On Fri, 30 Mar 2001, Mel Martinez wrote:
> --- Steve Downey <[EMAIL PROTECTED]> wrote:
>
> I must admit to not being totally sure whether
> co-opting the current 'Constants' and 'Options' class
> families is the best naming for this, but they do
> clearly indicate the difference in scope. I could
My preference would be to completely drop the Constants,
and keep each "constant" in the class where it belongs
( but this is personal taste ).
One big issue is that we should avoid over-engineering this.
We need to improve jasper and do a refactoring - as a number
of small code moves and interface changes. Then we can
evaluate the result and maybe repeat it.
The goal is to simplify and modularize, and IMHO the only
way that can be done right is via incremental steps, not a
"whaterfall" model ( and I don't want to start another
evolution/revolution, XP/waterfall flame war here :-).
I would sugest to move to a vote, create the proposal/jasper34,
do a first step ( like the JspServlet changes you want, or
the new packages - but with minimal interface changes ), evaluate
the result and maybe do another iteration.
Mel, please don't try to resolve all the problems in one step
or as a "block" design ( or a "perfect" initial design )
> That is always good stuff to hear! Costin deserves a
> lot of the credit here, since that performance gain is
> primarily (probably) coming from his JspInterceptor
> implementation. I know we can never get JspServlet to
Thanks - but I haven't started yet :-)
Jsp "real" performance tunning can start only after refactoring. The goal
of 3.3 was to make the code simple and modular enough so we can _start_
optimizing it.
Costin
