Local Installation

Install dependencies, e.g. via brew.

brew install postgresql memcached libmemcached npm

Register postgresql as a service

brew services start postgresql

Clone this repository:

git clone https://github.com/ghga-de/datameta.git

Change directory into your newly created project if not already there. Your current directory should be the same as this README.md file and setup.py.

cd datameta

Create a Python virtual environment, if not already created.

This can be done via venv:

python3 -m venv <environment_name>

or conda:

conda create -y -n <environment_name> 'python>3'

Activate the environment

With venv (docs):

source <path/to/environment>/bin/activate

With conda

conda activate <environment_name>

Upgrade packaging tools, if necessary.

pip install --upgrade pip setuptools

Install NPM dependencies

npm install --prefix datameta/static/

Install the project in editable mode with its testing requirements.

pip install -e ".[testing]"

Create a postgresql database, then add the path to the database to the development.ini found in datameta/config, e.g. sqlalchemy.url = postgresql://localhost/<dbname>. A copy of the development.ini can be placed at an arbitrary location named /path/to/config/ in the following.

createdb <dbname>

Initialize the database using Alembic.

<path/to/environment>/bin/alembic -c <path/to/config>/development.ini upgrade head

Load default data into the database using a script. Change the initial user and group information to your requirements.

<path/to/environment>/bin/initialize_datameta_db \
        -c <path/to/config>/development.ini \
        --initial-user-fullname "First User Fullname" \
        --initial-user-email "first@user.email" \
        --initial-user-pass "initialPassword" \
        --initial-group "My Organization"

Run your project’s tests.


Start the memcached process.

nohup memcached &

Run your project.

<path/to/environment>/bin/pserve /path/to/config/development.ini