If wicket policy prohibits isVisible() to depend on parent page,
you might as well make it final.
This is acceptable if you provide some other hook that may depend on
parent page.
Johan Compagner wrote:
>
> Moreover, this is an inconsistent behavior. System should have the
> same
> beh
Moreover, this is an inconsistent behavior. System should have the same
behavior regardless whether log is enabled or not.
thats true, i was just saying your isVisible() shouldn't absolutely depend
on a parent page..
It could be used in none logging code also. The parent hierarchy doesn't
hav
This breaks the concept of Model being evaluated when component is
rendered and not beforehand.
Moreover, this is an inconsistent behavior. System should have the same
behavior regardless whether log is enabled or not.
Johan Compagner wrote:
> seeing if a component is visible or not in the toStri
seeing if a component is visible or not in the toString() impl is pretty
nice to have
I think you should make your isVisible() a bit more safe.
What if you debug log your self a bit more about this ? and you use it in a
listview or something
then it can also be removed from the page.
Or if you us
Issued:
https://issues.apache.org/jira/browse/WICKET-622
Thanks,
Nili
Eelco Hillenius wrote:
>> 1. Why isn't toString() of Component a safe method ?
>>
>
> It should be, and later versions tweaked this. However, it's still
> defaulting to the extended debug info, which is maybe not the best
>
> 1. Why isn't toString() of Component a safe method ?
It should be, and later versions tweaked this. However, it's still
defaulting to the extended debug info, which is maybe not the best
idea. Can you open a JIRA issue for this please?
Thanks,
Eelco
---
I come across a very puzzling behavior in wicket-1.2.2.
I add a component to its container when the page is initialized, BEFORE
it is rendered and before page model finished to initialize.
if log4j is configured to enable wicket debug, the container toString()
is called.
This method invoked isVisib
