click-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "James P Brown (JIRA)" <j...@apache.org>
Subject [JIRA] Commented: (CLK-497) FieldSet isDisabled and isReadonly methods broken
Date Mon, 02 Mar 2009 21:17:55 GMT

    [ http://issues.apache.org/click/browse/CLK-497?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=11717#action_11717
] 

James P Brown commented on CLK-497:
-----------------------------------

Thank you for the speedy reply.

I actually like that the behaviour of the parent to child state is managed through the isDisabled/isReadonly
methods, because it keeps the component state and behaviours somewhat independent of each
other. With the approach you suggest above, I think you may get into trouble because the container
would only get one chance to "push" to its children: at the time set is invoked. However,
if child components are added subsequent to the state change, then they would miss the event.
As well, if a child component changes its state after the event, then it would be out of sync
with the event. The virtue of your reliance on the isDisabled invocation is that it neatly
side steps the reliance on state change events, and puts everything into the runtime state
evaluations. Otherwise, you would need a kind of "failsafe" set invocation in the lifecycle
that forces all components to be in the correct state, which I think isn't the best way to
go.

Nonetheless, I think the framework is very good, and my use cases shouldn't impact your design
decisions. I should also point out that you did in fact resolve my original issue, and that
I am abusing your good nature by expanding the scope of my original defect request. I think
the onus is on me to think this out a bit more, and make a better effort to understand the
internals of the code.

Thank you,

James


> FieldSet isDisabled and isReadonly methods broken
> -------------------------------------------------
>
>                 Key: CLK-497
>                 URL: http://issues.apache.org/click/browse/CLK-497
>             Project: Click
>          Issue Type: Bug
>          Components: core
>    Affects Versions: 2.0.1
>         Environment: Windows XP, Java 6
>            Reporter: James P Brown
>         Attachments: FieldSet_isDisabled_isReadonly_bug.zip
>
>
> The FieldSet is supposed to force its child components to be disabled/readonly when it
is set to disabled/readonly. I did not observe this when I attempted to create a FieldSet
with a child component.
> I believe the code to support this is not working as anticipated. The Field class has
modified methods for isDisabled/isReadonly that specifically check if the parent component
(i.e. container) is an instanceof FieldSet (or Form, which is working AFAIK). The problem
is that the design of FieldSet relies on an instance of its private inner class FieldSet.InnerContainerField
for managing those child elements. When I step through the code in debug mode, the class instance
is of this inner class type (InnerContainerField) not FieldSet. Since InnerContainerField
is not a type of FieldSet, the subsequent logic is ignored.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message