.. -*- mode: rst -*-

=========
 INSTALL
=========

This document should help you to install the **GromacsWrapper**
package. Please raise and issue in the `Issue Tracker`_ if problems
occur or if you have suggestions on how to improve the package or
these instructions.

.. _Issue Tracker: https://github.com/Becksteinlab/GromacsWrapper/issues
.. _setuptools: https://pypi.org/project/setuptools/


Quick installation instructions
===============================

The latest release can be directly installed
from the internet::

  pip install GromacsWrapper

This will automatically download and install the `latest version of
GromacsWrapper from PyPi`_.

.. _`latest version of GromacsWrapper from PyPi`:
   https://pypi.org/project/GromacsWrapper/

Manual Download
===============

If your prefer to download manually, get the latest stable release
from https://github.com/Becksteinlab/GromacsWrapper/releases and
either ::

  pip install GromacsWrapper-0.7.0.tar.gz

or install from the unpacked source::

  tar -zxvf GromacsWrapper-0.7.0.tar.gz
  cd GromacsWrapper-0.7.0
  python setup.py install



Source code access
==================

The tar archive from https://github.com/Becksteinlab/GromacsWrapper/releases
contains a full source code distribution.

In order to follow code development you can also browse the code
**git** repository at https://github.com/Becksteinlab/GromacsWrapper or
clone the git repository from

   git://github.com/Becksteinlab/GromacsWrapper.git
 
and checkout the *master& branch::

   git clone https://github.com/Becksteinlab/GromacsWrapper.git
   cd GromacsWrapper


Requirements
============

Python_ 2.7.x or Python >= 3.4 and Gromacs_ (4.6.x, 2016, 2018) must
be installed. ipython_ is very much recommended. These packages might
already be available through your local package manager such as
``aptitude/apt``, ``yum``, ``yast``, ``fink`` or ``macports``.

.. _Python: http://www.python.org
.. _Gromacs: http://www.gromacs.org
.. _ipython: http://ipython.scipy.org


System requirements
-------------------

Tested with Python 2.7.x and Python 3.5/3.6 on Linux and Mac
OS X. Earlier Python versions are not supported.

.. Note:: Python 3 support is currently in alpha state; in principle
  it is fully supported but if you find bugs please report them
  through the `Issue Tracker`_.


Required Python modules
-----------------------

The basic package makes use of numpy_ and numkit_ (which uses scipy_);
all dependencies are installed during a normal installation process.

.. _numpy: http://numpy.scipy.org
.. _numkit: https://github.com/Becksteinlab/numkit
.. _scipy: https://www.scipy.org/scipylib/index.html