.. _sec_installation:

############
Installation
############

There are two basic options for installing ``msprime``: either through
pre-built binary packages using :ref:`sec_installation_conda` or
by compiling locally using :ref:`sec_installation_pip`. We recommend using ``conda``
for most users, although ``pip`` can be more convenient in certain cases.


.. _sec_installation_requirements:

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

Msprime requires Python 2.7 or Python 3.3+
and the `GNU Scientific Library <http://www.gnu.org/software/gsl/>`_.
GSL can either be installed automatically as part of a
:ref:`sec_installation_conda` installation **or** via system packages
(see :ref:`sec_pip_platform_specific_installation`)

.. warning::
    If you installed Python using anaconda/miniconda, please install
    msprime using ``conda``. Link/load time errors will occur if you mix
    a conda installed Python with system libraries.


.. _sec_installation_conda:

=====
conda
=====

Pre-built binary packages for ``msprime`` are available through
`conda <https://conda.io/docs/>`_, and built using `conda-forge <https://conda-forge.org/>`_.
Packages for Python 2.7, 3.5 and 3.6 are available for Linux and OSX.
For Windows, only Python 3.5 and 3.6 are supported.

We strongly recommend using Python 3.

***********
Quick Start
***********

1. Install ``conda`` using `miniconda <https://conda.io/miniconda.html>`_.
   Make sure you follow the instructions to fully activate your ``conda``
   installation!
2. Set up the `conda-forge channel <https://conda-forge.org/>`_ using
   ``conda config --add channels conda-forge``.
3. Install msprime: ``conda install msprime``.
4. Try it out: ``msp --version``.


There are several different ways to obtain ``conda``. Please see the
`anaconda installation documentation <https://docs.anaconda.com/anaconda/install/>`_
for full details.


************
Full example
************

In this example we create a
`conda environment <https://conda.io/docs/user-guide/tasks/manage-environments.html>`_
and install ``msprime`` into it on an OSX machine.
We assume that ``conda`` has been installed  and bash shell is being used (Windows users will need to modify the
commands slightly).

.. code-block:: none

    $ conda create --name msprime-env
    Solving environment: done

    ## Package Plan ##

      environment location: /Users/jk/miniconda3/envs/msprime-env


    Proceed ([y]/n)? y

    Preparing transaction: done
    Verifying transaction: done
    Executing transaction: done
    #
    # To activate this environment, use
    #
    #     $ conda activate msprime-env
    #
    # To deactivate an active environment, use
    #
    #     $ conda deactivate

    $ source activate msprime-env
    (msprime-env) $ conda install -c conda-forge msprime
    Solving environment: done

    ## Package Plan ##

      environment location: /Users/jk/miniconda3/envs/msprime-env

      added / updated specs:
        - msprime


    The following NEW packages will be INSTALLED:

        ca-certificates: 2018.1.18-0           conda-forge
        certifi:         2018.1.18-py36_0      conda-forge
        gsl:             1.16-0                conda-forge
        hdf5:            1.10.1-2              conda-forge
        intel-openmp:    2018.0.0-h8158457_8
        libgfortran:     3.0.1-h93005f0_2
        mkl:             2018.0.1-hfbd8650_4
        msprime:         0.5.0b2-py36_3        conda-forge
        ncurses:         5.9-10                conda-forge
        numpy:           1.14.1-py36h8a80b8c_1
        openssl:         1.0.2n-0              conda-forge
        pip:             9.0.1-py36_1          conda-forge
        pyparsing:       2.2.0-py36_0          conda-forge
        python:          3.6.4-0               conda-forge
        readline:        7.0-0                 conda-forge
        setuptools:      38.5.1-py36_0         conda-forge
        six:             1.11.0-py36_1         conda-forge
        sqlite:          3.20.1-2              conda-forge
        svgwrite:        1.1.12-py_0           conda-forge
        tk:              8.6.7-0               conda-forge
        wheel:           0.30.0-py36_2         conda-forge
        xz:              5.2.3-0               conda-forge
        zlib:            1.2.11-0              conda-forge

    Proceed ([y]/n)? y

    Preparing transaction: done
    Verifying transaction: done
    Executing transaction: done
    (msprime-env) $ python
    Python 3.6.4 | packaged by conda-forge | (default, Dec 23 2017, 16:54:01)
    [GCC 4.2.1 Compatible Apple LLVM 6.1.0 (clang-602.0.53)] on darwin
    Type "help", "copyright", "credits" or "license" for more information.
    >>> import msprime
    >>> tree = msprime.simulate(5).first()
    >>> print(tree.draw(format="unicode"))
      8
    ┏━┻━━┓
    ┃    7
    ┃  ┏━┻━┓
    ┃  ┃   6
    ┃  ┃  ┏┻┓
    ┃  5  ┃ ┃
    ┃ ┏┻┓ ┃ ┃
    1 0 4 2 3


Please see the `conda documentation <https://conda.io/docs/index.html>`_ for
more details on managing packages and environments.


.. _sec_installation_pip:

===
pip
===

Installing using ``pip`` is more flexible than ``conda`` as it
can support more versions of Python, and the locations of the
various dependencies can be specified.

Installing via ``pip`` is the recommended method when using the
system provided Python installations.

.. _sec_installation_system_requirements:

******************************
Installing system requirements
******************************

When installing via ``pip``, GSL must be installed using system packages
(we do not recommend installing GSL from source). GSL packages are available
on all major platforms.

For example, to install on Debian/Ubuntu use (as root)::

    $ apt-get install python-dev libgsl0-dev

For Redhat/Fedora use::

    $ yum install gsl-devel

On FreeBSD we can use ``pkg`` to install the requirements::

    $ pkg install gsl

To install the dependencies on OS X, we can use `Homebrew <http://brew.sh/>`_::

    $ brew update
    $ brew install gsl


************
Installation
************

We can install ``msprime`` easily using pip::

    $ python -m pip install msprime

(It is generally better to use ``python -m pip`` rather than call ``pip``
directly since this allows you to control which installation of Python the
package is installed to.) This will work in most cases, once the GSL has been
installed. See below for platform specific build instructions when this fails.

If you do not have root access to your machine, you can install
``msprime`` into your local Python installation as follows::

    $ python -m pip install msprime --user

To use the ``mspms`` program you must ensure
that the ``~/.local/bin`` directory is in your ``PATH``, or
simply run it using::

    $ ~/.local/bin/mspms

To uninstall ``msprime``, simply run::

    $ python -m pip uninstall msprime


.. _sec_pip_platform_specific_installation:

******************************
Platform specific installation
******************************

This section contains instructions to build on platforms
that require build time flags for GSL.

+++++++++++++
Debian/Ubuntu
+++++++++++++

To install and run ``msprime`` on a fresh Ubuntu 15.10 installation, do the
following:

.. code-block:: bash

    $ sudo apt-get install pkg-config python-dev python-pip libgsl0-dev
    $ pip install msprime --user
    $ ~/.local/bin/mspms 2 1 -t 1
    /usr/local/bin/mspms 2 1 -t 1
    5338 8035 23205

    //
    segsites: 3
    positions: 0.014 0.045 0.573
    100
    011


++++++++++++
FreeBSD 10.0
++++++++++++

Install the prerequisitites, and build ``msprime`` as follows::

    $ pkg install gsl
    $ CFLAGS=-I/usr/local/include LDFLAGS=-L/usr/local/lib pip install msprime

This assumes that root is logged in using a bash shell. For other shells,
different methods are need to set the ``CFLAGS`` and ``LDFLAGS`` environment
variables.

+++++++++++++
OS X Homebrew
+++++++++++++

We recommend using :ref:`sec_installation_conda` to install ``msprime`` on OS X.
However, it is also possible to install using Homebrew to satisfy the
GSL dependency.

First, ensure that Homebrew is installed and up-to-date::

    $ brew update

We need to ensure that the version of Python we used is installed via Homebrew
(there can be linking issues if we use the built-in version of
Python or a version from Anaconda). Therefore, we install Python 3 using
homebrew::

    $ brew install python3
    $ pip3 install --upgrade pip setuptools

The previous step can be skipped if you wish to use your own Python installation,
and already have a working pip.

Now install the dependencies and msprime::

    $ brew install gsl
    $ pip3 install msprime

Check if it works::

    $ mspms 10 1 -T