Scott Anderson wrote:
Thanks Ate! I have tested the updated FilterPortlet and your fixes have
resolved my issues.
Great!
I'll close the JIRA issues.
Scott
On 9/16/06, Ate Douma <[EMAIL PROTECTED]> wrote:
I've created the following JIRA issues for this:
http://issues.apache.org/jira/browse/PB-46
http://issues.apache.org/jira/browse/PB-47
http://issues.apache.org/jira/browse/PB-48
Fixes already committed, ready for review :)
Regards,
Ate
Ate Douma wrote:
> Shinsuke SUGAYA wrote:
>> Scott Anderson wrote:
>>> The portletConfig instance variable is used but never set. In the
>>> methods
>>> getInitParameter, getInitParameterNames, getPortletContext,
>>> getPortletName,
>>> & getResourceBundle you will get a NullPointerException if the
>>> Portlet being
>>> managed (portlet instance variable) is not an instance of
PortletConfig.
>>
>> I do not see that issue.
> I just looked at the code and he's right. When the portlet managed is
> *not* an instance of PortletConfig, the FilterPortlet will use its own
> provided PortletConfig for resolving the request.
> But FilterPortlet doesn't save its PortletConfig in its init method,
> hence a NPE.
>
> After reviewing Scott's request to be able to use the FilterPortlet
with
> ServletContextProvider (which requires the portlet, in this case
> FilterPortlet) to extend GenericPortlet, I think it can be easily
> "fixed", without introducing side-effects or new dependencies.
> As the FilterPortlet currently only implements Portlet and
> PortletConfig, which GenericPortlet does as well, simply extending
> GenericPortlet and calling super.init(PortletConfig) in its
> init(PortletConfig) method will resolve both issues.
>
> Additionally, I see the FilterPortlet contains another bug: in its
> destroy() method it doesn't call the managed portlet destroy() method.
>
> I will create a new JIRA issue for these issues, commit my proposed
> fixes to FilterPortlet and define portals.bridges.jsf.version=1.0.1-dev
> in project.properties.
>
> Then Scott can test if this indeed resolves his issues, and maybe you
> can validate this doesn't cause any side-effects for your usecases.
>
> Regards,
>
> Ate
>
>> Please provide(or file a bug) more information with a test case.
>>
>> Thanks,
>> shinsuke
>>
>>>
>>> Scott
>>>
>>
>> --------------------------------------
>> [10th Anniversary] special auction campaign now!
>> http://pr.mail.yahoo.co.jp/auction/
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: [EMAIL PROTECTED]
>> For additional commands, e-mail: [EMAIL PROTECTED]
>>
>>
>>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
>
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
