Isn't there some sort of ExpressionEvaluator service in HiveMind now? Is that where you'd plug in another expression language?
-----Original Message----- From: Ben Eng [mailto:[EMAIL PROTECTED] Sent: Tuesday, August 08, 2006 3:33 PM To: Tapestry users Subject: Re: Tapestry 5 Discussions Howard, I was only offering up an idea in response to your comment about OGNL not providing an adequate solution for Tapestry 5. You seemed to be searching for an expression language that could translate into both server-side and client-side implementations. Ben On Tue, Aug 08, 2006 at 09:07:13AM -0700, Howard Lewis Ship wrote: > Everything will be pluggable, just like in Tapestry 4. Why not create this > for Tapesty 4 today? > > On 8/8/06, Ben Eng <[EMAIL PROTECTED]> wrote: > > > >Howard, > > > >How about adopting XPath expressions as an alternative to OGNL? Apache > >Commons JXPath could provide a server-side implementation, while > >something like Google AJAXSLT can implement XPath (as a part of XSLT) > >in JavaScript on the client side. See > >http://goog-ajaxslt.sourceforge.net/ > > > >Ben > > > >On Thu, Aug 03, 2006 at 11:47:46AM -0700, Howard Lewis Ship wrote: > >> As per my early blog post ( > >> > >http://howardlewisship.com/blog/2006/03/from-fanciful-ideas-category.html), > >> I would like to see object editting be dirt simple in Tapestry 5, using > >> built in components. I'll be discussing some of this with Chris Nelson > >this > >> weekend. > >> > >> I would like to see Trails5 be an Apache project next to Tapestry5. > >> > >> One thing I hope to do is bridge the gap between Tapestry components and > >the > >> entity objects that contain the properties being updated. To wit, a > >> component such as TextField should be able to see the setter method that > >it > >> will ultimately invoke, and be able to convert annotations (both defined > >by > >> Tapestry and defined by external sources such as EJB3 and Hibernate) > >into > >> server-side and client-side validation logic. > >> > >> The upside is automatic coordination of validation at multiple layers. > >> > >> The downside is that we may leave OGNL behind in the process, since its > >APIs > >> don't support anything like this. I'll be discussing that with Drew > >> Davidson next week. Of course, synthetic properties and instant class > >> reloading will reduce the necessity of OGNL. > > > >--------------------------------------------------------------------- > >To unsubscribe, e-mail: [EMAIL PROTECTED] > >For additional commands, e-mail: [EMAIL PROTECTED] > > > > > > > -- > Howard M. Lewis Ship > TWD Consulting, Inc. > Independent J2EE / Open-Source Java Consultant > Creator and PMC Chair, Apache Tapestry > Creator, Apache HiveMind > > Professional Tapestry training, mentoring, support > and project work. http://howardlewisship.com --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
