*Scripting Chart*
Hi,
One of the major limitations of the Chart module is the inability to
script and automate many of its processes. A direct consequence of this
is the inability for common users to easily extend the functionality of
Chart.
CURRENT LIMITATIONS
====================
- monolithic code: everything is hardcoded inside Chart
- inability to write scripts for Chart
-- no possibility for automation
- inability to easily extend functionality
-- this can be performed only by hardcoding new functions inside OOo
-- only a limited number of programmers can do this
-- it takes eons to implement a new feature
-- when implementing new functionality: code becomes bigger, more
complicated
-- therefore we have to deal with the many limitations of such code
(difficult to debug and maintain)
SOLUTION
=========
- break monolithic structure
- layer the Chart module (Virtual Machine/ internal scripts)
- implement a powerful scripting engine
- implement the various functions as internal scripts
1. Implement a powerful mathematical and graphical (charting) scripting
language. Actually there are a number of open source programs that have
such scripting capabilities.
- mathematically oriented scripting: octave, gnuplot, R
- scientific annotation: asymptote – this is a graphics annotation
language and is highly specialized for this task
-- see the Chart wiki page and http://sourceforge.net/projects/asymptote/
2. Internally implement the functions as scripts. Ease the creation of
new functions/ features using the scripting engine.
ADVANTAGES
============
- *many users* would be able to extend the functionality of Chart
- no need for extensive C++/ OOo coding knowledge
-- NOT limited anymore to OOo staff
-- implementing new features would be much more easy and much more
*faster*
- slim code, easier to debug; only users needing those features would
download them;
- scripting would allow automation of many processes; (especially in
enterprise environments)
The layering of the chart application could allow many more users to
write extensions without having extensive knowledge of OOo. This is
beautifully illustrated in the R statistical program. R itself is more
of an environment and it builds heavily on the 500+ packages, many
written in the R-language itself (actually S+ language).
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
