Activation context for a component?

2010-02-14 Thread Nathan Kopp
Hey everyone! I'm new here and I apologize for not lurking long before posting. I have searched the archives extensively for this issue and have come up empty, so I'm just going to jump in. I'm working in Tapestry 5.1.0.5 (the latest as far as I know) and I've run into a use-case that I can't

Re: [T5.2.0-SNAPSHOT] Block to String without MarkupWriter

2010-02-14 Thread Howard Lewis Ship
Nothing obvious is coming to me; a Block represents markup and needs a markup writer to execute. You can leverage some of the internal Tapestry services to render the content much like a partial page render, perhaps. On Thu, Jan 28, 2010 at 8:36 AM, Joost Schouten (ml) joost...@jsportal.com

Re: Activation context for a component?

2010-02-14 Thread P . Stavrinides
Hi Nathan, Anywhere on my site where I display a Donkey, I want to let the user edit the instance. So then do exactly that... Tapestry 5 is sometimes simpler than you might expect, to me it seems you are complicating things unnecessarily... why are you creating new instances when editing?

Re: Activation context for a component?

2010-02-14 Thread Thiago H. de Paula Figueiredo
Hi! On Sun, 14 Feb 2010 00:10:02 -0200, Nathan Kopp nathan.k...@gmail.com wrote: There are two things preventing this from working: First, Tapestry does not replay the loop when processing the eventlink (like it did in the 3.x and 4.x days). Therefore, fooLoopVar never gets changed

Re: Activation context for a component?

2010-02-14 Thread Nathan Kopp
Good thought, but the copy of Donkey isn't the problem. (Actually, when using Hibernate or JPA, you'd want to make a copy to allow cancel to work, because altering a JPA entity by binding directly to the real instance could cause the database to be updated immediately... JPA/Hibernate doesn't

Re: Activation context for a component?

2010-02-14 Thread Thiago H. de Paula Figueiredo
On Sun, 14 Feb 2010 17:33:43 -0200, Nathan Kopp nathan0...@nathankopp.com wrote: Even with your suggested change, this would still generate a null pointer exception (NPE) in the same place, since the problem lies with fooLoopVar being null when the eventlink is click is processed by