Skip to content

Latest commit

 

History

History
103 lines (72 loc) · 3.42 KB

DEVELOPMENT.md

File metadata and controls

103 lines (72 loc) · 3.42 KB
Raw

Developing HADDOCK3

This file provides information on how to setup a development environment for HADDOCK3.

System requirements

  • Python 3.9
  • OpenMPI

Installing system dependencies

Below the instructions are provided for a Ubuntu system using apt-get. If you are using a different system, please refer to the respective package manager - yum, dnf, pacman, homebrew, port, etc.

sudo apt-get update &&
    # Needed for building Python
    sudo apt-get install build-essential python-setuptools python-pip &&
    sudo apt-get install libncursesw5-dev libgdbm-dev libc6-dev &&
    sudo apt-get install zlib1g-dev libsqlite3-dev tk-dev &&
    sudo apt-get install libssl-dev openssl &&
    sudo apt-get install libffi-dev &&
    # Needed for haddock3 installation
    sudo apt-get install git gcc &&
    # Needed to run the MPI related tests
    sudo apt-get install openmpi-bin libopenmpi3 libopenmpi-dev

After all the system-dependencies are in place, download and compile python.

Conda is NOT recommended for development

wget https://www.python.org/ftp/python/3.9.6/Python-3.9.6.tgz
tar -xf Python-3.9.6.tgz
cd Python-3.9.6
./configure --enable-optimizations
sudo make altinstall -j 8

Then python3.9 should be available on your system at /usr/local/bin/python3.9

On OSX, you can use a package-manager such as brew to install Python 3.9.

Please keep in mind installing python with a package manager can mask system dependencies that your development might add. That's why we recommended you install it from source.

brew install [email protected]

Setting up the development environment

Clone the repository

git clone https://github.com/haddocking/haddock3.git
cd haddock3

Python environment

We recommend you use Python's native virtual environment to manage the dependencies.

python3.9 -m venv .venv
source .venv/bin/activate

Install haddock3 in development mode

Install both project dependencies and test dependencies using pip.

pip install -e '.[dev,docs]'

If you are using a Mac, if the installation of mpi4py fails, run first brew install mpi4py

Running tests

In haddock3 we use the pytest framework, make sure you check it's documentation the tests are located in tests/ (unit) and integration_tests/ directories.

pytest tests/
pytest integration_tests/

Installation in an HPC environment

Please get in contact with the system administrator before doing development in a shared HPC environment.

For installation in an HPC environment we recommend to check the installed Python versions on the system and also importantly if an openmpi (or other custom MPI) installation is available on the system. Those are often offered via the module command. If you only intend to develop haddock3 using the multiprocessing scheduler, the above instructions should be fine. But to harvest the MPI capabilities of an HPC system it is best to build haddock3 using the installed MPI version on the HPC system.