Installing pyfusion on CentOS 4.8¶

Release:	0.81
Date:	February 01, 2017

Because CentOS 4.8 has an old version of python (version 2.3), which won’t work with some of the libraries required by pyfusion, we will install a newer version of python and some libraries in the user’s home directory.

Environment variables¶

First, set up some environment variables:

export PYTHONPATH=$PYTHONPATH:$HOME/code/python
mkdir -p code/python

export SOURCEDIR=$HOME/source
mkdir $SOURCEDIR
export LOCALDIR=$HOME/local
mkdir $LOCALDIR

export PATH=$LOCALDIR/bin:$PATH

The PYTHONPATH and PATH (and LOCALDIR if PATH refers to it) exports should also go in your $HOME/.bashrc file.

Python¶

Now install Python - the version number is the latest in December 2010, you can use a later one if it exists. If you wish to use Python 3.x, check you have the latest pyfusion download from Boyd’s fork, and veiw the issues file

cd $SOURCEDIR
wget http://www.python.org/ftp/python/2.7.1/Python-2.7.1.tar.bz2
tar -xjf Python-2.7.1.tar.bz2
cd Python-2.7.1
./configure --prefix=$LOCALDIR
make
make install

Make sure the default python is now python2.7 (it should be this because $LOCALDIR/bin is first in your PATH environment variable):

> python

Python 2.7.1 (r271:86832, Dec 26 2010, 03:33:20)
[GCC 3.4.6 20060404 (Red Hat 3.4.6-11)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>>

Setuptools¶

We need setuptools so we can install pip:

cd $SOURCEDIR
wget http://pypi.python.org/packages/2.7/s/setuptools/setuptools-0.6c11-py2.7.egg
sh setuptools-0.6c11-py2.7.egg --prefix=$LOCALDIR

pip¶

We use pip to install other libraries:

cd $SOURCEDIR
wget http://pypi.python.org/packages/source/p/pip/pip-0.8.2.tar.gz
tar -xzf pip-0.8.2.tar.gz
cd pip-0.8.2
python setup.py install --prefix=$LOCALDIR

numpy¶

Either using sudo or log in as root to install the dependencies:

yum install blas lapack

Then use pip to install numpy:

pip -v install numpy

sqlalchemy¶

Install using pip:

pip -v install sqlalchemy

ipython¶

Optional, but very useful:

pip -v install ipython

git¶

Git is used for pyfusion revision control, and makes it easy for you to update pyfusion and help with development:

cd $SOURCEDIR
wget http://kernel.org/pub/software/scm/git/git-1.7.3.4.tar.bz2
tar -xjf git-1.7.3.4.tar.bz2
cd git-1.7.3.4
./configure --prefix=$LOCALDIR
make
make install

pyfusion¶

We install with git:

cd $HOME/code/python
git clone git://github.com/dpretty/pyfusion.git
cd pyfusion
git checkout -b dev origin/dev

Setting up mysql¶

Because pyfusion uses sqlalchemy, you can choose from many different types of SQL servers, here we show how to set up MySQL.

As root (or sudo), install the required packages:

yum install mysql mysql-server mysql-devel

Still as root, start MySQL:

/etc/init.d/mysql start

and create a MySQL user for pyfusion:

mysql
> GRANT ALL PRIVILEGES ON *.* TO 'pyfusionuser'@'localhost' IDENTIFIED BY 'mypassword' WITH GRANT OPTION;

Now install (not as root) the python MySQL libraries:

pip -v install MySQL-python

and create a test database to use with pyfusion:

mysql -p
> create database pyfusion_test;

Now, edit $HOME/.pyfusion/pyfusion.cfg to tell pyfusion to use this database, if the directory doesn’t exist, make it:

mkdir $HOME/.pyfusion

and then in $HOME/.pyfusion/pyfusion.cfg:

[global]
database = mysql://pyfusionuser:mypassword@localhost/pyfusion_test

Matplotlib¶

You’ll also want matplotlib installed to visualise the data. As root, install the dependencies (we’ll use pyqt for the graphics backend):

yum install freetype-devel libpng-devel qt-devel

And as user:

cd $SOURCEDIR
# for some reason pip -v install matplotlib failed for me, so I downloaded the source file separately:
wget http://sourceforge.net/projects/matplotlib/files/matplotlib/matplotlib-1.0/matplotlib-1.0.0.tar.gz/download
pip -v install matplotlib-1.0.0

Now we setup the pyqt backend:

cd $SOURCEDIR
wget http://www.riverbankcomputing.com/static/Downloads/sip4/sip-4.12.tar.gz
tar -xzf sip-4.12.tar.gz
cd sip-4.12
python configure.py
make
make install

cd $SOURCEDIR
wget http://www.riverbankcomputing.co.uk/static/Downloads/PyQt3/PyQt-x11-gpl-3.18.1.tar.gz
tar -xzf PyQt-x11-gpl-3.18.1.tar.gz
cd PyQt-x11-gpl-3.18.1
export QTDIR=/usr/lib/qt-3.3
python configure.py
make
make install

and configure matplotlib to use this backend:

mkdir $HOME/.matplotlib
cp $HOME/local/lib/python2.7/site-packages/matplotlib/mpl-data/matplotlibrc .matplotlib/.

and edit $HOME/.matplotlib/matplotlibrc to use the setting:

backend      : QtAgg