Re: Cached Page State

In Cayenne there is no "unit of work" (which is a separation of read and 
write contexts). The closest concept is DataContext (ala EOF 
EOEditingContext). 

Anyway, do you have different web sessions (and DataContexts) in steps 4 & 
5? If so, there maybe some caching issues. You can post the details on 
cayenne-user list. Myself or other folks will most definitely try to help. 

Also, you may know this already, but just in case. 

1. To do logging of individual selects, do something like:
myQuery.setLoggingLevel(Level.WARN) 

2. To do logging of individual commits, do
myDataContext.commitCHanges(Level.WARN) 

3. To lower the logging threshold  of *all* database operations, configure 
log4j for "org.objectstyle.cayenne.access.QueryLogger" with level lower then 
INFO. For instance in the code you can force using your custom logging file 
by calling static method: 

Configuration.configCommonLogging(urlOfLog4JConfig) 

Cheers 

Andrus 


Geoff Longman writes: 

> Usually when I need to debug these things (the closest example I have is
> when using TOPLink) is
> to watch the sql log and monitor the backend interaction. I don't know if
> you are TOPLink savy but we had a similar case where
> we thought that changes were not being reflected but it turned out that the
> two 'views' were coming off separate UnitOfWork instances (one was out of
> sync of course). 
> 
>  
> 
> ----- Original Message -----
> From: "Neil Clayton" <[email protected]>
> To: "Geoff Longman" <[email protected]>;
> <[email protected]>
> Sent: Tuesday, November 12, 2002 2:41 AM
> Subject: Re: [Tapestry-developer] Cached Page State 
> 
> 
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1 
> 
> Sorry, I'll be more specific: 
> 
> 1) I go to a page to adminster data - 'recent change' data. Each element
> representing some chunck of work.
> 2) T renders the page, using data in a collection, which came from Cayenne
> 3) I click on some element to remove it
> 4) the page is resubmitted, and the element removed
> 5) If I then go to the user based recent changes page, where all of the data
> is displayed, the change is not visiable (e.g: the record is still there) -
> even though it's no longer in the DB. 
> 
> Having just typed all of this, I realise that it could be Cayenne, and that
> there is more I can do debugging wise before throwing my hands in the air! 
> 
> I will go debug.
> But at least you know where I'm coming from.
> It seems ... weeeeird. 
> 
> Neil 
> 
> On Monday 11 Nov 2002 9:09 pm, Geoff Longman wrote:
>> What do you mean when you say "I've got some pages backed by Cayenne"?
>> What is being provided from the backend? Where is it getting put? 
>>
>> You can expire pages by calling 
>>
>> public void discardPage(String name) 
>>
>> on the request cycle although there may be side effects. 
>>
>> from the javadoc: 
>>
>> /**
>> * Finds the page and its page recorder, creating the page recorder if
>> necessary.
>> * The page recorder is marked for discard regardless of its current state.
>> *
>> * <p>This may make the application stateful even if the page recorder does
>> * not yet exist.
>> *
>> * <p>The page recorder will be discarded at the end of the current request
>> cycle. 
>>
>> Geoff 
>>
>> ----- Original Message -----
>> From: "Neil Clayton" <[email protected]>
>> To: <[email protected]>
>> Sent: Monday, November 11, 2002 2:59 PM
>> Subject: [Tapestry-developer] Cached Page State 
>>
>> > -----BEGIN PGP SIGNED MESSAGE-----
>> > Hash: SHA1
>> >
>> > What kind of caching does Tapestry perform per page?
>> >
>> > I've got some pages backed by Cayenne, and when I make changes, those 
>>
>> changes 
>>
>> > do not appear. If I switch browsers, or restart the web container they
> do
>> > appear.  Also, they do appear if I set
>> > - -Dnet.sf.tapestry.disable-caching=true.
>> >
>> > Is there a better way to selectively disable caching? Or expire pages?
>> >
>> > I've already got
>> > <meta http-equiv="expires" content="0"/>
>> > in the head section of the HTML (again, to little effect in terms of
> this
>> > particular problem!)
>> >
>> > Am I going mad?
>> > Has computing suddenly got beyond me? [I am beginning to wonder!]
>> >
>> > - --
>> > Regards,
>> > Neil Clayton
>> >
>> > (PS: If you see strange text you don't understand underneath my email, 
>>
>> don't 
>>
>> > worry - it's just my PGP signature)
>> > -----BEGIN PGP SIGNATURE-----
>> > Version: GnuPG v1.0.7 (GNU/Linux)
>> >
>> > iD8DBQE90AwLLXcfQF3yrNoRAjmKAJ0cLhtCS4MHuLJYwnZTPCz0U1S5NgCeNPQi
>> > n9/j+vGHSVnDWElT3lwKHHU=
>> > =yS/w
>> > -----END PGP SIGNATURE-----
>> >
>> >
>> >
>> > -------------------------------------------------------
>> > This sf.net email is sponsored by:ThinkGeek
>> > Welcome to geek heaven.
>> > http://thinkgeek.com/sf
>> > _______________________________________________
>> > Tapestry-developer mailing list
>> > [email protected]
>> > https://lists.sourceforge.net/lists/listinfo/tapestry-developer
> 
> - --
> Regards,
> Neil Clayton 
> 
> (PS: If you see strange text you don't understand underneath my email, don't
> worry - it's just my PGP signature)
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.0.7 (GNU/Linux) 
> 
> iD8DBQE90LCgLXcfQF3yrNoRAlB6AJ0XajQYqkAOo6gfdAEQnbNJz2gMzQCcDzWL
> 8h8tsGQ6jhNJ3bOIiCOUMJk=
> =YtDV
> -----END PGP SIGNATURE----- 
> 
>  
> 
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Tapestry-developer mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/tapestry-developer 
> 
>  
> 
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Tapestry-developer mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/tapestry-developer
 


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf