Developer Getting Started Guide

This guide explains how to set up the recommended developer setup on a Linux machine. Other development setups are possible, but this one provides a good starting point.

  • Install VS Code and Inmanta extension.

  • Setting up Python virtual environments.

  • Setting up a project.

  • Set project sources

  • Setting up a module

  • Run tests

  • Module developers guide

  • Required environment variables

The examples below are using pip your system might require you to use pip3.

Install VS Code and Inmanta extension

The developer setup is based on VSCode with the Inmanta extension.

In order to install VS Code, you can refer to this page.

Inmanta’s extension in VS Code marketplace can be found here.

Further information about Inmanta VS Code extension is available on this page.

Setting up Python virtual environments

For every project that you work on, we recommend using a new virtual environment. If you are unfamiliar with venv’s, you can check out this page.

To create a virtual environment:

python3 -m venv ~/.virtualenvs/my_project

Then activate it by running:

source ~/.virtualenvs/my_project/bin/activate

Upgrading your pip will save you a lot of time and troubleshooting.

You can do so by running:

pip install --upgrade pip wheel

Setting up a project

At the time of this writing, linting and code navigation in IDEs work only if you have a project, so even if you only work on a single module, it is best to have a project.

There are two scenarios:

  1. Working on a New Project.

  2. Working on an Existing Project.

Working on a New Project

To create a new project you need to install some essential packages as follows:

pip install inmanta-core pytest-inmanta

Create a new project using the inmanta-project-template:

pip install cookiecutter

cookiecutter https://github.com/inmanta/inmanta-project-template.git

Navigate into the project and install the module dependencies using the inmanta CLI tool:

cd <project_name>

inmanta project install

V1 modules will be downloaded to the downloadpath configured in the project.yml file. V2 modules are installed in the active Python environment. For more details go here. Once you are done with creating a project, you can open VS Code by running:

code .

Working on an Existing Project

When working on an existing project, you need to clone them first:

git clone <project_url>

They also come with a requirements.dev.txt to install the development dependencies:

cd <project_name>

pip install -r requirements.dev.txt

The module dependencies are installed using the inmanta CLI tool:

inmanta project install

Set project sources

When starting a new project, the next step is to set the sources of your project so that it knows where to get its required modules from.

V1 module source

If you only use opensource v1 modules as provided by Inmanta, you can skip below step.

  1. Find the module you want to work on

  2. Copy the SSH URL of the repo

  3. In your VS code, open the project.yml file and under repo:, add the copied line there but keep in mind to replace the name of a specific module with a place holder, like below example:

code project.yml

repo:
    - url: git@code.inmanta.com:example/my_module.git
      type: git

Becomes:

repo:
    - url: git@code.inmanta.com:example/{}.git
      type: git

  • Now, in your main.cf file, if you import a module like, import <my_module> and save the file, you can get code completion. If you are working on an existing project with a populated main.cf file, code completion will work as expected.

Please note, code completion and navigation work on modules that are imported in the main.cf file.

Pip index for V2 modules and V1 modules’ dependencies

Add the pip index where your modules and dependencies are hosted to project.yml in the pip.index-url section. For example, for modules hosted on PyPi:

pip:
    index-url: https://pypi.org/simple/

Setting up a module

Like projects, there are also two scenarios:

  1. Working on a New Module.

  2. Working on an Existing Module.

Working on a New Module

Same as Working on a New Project part, modules can also be created like:

pip install cookiecutter
cookiecutter --checkout v1 https://github.com/inmanta/inmanta-module-template.git

for a v1 module. If you want to use the module in a project, make sure to put it in the project’s module path.

For a v2 module, use the v2 cookiecutter template, then install the module:

pip install cookiecutter
cookiecutter https://github.com/inmanta/inmanta-module-template.git
pip install -e ./<module-name>

This will install a Python package with the name inmanta-module-<module-name> in the active environment.

If you want to use the v2 module in a project, make sure to set up a v2 module source as outlined in the section above, then add the module as a dependency of the project as described in Working on an Existing Module. The location of the module directory is not important for a v2 module.

For more information on how to work on modules, see Understanding Modules and the module template documentation.

Working on an Existing Module

Modules that you want to work on, have to be added to your Inmanta project using the following command. This command also installs the module into the project.

inmanta module add --v1 <module-name>

for a v1 module or

inmanta module add --v2 <module-name>

for a v2 module. The latter will implicitly trust any Python package named inmanta-module-<module-name> in the project’s configured module source.

When starting to work on an existing module, it is recommended to check the readme.md file that comes with the module to see the instructions on how to install and use them.

Running Test

To run test on modules, it is recommended to set the INMANTA_TEST_ENV environment variable to speed up your tests and avoid creating virtual environments at each test run.

  1. Create a temp directory and export the path:

export INMANTA_TEST_ENV=$(mktemp -d)

  1. Install required dependencies

pip install -r requirements.txt -r requirements.dev.txt

  1. Run the test

python -m pytest tests