Contents

    1. Introduction
    2. Dependencies
    3. Installation
    4. Configuration file
    5. License
    6. Credits
    7. Contact

-------------------------------------------------------------------------------
1. Introduction

SciTools is a Python package containing many useful tools for scientific 
computing in Python. The package is built on top of other widely used 
packages such as NumPy, SciPy, ScientificPython, Gnuplot, etc.

-------------------------------------------------------------------------------
2. Dependencies

SciTools only requirements are Python (see http://python.org)
and Numerical Python (see http://numpy.org). In the past, SciTools
worked transparently with NumPy, Numeric, and Numarray, through its
numpytools module. Although this module still works, we recommend
users to use NumPy only.

-------------------------------------------------------------------------------
3. Installation

There are several different ways of installing SciTools:

 - Normal install ([, ... ] means optional)
  
     python setup.py install [,--prefix=$PREFIX]

   This install makes SciTools use Matplotlib as plotting engine, if a
   a working Matplotlib is found. Otherwise Gnuplot is used (which requires
   both the Gnuplot program in C and the Python interface Gnuplot.py to
   be installed). If neither Matplotlib nor Gnuplot is found, Matplotlib
   is still used as default plotting engine. (The rest of SciTools will
   work well, of course.)

   Installing Matplotlib is easy on most platforms: just download the
   tarball, pack it out, and run python setup.py install.

   You can explicitly specify the desired plotting engine on the command line,
   using the --easyviz_backend option:

     python setup.py install --easyviz_backend gnuplot [,--prefix=$PREFIX]

   The default plotting engine is specified in the file scitools.cfg in the
   scitools package directory, and the --easyviz_backend option leads to
   an automatic edit of the "backend" line in the [easyviz] section of 
   that configuration file.

   Don't edit the scitools.cfg file manually before installation (use
   the --easyviz_backend option) because setup.py will override your
   edits of the easyviz backend.  However, all other edits of the
   configuration file can be edited.  For example, the Matplotlib GUI
   is set to TkAgg (i.e., using Tkinter, which is standard in most
   Python installations). The GUI can be set to other values, such as
   Qt4Agg (provided you have Qt4 and a Python interface to it).

   After SciTools is installed, you may edit the installed version of
   scitools.cfg, or (better) have your own .scitools.cfg file in your
   home directory.

 - Setuptools using eggs

   First build the egg with the following command	
    
     python setupegg.py [, egg_info --tag-svn-revision ] bdist_egg

   The --easyviz_backend option can be used, as described above, to
   change the default plotting engine.

   Then install the created egg using easy_install

     easy_install [, --prefix=$PREFIX] dist/Scitools-0.7-py2.6.egg 

   (Version numbers of SciTools and Python in this filename may vary.)

 - Manual installation

   Put the lib subdirectory in a directory contained in PYTHONPATH,
   and the bin subdirectory in a directory contained in PATH.
   (This gives access to the directories for scitools, easyviz, etc.)
   Edit the scitools.cfg file to set the desired default plotting engine
   (look for the "backend =" line in the [easyviz] section).

 - Debian/Ubuntu packages

   If you have Debian, or a Debian based platform like Ubuntu, you can
   do 

     sudo apt-get install python-scitools

   as SciTools is now a part of Debian. You can also find newer Debian 
   packages for SciTools at googlecode.com.


 - Extended (not supported anymore)

   Run build_scitools++.py. This makes a new directory build/scitools++
   that you can move to any convenient place. scitools++ contains
   scitools plus some standard Python modules like Gnuplot, IPython,
   epydoc, etc. Put scitools++/lib in PYTHONPATH and scitools++/bin in
   PATH. Or you may run scitools++/setup.py to install scitools++ in the
   official installation directories on your system. 

-------------------------------------------------------------------------------
4. Configuration File

The behavior of many parts of SciTools and in particular the
subpackage Easyviz (for plotting) can be controlled in a configuration
file.  Please read the subsection "Setting Parameters in the
Configuration File" under the section "Advanced Easyviz Topics" in the
Easyviz tutorial (pydoc scitools.easyviz will show the tutorial). In
particular, if you use Matplotlib as the default plotting engine, you
may want to turn on the use of LaTeX for rendering legends, titles,
and numbers.  By default, LaTeX is turned off when SciTools is
installed. It is easy to change this by locating the scitools.cfg file
in the folder where the SciTools package is installed and edit the
line with text.usetex in the [matplotlib] section of this file. A
better solution is to copy the system scitools.cfg file to
.scitools.cfg in your home folder and edit that file. You can also
change the GUI used by Matplotlib. Users who do not apply Matplotlib
for plotting will seldom need to edit the configuration file.

-------------------------------------------------------------------------------
5. License

SciTools is licensed under the new BSD license, see the LICENSE file.

Lumpy.py and Gui.py are licensed under GPL, however, permission is
granted by Allen Downey to include these under a BSD license.

-------------------------------------------------------------------------------
6. Credits

SciTools was initially mainly developed by Hans Petter Langtangen
<hpl@simula.no> for his book "Python Scripting for Computational
Science" (Springer, 1st edition 2003, 3rd edition 2009). 
The Easyviz package was mainly developed by Johannes H. Ring 
<johannr@simula.no>. Johannes H. Ring has been the principal
maintainer of SciTools.

Some modules included in SciTools are written by others:

Allen B. Downey <downey@allendowney.com> wrote Lumpy.py and Gui.py
Imri Goldberg <lorgandon@gmail.com> wrote aplotter.py
Fred L. Drake, Jr. <fdrake@acm.org> wrote pprint2.py
Gael Varoquaux <gael.varoquaux@normalesup.org> wrote pyreport

Code contributors include:

Rolv E. Bredesen <rolveb@simula.no>
Joachim B. Haga <jobh@simula.no>
Mario Pernici <Mario.Pernici@mi.infn.it>
Ilmar Wilbers <ilmarw@simula.no>
Arve Knudsen <arvenk@simula.no>

-------------------------------------------------------------------------------
7. Contact

Comments, suggestions, bug fixes should be sent to

    scitools@googlegroups.com

The latest release of SciTools can be obtained from

    http://scitools.googlecode.com

while the latest Subversion revision can be obtained either anonymously 
(for non-members) by

    svn co http://scitools.googlecode.com/svn/trunk/ scitools

while SciTools members should use

    svn co https://scitools.googlecode.com/svn/trunk/ scitools --username USERNAME

See http://code.google.com/p/scitools/source/checkout for details.