dev@javaserverfaces.java.net

Re: Issue 484 - Is f:subview still needed?

From: Ryan Lubke <Ryan.Lubke_at_Sun.COM>
Date: Fri, 02 Feb 2007 13:33:53 -0800

Ryan Lubke wrote:
> Adam Winer wrote:
>> Ryan Lubke said the following on 2/2/07 11:54 AM PT:
>>> Adam Winer wrote:
>>>> Is this strictly an issue for tag files, or does it
>>>> have repercussions for jsp:include?
>>> jsp:include is handled. We detect it by a servlet required request
>>> attribute that will
>>> be present, so it's strictly tag files.
>>>>
>>>> I would have thought that a tag file would have a separate
>>>> JspContext instance from the page in which it is used, but is
>>>> that not necessarily so?
>>> Right, it does. I kicked some ideas around with detecting the
>>> the fact that a Tag file will have a clean page scope, but they didn't
>>> quite work out.
>>>
>>> I guess one could just track each PageContext and for each new one we
>>> encounter mundge the IDs appropriately....
>>
>> If we did this, would it obviate the need for
>> detecting includes via the request attribute?
> It probably would, but I personally prefer continuing to detect it.
> The IDs are generated so that it's very easy to tell if they were an
> include or not - but that's me.
>>
>> But I'd be extremely concerned that a JSP environment
>> might reuse PageContext instances. Given the JSP EG's
>> (at least past) adoration of object reuse, they might
>> have gone down this path.
> True.
I think I've got a fix - at least the Tobago demo and cactus tests are
running.

I'll post a change bundle shortly.
>>
>>
>> -- Adam
>>
>>>
>>>
>>>>
>>>>
>>>> -- Adam
>>>>
>>>>
>>>> Ryan Lubke said the following on 2/2/07 11:32 AM PT:
>>>>> Hey Folks,
>>>>>
>>>>> I'm out of ideas on this one. So far, I've tried to come up with
>>>>> a portable way to
>>>>> detect that JSF is executing within the context of a JSP Tag file
>>>>> in order to handle
>>>>> the JSP ids, but there doesn't seem to be any way at this point.
>>>>>
>>>>> My last entry in the bug led me to believe a potential issue with
>>>>> the JSP implementation,
>>>>> but after bringing it up to the JSP spec leads, they state it's a
>>>>> case of bad wording
>>>>> and Jasper is doing the right thing.
>>>>>
>>>>> So, hopefully someone out here has an idea or two on how we might
>>>>> be able to
>>>>> tackle this one, otherwise, the only solution I see is to use
>>>>> f:subview.
>>>>>
>>>>> Please review
>>>>> https://javaserverfaces.dev.java.net/issues/show_bug.cgi?id=484
>>>>>
>>>>> Thanks,
>>>>>
>>>>> -rl
>>>>>
>>>>> ---------------------------------------------------------------------
>>>>> To unsubscribe, e-mail: dev-unsubscribe_at_javaserverfaces.dev.java.net
>>>>> For additional commands, e-mail:
>>>>> dev-help_at_javaserverfaces.dev.java.net
>>>>>
>>>>
>>>> ---------------------------------------------------------------------
>>>> To unsubscribe, e-mail: dev-unsubscribe_at_javaserverfaces.dev.java.net
>>>> For additional commands, e-mail: dev-help_at_javaserverfaces.dev.java.net
>>>>
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: dev-unsubscribe_at_javaserverfaces.dev.java.net
>>> For additional commands, e-mail: dev-help_at_javaserverfaces.dev.java.net
>>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe_at_javaserverfaces.dev.java.net
>> For additional commands, e-mail: dev-help_at_javaserverfaces.dev.java.net
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe_at_javaserverfaces.dev.java.net
> For additional commands, e-mail: dev-help_at_javaserverfaces.dev.java.net
>