You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@tapestry.apache.org by Paul Stanton <pa...@gunnsoft.com.au> on 2004/05/18 01:23:24 UTC
abstract Pages and property-specifications
i fail to see the benefit of using implied instance variables in the
form of abstract accessor methods and <property-specification> elements
in the .page file. It would seem more efficient to implement the
instance variables and getter/setter methods and detach method.
1. the number of files edited is greater (.java & .page)
2. the number of characters typed is greater (xml element in .page)
3. ognl has to do extra processing to subclass and generate the
implementation
the only possible benefit i can see (the book states) "Properties can be
defined as persistent, meaning that they are stored on the server
between request cycles."
does anyone have an argument on this?
---------------------------------------------------------------------
To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
Re: abstract Pages and property-specifications
Posted by Jamie Orchard-Hays <ja...@dang.com>.
Yes. Since much of the work with those properties is done in the .page/.jwc
files, you can define them and see them in that file rather than having to
bounce back to the java file. Furthermore, if you want to make dynamic
changes while your app is running, this can be done with the current model.
If you leave it to always making changes in the java file, then you must
recompile between changes, rather than just refreshing the page in the
browser. I'll take a browser refresh over a recompile/redeploy/refresh cycle
any day.
Jamie
----- Original Message -----
From: "Paul Stanton" <pa...@gunnsoft.com.au>
To: "Tapestry users" <ta...@jakarta.apache.org>
Sent: Monday, May 17, 2004 7:23 PM
Subject: abstract Pages and property-specifications
> i fail to see the benefit of using implied instance variables in the
> form of abstract accessor methods and <property-specification> elements
> in the .page file. It would seem more efficient to implement the
> instance variables and getter/setter methods and detach method.
>
> 1. the number of files edited is greater (.java & .page)
> 2. the number of characters typed is greater (xml element in .page)
> 3. ognl has to do extra processing to subclass and generate the
> implementation
>
> the only possible benefit i can see (the book states) "Properties can be
> defined as persistent, meaning that they are stored on the server
> between request cycles."
>
> does anyone have an argument on this?
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
Re: abstract Pages and property-specifications
Posted by Glen Stampoultzis <gs...@iinet.net.au>.
The reason they're a good thing is that the initialization and teardown is
handled automatically for you. This stuff is very easy to miss if you're
doing it manually.
I'm not sure about point 2. It seems to me it's less typing.
-- Glen
At 09:23 AM 18/05/2004, you wrote:
>i fail to see the benefit of using implied instance variables in the form
>of abstract accessor methods and <property-specification> elements in the
>.page file. It would seem more efficient to implement the instance
>variables and getter/setter methods and detach method.
>
>1. the number of files edited is greater (.java & .page)
>2. the number of characters typed is greater (xml element in .page)
>3. ognl has to do extra processing to subclass and generate the implementation
>
>the only possible benefit i can see (the book states) "Properties can be
>defined as persistent, meaning that they are stored on the server between
>request cycles."
>
>does anyone have an argument on this?
>
>---------------------------------------------------------------------
>To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
>For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
>
Glen Stampoultzis
gstamp@iinet.net.au
http://members.iinet.net.au/~gstamp/glen/