Python control layer for the synthetic EO scene simulator service of the ESA DTE-S2GOS project.

Note, this project and its documentation is still in an early development stage.

This project hosts two Python packages:

• s2gos-client - CLI, GUI, and API clients that allow for interaction with the S2GOS scene simulator service.
• s2gos-server - a FastAPI-based implementation of the OGC API - Processes that is wrapped around the S2GOS scene simulator service based on Apache Airflow.

The development of the S2GOS controller paved the way for the development of the more general Eozilla Suite of tools supporting the interaction with EO workflow orchestration and processing systems. The S2GOS controller therefore build on the Eozilla packages Cuiman, Wraptile and others.

The S2GOS controller packages are not yet deployed on PyPI, therefore installing it as a package using pip is not yet available.

The S2GOS controller packages are not yet deployed on conda-forge, therefore installing it using as a conda package using conda or mamba is not yet available.

The S2GOS controller packages are not yet deployed on conda-forge, therefore installing it as a conda package using pixi is not yet available.

To install the S2GOS controller packages from their sources on GitHub you'll need to install both git and pixi first. Then:

git clone https://github.com/eo-tools/sen4cap-client.git
cd s2gos-controller
pixi install
pixi shell

The installed development environment includes also JupyterLab so the recommended way to get started is to take a look at the notebooks in the notebooks folder.

cd notebooks
jupyter-lab

After installing the S2GOS controller packages in your Python environment and activating it (conda/mamba: conda activate <your-env>, pixi: pixi shell) make sure the respective command-line tools are accessible: Type

s2gos-server --help

and

s2gos-client --help

to get an overview of the available commands and options.

You can then run the S2GOS server with one of the development services or use the deployed server. To run the development service with a local process executor run

s2gos-server dev -- wraptile.services.local:service

To run the S2GOS gateway server with a local Airflow instance (assuming the local Airflow webserver runs on http://localhost:8080):

s2gos-server dev -- wraptile.services.airflow:service --airflow-password=abcd1234

The next step is configure the client, which will also serve as default configuration for the client's Python API and its GUI:

s2gos-client configure

Test:

s2gos-client list-processes

Install the S2GOS controller packages as described in Installation / Using GitHub above.

To run all checks, execute

pixi run checks

To run all tests, execute

pixi run tests

To generate a coverage report, execute

pixi run coverage

The S2GOS controller code relies heavily on the Eozilla packages

• wraptile, which provides the gateway server implementation,
• cuiman, which provides the client CLI, GUI, and API implementations, and
• gavicore which provides common OGC model classes and basic utilities.

Should S2GOS controller require non-S2GOS-specific enhancements it would likely be best to implement the required changes in the respective Eozilla packages. For this, check out the Eozilla sources directly next to this project's sources to achieve this folder structure:

    <projects>/
    ├── s2gos-controller/
    │   ├── s2gos-client/
    │   ├── s2gos-server/
    │   └── ...
    └── eozilla/
        ├── appligator/
        ├── cuiman/
        ├── gavicore/
        ├── procodile/
        ├── wraptile/
        └── ...

For development, change the root pyproject.toml file as follows

1. Comment out the dependencies cuiman, gavicore, etc. in the [tool.pixi.dependencies] table.

2. Uncomment the editable PyPI dependencies for cuiman, gavicore, etc. in the [tool.pixi.pypi-dependencies] table.

The run

pixi i

to make the changes effective. Check with

pixi ls

which should now list the Eozilla packages as editable.