[Pymerase-cvs-commit] pymerase INSTALL,NONE,1.1

[<ki...@us...>]

Update of /cvsroot/pymerase/pymerase
In directory sc8-pr-cvs1:/tmp/cvs-serv1440

Added Files:
	INSTALL 
Log Message:
Some how was missing form cvs


--- NEW FILE: INSTALL ---
* How to install?
-----------------

Pymerase depends on several components depending on what you'd like to
do with it.


** GeneX Table Definition Format
--------------------------------

Currently the GeneX Table definition format is the best supported file
format (As it was the first mode I started developing against.)

However the XMI format is coming along as well.

At minimum it requires an XML parser. The default python XML parser is
called expat, newer versions are available at
http://sourceforge.net/projects/pyxml. Later versions of this might
work with the current table.dtd

However the version of expat I was using didn't properly support
entities with the XML files were using so I ended up using the
xerces-c XML parser. 

A project called pirxx is providing the python wrapper for xerces-c.

xerces-c: http://xml.apache.org/xerces-c/index.html
pirxx: http://sourceforge.net/projects/pirxx

** XMI file format
------------------

The key thing for XMI files is you need to generate them somehow,
ArgoUML is the tool we're currently using for testing. 

http://argouml.tigris.org

The current XMI parser was written by Teemu Toivanen and is available
from the MGED/Mage sourceforge site in the directory lib/Python/pyXMI.

** DB Support
-------------

The code generated by CreateDBAPI, needs the mxTools Date Time
extension as well as the postgresql python db api module.

The mx tools is available from:
http://www.egenix.com/files/python/mxDateTime.html 

The database interface, pgdb, is distributed with postgresql
(http://www.postgresql.org) 

Also, the SQL code generated by CreateSQL is using postgresql's sequence
feature, and so is currently dependent on postgresql.