Title: Message
|
Many WW2 tags seem to use the attributes 'name' and
'value' for non-evaluated and evaluated parameters respectively.
Wouldn't an evaluated 'name' attribute be a little confusing? Maybe some
other convention is required.
----- Original Message -----
Sent: Tuesday, October 21, 2003 8:29
PM
Subject: RE: [OS-webwork] ActionTag
changes
Rather than
supporting name and value, we’ll just make name an evaluated _expression_ –
unless 1.3 didn’t do that (we need to keep compatibility if
possible).
-----Original
Message-----
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of John Patterson
Sent: Tuesday, October 21, 2003 12:26
AM
To: [EMAIL PROTECTED]
Subject: Re: [OS-webwork] ActionTag
changes
Great. The Action tag will
be much more useful after those changes. One other improvement I would
like to suggest is to have a 'value' attribute in addition to the 'name'
attribute. Then I will eventually be able to replace all my
<ww:include's with <ww:action's
----- Original Message -----
Sent:
Tuesday, October 21,
2003 2:47
AM
Subject: RE:
[OS-webwork] ActionTag changes
-----Original
Message-----
From: Michal
Mosiewicz [mailto:[EMAIL PROTECTED]
Sent: Monday, October 20, 2003 3:11
PM
To: [EMAIL PROTECTED]
Subject: Re: [OS-webwork] ActionTag
changes
Is it correct
that the main difference between using the Action tag and the Include
tag for 'componentising' a page is that the Action tag uses the same
value stack?
It seems include
includes servlet's output (copying stream data) and it's really not
related to actions at all, while Action tag uses actions to eventually
dispatch to view components.
Does the Action
tag push the action onto the stack?
I hope that's a good time to make
sure Pat remembers my suggestions. IMHO Action tag executing results
should push the action onto the stack just before it does execute result,
and then it should pop the action off the stack as soon as result is
finished.
This way you would (logically)
access both actions in the included view.
The problem was that action
invocations was assumed never take actions off the stack, but I posted
apropriate patches to deal with that.
|
