Mailing-List: contact dev-help@myfaces.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "MyFaces Development" <dev@myfaces.apache.org>
Received-SPF: neutral (asf.osuosl.org: local policy)
Message-ID: <43850813.6070707@obsidium.com>
Date: Thu, 24 Nov 2005 13:23:47 +1300
From: Simon Kitching <skitching@obsidium.com>
User-Agent: Thunderbird 1.4 (X11/20050908)
MIME-Version: 1.0
To: MyFaces Development <dev@myfaces.apache.org>
Subject: Release method in UIComponentTag (was  Reconnecting UIComponentTag
 instances ...)
References: <4383A60B.3020205@obsidium.com>
	 <cb51ede40511221532m7de8f724j9e6f3a43e34ce6@mail.gmail.com>
	 <4383AF61.5000607@obsidium.com>
	 <cb51ede40511231325o41b0fd19q192f15f4c17b5afe@mail.gmail.com>
 <6dac79b90511231425u3f80a8eah7ea31079853d07fe@mail.gmail.com>
In-Reply-To: <6dac79b90511231425u3f80a8eah7ea31079853d07fe@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

I see that originally _id etc were reset in internalRelease, and they 
were moved to the release method by this commit:

------------------------------------------------------------------------
r166747 | manolito | 2004-04-27 00:01:39 +1200 (Tue, 27 Apr 2004) | 2 lines

more reluctant releasing of members for Resin compatibility


So it looks like this is a workaround for a Resin bug. I dont quite 
understand that; I can't see what bug Resin might have that would be 
fixed by this but wouldn't break many other things. Is "manolito" around 
to explain further?

Regards,

Simon

Adam Winer wrote:
> Sounds like it is a bug, though that's a very quick impression
> without serious thought.
> 
> Re: what happens if the number of tags change from one request
> to the next:  this is why JSF tags inside of c:if are required to have
> an explicitly set ID (think this is stated explicitly in the JSF spec).
> 
> This whole mess is why I proposed JspIdConsumer for JSP 2.1,
> which lets you do this reconnecting in a robust manner.
> 
> Cheers,
> Adam Winer
> 
> On 11/23/05, Travis Reeder <treeder@gmail.com> wrote:
>> Does anyone have any comment on whether this is a bug or not:
>>
>>> As you can see UIComponentTag.doEndTag calls internalRelease(). But
>>> strangely, members _id, _rendered and _binding are cleared only in
>>> release(), not internalRelease(). Maybe that's a bug?
>>>
>