Virtual Python Environment builder Home-page: Author: Author-email: License: Location: /usr/lib/python3/dist-packages Requires: distlib, filelock, platformdirs Required-by:

If missing install it:

python -m pip install virtualenv

Install LSP Server Dependencies

For this how to, we assume that you install your virtual environment in your home directory. You can, however, install it anywhere. Be sure to adjust paths accordingly.

cd $HOME

Generate virtual environment in directory .venv using the standard library venv or use the tool virtualenv.

Using venv:

python -m venv .venv

Using virtualenv:

virtualenv .venv

Activate virtual environment .venv

source .venv/bin/activate

This will result in the following output: (.venv) user@hostname:~$ Where user will be your username and hostname will be the name of your computer.

Install antlr4-python3-runtime, Jinja2, mock, pygls and pytest package into virtual environment .venv

python -m pip install antlr4-python3-runtime Jinja2 mock pygls pytest

The last line of this operation should look like this: Successfully installed Jinja2-3.1.3 MarkupSafe-2.1.5 antlr4-python3-runtime-4.13.1 attrs-23.2.0 cattrs-23.2.3 iniconfig-2.0.0 lsprotocol-2023.0.1 mock-5.1.0 packaging-23.2 pluggy-1.4.0 pygls-1.3.0 pytest-8.0.0

The version numbers reflect the state of these packages while writing this documentation. They may differ in future.

Update npm and node

In case you want to use VSCode as editor, you need JavaScript. Therefore, install and update JavaScript package manager which is used by VSCode.

sudo apt install npm

Check npm version

npm --version

Outputs, e.g., 10.4.0 or a later version

In case the package is not up-to-date, install latest npm version globally (-g or --global)

sudo npm install -g npm@latest`

Output: ``` changed 14 packages in 1s

24 packages are looking for funding run npm fund for details ``` Depending on your installation, the numbers may differ.

Install latest node version

sudo n latest

Output: copying : node/21.6.1 installed : v21.6.1 (with npm 10.2.4)

The version numbers may differ in your installation.

Install the dependencies to the global mode (-g or --global)

sudo npm update -g

Output: ``` removed 25 packages, and changed 47 packages in 1s

24 packages are looking for funding run npm fund for details ``` The version numbers may differ in your installation.

Install and setup VScode

Install VScode in Ubuntu:

$ sudo apt install code

Install Python extensions in VSCode

Start VScode by typing:

$ code

You can no choose a theme, if you like.

Click on extensions. This is the icon on the left made out of squares. If unsure hover with the mouse over the icons on the left until you find extensions.

Search for the Python plugin as depicted in the screenshot below. You can either browse the list or type the name in the search field at the top.