ESA Sen4CAP Client

Python client for the processing service of the ESA Sen4CAP project

---

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

Installation

Using pip

The sen4cap-client package is available on PyPI, and can be installed with pip:

pip install sen4cap-client

Using conda/mamba

The sen4cap-client package is available on conda-forge and can be installed using mamba or conda. To install into an existing, activated conda environment:

mamba install --channel conda-forge sen4cap-client

To create and activate a new environment containing sen4cap-client:

mamba create --channel conda-forge --name sen4cap-client sen4cap-client
mamba activate sen4cap-client

Using pixi

You can use the pixi package manager to install the conda-forge package:

mkdir sen4cap-test
cd sen4cap-test

pixi init
pixi add python
pixi add sen4cap-client
pixi shell

Pixi can also install the PyPI package:

mkdir sen4cap-test
cd sen4cap-test

pixi init
pixi add python
pixi add --pypi sen4cap-client
pixi shell

Using GitHub

To install and use the sen4cap-client package from its 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 sen4cap-client
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

Getting started

After installing the sen4cap-client package in your Python environment and activating it (conda/mamba: conda activate <your-env>, pixi: pixi shell) make sure the sen4cap-client command-line tool is accessible: Type

sen4cap-client --help

to get an overview of the available commands and options. The first step is to configure the client, which will also serve as default configuration for the client's Python API and its GUI:

sen4cap-client configure

List the available processes of the Sen4CAP processing service:

sen4cap-client list-processes

Development

Install the sen4cap-client as described in Installation / Using GitHub above.

Linting and Testing

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

Implementing Enhancements

The sen4cap-client code relies heavily on the Eozilla packages

• cuiman, which provides the client implementation, and
• gavicore which provides common OGC model classes and basic utilities.

Should sen4cap-client require non-Sen4CAP-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 into the project root:

git clone https://github.com/Sen4CAP/sen4cap-client.git
cd sen4cap-client
git clone https://github.com/eo-tools/eozilla.git
pixi install
pixi run tests

The resulting folder structure:

    ${projects}/
    └── sen4cap-client/
        ├── ...
        └── eozilla/
            ├── cuiman/
            ├── gavicore/
            └── ...

Then, during development, change sen4cap-client/pyproject.toml as follows

1. Comment out the dependencies cuiman and gavicore in the [project.dependencies] table.

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

Then run once more

pixi install

Integration tests

This is how to run a simple server integration test:

git clone https://github.com/Sen4CAP/sen4cap-client.git
cd sen4cap-client
git clone https://github.com/eo-tools/eozilla.git
pixi install
pixi shell
sen4cap-client configure
pytest -s scripts/integration_test.py

Running the client in a remote VM (Windows 11)

JupyterLab

On the remote VM start JupyterLab without opening a browser and listening on all interfaces:

cd sen4cap-client
pixi shell
jupyter lab --no-browser --ip=0.0.0.0 --port=8888

It will print something like http://127.0.0.1:8888/lab?token=1e751cd... - keep this running. On your local desktop machine (e.g., Windows or WSL2) SSH into the VM with port forwarding:

ssh -L 8888:localhost:8888 user@remote-vm

Now open your local browser:

http://127.0.0.1:8888/lab?token=1e751cd20cd...