[prev in list] [next in list] [prev in thread] [next in thread] 

List:       myfaces-user
Subject:    Re: Maximum number of tokens in pageFlowScope
From:       "Scott O'Bryan" <darkarena () gmail ! com>
Date:       2008-05-27 20:46:17
Message-ID: 483C7319.1050204 () gmail ! com
[Download RAW message or body]

No, I totally agree.  :)  I was just commenting because this thing 
hasn't worked for YEARS...  Thanks for finding it..

Dirk Krummacker wrote:
> Scott,
>
> Thanks for the confirmation. I have already created an issue earlier today:
>
> https://issues.apache.org/jira/browse/TRINIDAD-1095
>
> This setting might not be used too often, but we need it, so please
> don't delete ;-)
>
> Dirk
>
>
> On Tue, May 27, 2008 at 9:34 PM, Scott O'Bryan <darkarena@gmail.com> wrote:
>   
>> The source code is wrong so please log a JIRA ticket.  Process Scope was the
>> name of page flow scope before oracle donated the code to apache.  Obviously
>> somebody forgot to change the source to match the new name used in the
>> documentation.  :)  Shows how often that setting is used.
>>
>> Scott
>>
>> Dirk Krummacker wrote:
>>     
>>> Hello all,
>>>
>>> We are using pageFlowScope to "carry over" data from request to
>>> request. It quite nicely fits our requirements (better than
>>> t:saveState which we used before). We are using the token-based
>>> approach where the actual data is stored in the session and only a
>>> token ("_afPfm") is added as a request parameter. In order to allow
>>> for some degree of "back-button-history", the mechanics of the
>>> pageFlowScope store not only one instance of the data in the session,
>>> but the data for the last 15 requests (default value).
>>>
>>> Because of a certain usage pattern we expect from out users
>>> (right-click and external URLs) I wanted to increase this value to 30.
>>> Memory is not an issue, we expect only a fairly low number of users
>>> and the data is not so huge anyway. For this, I consulted the
>>> documentation in the developer's guide on
>>>
>>> http://myfaces.apache.org/trinidad/devguide/configuration.html#trinidad-config.xml
>>> and found the parameter "page-flow-scope-lifetime". This seemed to be
>>> exactly what I was searching for, and so I added it to my
>>> trinidad-config.xml.
>>>
>>> But that didn't work. Using a debugger and the Trinidad source, I
>>> found out that the code was actually checking for the parameter of the
>>> name "process-scope-lifetime". With that name, everything worked fine
>>> for me.
>>>
>>> Now my question is: Is this possibly a bug in the documentation, or is
>>> the source code wrong? The documented name "page-flow-scope-lifetime"
>>> makes more sense to me, plus I have never heard of the term
>>> "process-scope" in the context of the pageFlowScope. Maybe a remnant
>>> from ADF times?
>>>
>>> Thanks in advance,
>>>
>>> Dirk
>>>
>>>       
>>     

[prev in list] [next in list] [prev in thread] [next in thread]