commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dmitri Plotnikov <dplot...@yahoo.com>
Subject Re: Proposal: JSP/JSTL expression language
Date Thu, 25 Jul 2002 16:22:37 GMT
Craig,

--- "Craig R. McClanahan" <craigmcc@apache.org> wrote:
> 
> 
> On Thu, 25 Jul 2002, Dmitri Plotnikov wrote:
> 
> > Date: Thu, 25 Jul 2002 08:56:16 -0700 (PDT)
> > From: Dmitri Plotnikov <dplotnik@yahoo.com>
> >
> > --- "Geir Magnusson Jr." <geirm@adeptra.com> wrote:
> > > On 7/20/02 6:11 AM, "Stephen Colebourne"
> > > <scolebourne@btopenworld.com>
> > > wrote:
> > >
> > > > I knew I had a sense of deja vu. So the question becomes, can
> the
> > > JSP EL and
> > > > JEXL projects become simply plugins to, or part of the JEX
> project?
> > >
> > > I don't think that's the right way to go - I think that JEX
> should
> > > (if it
> > > wants to) support all of these things.  But my feeling is that
> any
> > > kind of
> > > scriptish project should be independent (for purposes of
> > > innovation...)
> > >
> >
> > I agree with Geir that individual languages should not be
> constrained
> > by something like JEX and therefore should not be mere plug-ins.
> > However, it might be a good idea to get them to conform to it.
> >
> > Contributing JEX I was thinking: with all these expression
> languages
> > popping up, each framework that needs one has to either choose a
> > specific language or have a pluggability mechanism.  An example of
> the
> > former approach is JSTL, which after several pre-releases allowing
> > multiple languages ended up with just one - EL.  An example of the
> > latter approach is Jelly that has a pluggability mechanism similar
> to
> > JEX.  To me this clearly calls for a set of common API/SPI.
> >
> 
> Having one and only one EL was always the goal for JSTL and JSP --
> the
> reason for the plugability in the early releases was solely to help
> people
> evaluate the alternatives that were being considered.

Sure, I know that, and IMO it is a good thing that they pegged a
specific language (I am not too ecstatic about the language itself
though).  This decision will allow them to _compile_ expressions to
Java, which is fantastic.

> > Of course, the "right" way to introduce such APIs would be a JSR
> > similar to JAXP in the XML parser world.  But I have no idea how to
> get
> > from here to there.
> >
> 
> Even without a JSR to formalize this (you can certainly do that if
> you
> want, by following the procedures on http://jcp.org), you might
> consider
> trying to establish a de facto standard first.
> 
> You would need to do something like what commons-logging did --
> define a
> common API, and a factory that followed a JAXP-style discovery
> mechanism,
> and then write some adapters for existing EL interpreters.  The
> "discovery" part of this is even easier now, because it's been
> abstracted
> out into the "discovery" package that you can just reuse.
This is a very good description of JEX.  This is exactly what it does
and what it is for.  

I haven't looked at the "discovery" package yet, but certainly will.

> > One more point. I wanted to make sure it is understood that I don't
> > really see JEX as "my" project and do not want to control its
> destiny.
> > I completely understand how considerations of ownership can stand
> in
> > its way.  I cordially invite everybody who has interest in the
> > unification of expression language APIs to join me as a
> co-contributor
> > on the JEX project.
> >
> > [snip]
> >
> > - Dmitri
> >
> 
> Craig
> 
- Dmitri

__________________________________________________
Do You Yahoo!?
Yahoo! Health - Feel better, live better
http://health.yahoo.com

--
To unsubscribe, e-mail:   <mailto:commons-dev-unsubscribe@jakarta.apache.org>
For additional commands, e-mail: <mailto:commons-dev-help@jakarta.apache.org>


Mime
View raw message