Python Hyperschema 3.0 is placed in the public domain by Randy Phillips
(hyperschema@rebeccaphillips.com) because you will need to edit this 
code to suit your own tastes on the HTML output.  If I release it under
the GPL that would be a real pain in the neck, as the GPL requires that
you giving me every single one of your code changes.  If you find a bug
or would like to make a contribution to the code, please let me know.  

Bug fixes and enhancements are most welcome!  I would like to put on the
hyperschema web site a list of users, please let me know if you used this
public domain source code.  Email me with any questions or comments,
and enjoy!

Many thanks to Norman Dunbar, ndunbar@lynxfinancialsystems.co.uk, who
pointed out the reverse backslashes (now fixed) at the end of the SQL
files, and suggested changing the SQL schema owner name to USER (done).
Norman also provided cross-compilation advice for the C source code.

The C file, hyperschema.c, is no longer maintained in favor of the
Python version hyperschema.py, although it is included in the Hyperschema
distribution.


Instructions:

1) Run make_dump_table.sql to create a dummy table to hold the database
   metadata.

2) Run extract_schema.sql to populate the dummy table, this may take a
   while if you have many tables - anyone who can help tune this please try!
   Current schema owner is USER, change as needed.

3) set linesize to 300 and pages to 32000, spool on, and select * from
   the dummy table ordering by table_name and column_id to sort the spool file.

4) PYTHON: Run hyperschema.py as "hyperschema.py filename", where filename
      is the name of the spool file.  You can test on the included example
      file table.lst.  You can also run as "hyperschema.py" and be prompted
      for a file name, I found this handy for developement work within IDLE.

   C: Run the compiled hyperschema.c as "hyperschema filename", where filename
      is the name of the spool file.  You can test on the included example
      file table.lst.  Please note that after version 3.0, the C source code
      will not be maintained in favor of Python.

5) Please consider contributing to hyperschema, as Norman Dunbar has done.


File list:

SQL files used to extract database metadata:
     make_dump_table.sql
     extract_schema.sql

Example SQL output file:
     table.lst

Hyperschema-generating python source code file:
     hyperschema.py

Hyperschema-generating C source code file (deprecated):
     hyperschema.c

Javascript library for mouseovers:
     overlib.js

GIF file for Production Support Team:
     prodsup.gif

HTML output when hyperschema is run for the example file table.lst:
     index.html
     ANB_CONTEXT.html
     BNB_FUNCTION.html
     CNB_PARTY.html
     CNB_POSTAL_ADDRESS.html