OTM M-N Mapping Issue

2004-02-29 Thread David Le Strat
Hello there,

I am using OTM with M-N mapping. I am having an issue
with OTM, where OJB violate foreign key constraints.

I have the following mapping:

PREFS_NODE -- PREFS_NODE_PROPERTY_KEY --
PREFS_PROPERTY_KEY

With OTM, OJB tries to delete PREFS_PROPERTY_KEY
before removing PREFS_NODE_PROPERTY_KEY therefore
violating foreign key constraint.  Has anybody else
experienced the same behavior.

Regards,

David Le Strat.

__
Do you Yahoo!?
Get better spam protection with Yahoo! Mail.
http://antispam.yahoo.com/tools

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Extent

2004-02-29 Thread Dirk Manske (Service Respond)
Hi,

I am using PB-API and DB2 7 for a few weeks now and have a question
regarding the ojb extent concept. 

I have a BaseClass which holds common attributes (timestamp, versionnr) for
model objects. Each model class has to extent BaseClass. Database tables
hold the common attributes and specific attributes for the appropriate model
class. 

In the repository-user.xml I declared the common attributes separately for
each model class-descriptor. My question is: Did I miss the ojb-extent
concept totally? Could I just declare a model class as an extent-class of
BaseClass and delete the common attributes from each (model)class-descriptor
because of the extent, ojb would handle these attributes automatically for
each extended class?

thx,
Dirk


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Re: Non decomposed problem

2004-02-29 Thread olivier.nouguier
Hi all

Some news on my problem ... In fact this behavior ( deletion of
roles
when removing the cross reference ) only occurs when I'm NOT using the
ObjectCacheDefaultImpl caching strategie.

It's strange to me ?


Le sam 28/02/2004 à 02:32, olivier.nouguier a écrit :
 Hi all,
   I'm really lost with auto-delete feature...
 
   The model:
 
0,n   0,n
 Actors  Roles
 
   The RDBMS schema:
 
 actors ( id, ... )
 roles ( id, ... )
 actors_roles ( actor_id, role_id )
 
   The problem:
 
 With the attached repository.xml, when I remove a role from an actor,
 the role is also deleted from the table. But when I remove the actor,
 the role is kept ?
 
 All attribute are ( auto-update, auto-delete ) set to false.
 
 ojb-RC5
 
 //
   
   Transaction tx = odmg.newTransaction();
   tx.begin();
   Actor actor = retrieveActor(id);
   tx.lock(actor, Transaction.WRITE);
   actor.removeRole(0); ( delegate actor.roles.remove(0) on a vector )
   
   tx.commit();
 //
 
 
 
 
 
 __
 -
 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]



Proxy objects for nonexisting references

2004-02-29 Thread Juergen Ebert
Hi all,

I've just changed my application to using dynamic proxies and I noticed
some unexpected behaviour: even if a foreign key attribute is 0, OJB
instantiates a proxy object (with 0 as primary key value). So all the
existing tests that expect a null value if this reference does not exist
now fail. To do the same test I now have to check the foreign key field
for 0, but that field should better not appear in the interface. Is this
by design or is there a way to change this?

Juergen


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Re: OTM M-N Mapping Issue

2004-02-29 Thread Oleg Nitz
Hi David,

If you use otm-dependent relations, then check the current CVS HEAD - I've 
just committed the fix. If not, then give more detailed description of your 
case.

Regards,
 Oleg

On Sunday 29 February 2004 08:09, David Le Strat wrote:
 Hello there,

 I am using OTM with M-N mapping. I am having an issue
 with OTM, where OJB violate foreign key constraints.

 I have the following mapping:

 PREFS_NODE -- PREFS_NODE_PROPERTY_KEY --
 PREFS_PROPERTY_KEY

 With OTM, OJB tries to delete PREFS_PROPERTY_KEY
 before removing PREFS_NODE_PROPERTY_KEY therefore
 violating foreign key constraint.  Has anybody else
 experienced the same behavior.

 Regards,

 David Le Strat.

 __
 Do you Yahoo!?
 Get better spam protection with Yahoo! Mail.
 http://antispam.yahoo.com/tools

 -
 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]



Re: OTM M-N Mapping Issue

2004-02-29 Thread David Le Strat
Thanks Oleg, just saw your commit ;).  I use OTM. I
will try the latest head.

Regards,

David.
--- Oleg Nitz [EMAIL PROTECTED] wrote:
 Hi David,
 
 If you use otm-dependent relations, then check the
 current CVS HEAD - I've 
 just committed the fix. If not, then give more
 detailed description of your 
 case.
 
 Regards,
  Oleg
 
 On Sunday 29 February 2004 08:09, David Le Strat
 wrote:
  Hello there,
 
  I am using OTM with M-N mapping. I am having an
 issue
  with OTM, where OJB violate foreign key
 constraints.
 
  I have the following mapping:
 
  PREFS_NODE -- PREFS_NODE_PROPERTY_KEY --
  PREFS_PROPERTY_KEY
 
  With OTM, OJB tries to delete PREFS_PROPERTY_KEY
  before removing PREFS_NODE_PROPERTY_KEY therefore
  violating foreign key constraint.  Has anybody
 else
  experienced the same behavior.
 
  Regards,
 
  David Le Strat.
 
  __
  Do you Yahoo!?
  Get better spam protection with Yahoo! Mail.
  http://antispam.yahoo.com/tools
 
 

-
  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]
 


__
Do you Yahoo!?
Get better spam protection with Yahoo! Mail.
http://antispam.yahoo.com/tools

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Re: Extent

2004-02-29 Thread Peter Wieland
Hi Dirk,

I'm not sure whether I understood your problem. Maybe I can help anyway...

First, I'm not sure whether you really need to declare the extents for all
your classes. Wouldn't it be sufficient to simply have all model classes
having the same superclass BaseClass. If this is the case, simply model
write a class-descriptor for each model class without any extent
declarations. The BaseClass doesn't even need an entry in the repository
file.

Of course, if you need to do queries using the BaseClass, this approach
won't work. You would then need to declare the BaseClass in the repository
with all it's extents. This would probably look like

class-descriptor class=foo.bar.BaseClass
  extent-class class-ref=foo.bar.ModelClass1/
  extent-class class-ref=foo.bar.ModelClass2/
  ...
  extent-class class-ref=foo.bar.ModelClassN/
  field-descriptor autoincrement=true primarykey=true column=id
jdbc-type=VARCHAR name=id/
  field-descriptor column=version jdbc-type=VARCHAR name=version/
/class-descriptor

In the latter case, I think you would have to redeclare the id and version
fields in each extent class together with the specific attributes for each
extent class:

class-descriptor class=foo.bar.ModelClass1
  field-descriptor autoincrement=true primarykey=true column=id
jdbc-type=VARCHAR name=id/
  field-descriptor column=version jdbc-type=VARCHAR name=version/
  field-descriptor column=fieldAFromClass1 jdbc-type=VARCHAR
name=fieldAFromClass1/
  field-descriptor column=fieldBFromClass1 jdbc-type=VARCHAR
name=fieldBFromClass1/
/class-descriptor

I'm not quite sure, whether it is really necessary to redeclare the common
fields in each extent class, but I'm sure that it works this way as we are
using this approach. The column names for the extent classes do not need to
be the same for the same fields in different extent classes.

Using the extent-approach, you will have one separate table for each model
class, containing all the needed columns, included the inherited ones.

Hope that helps,

Peter


 Hi,

 I am using PB-API and DB2 7 for a few weeks now and have a question
 regarding the ojb extent concept.

 I have a BaseClass which holds common attributes (timestamp, versionnr)
 for
 model objects. Each model class has to extent BaseClass. Database tables
 hold the common attributes and specific attributes for the appropriate
 model
 class.

 In the repository-user.xml I declared the common attributes separately for
 each model class-descriptor. My question is: Did I miss the ojb-extent
 concept totally? Could I just declare a model class as an extent-class of
 BaseClass and delete the common attributes from each
 (model)class-descriptor
 because of the extent, ojb would handle these attributes automatically for
 each extended class?

 thx,
 Dirk


 -
 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]



AW: Extent

2004-02-29 Thread Dirk Manske (Service Respond)
Hi Peter,

my first repository-user.xml does not contain the extent concept at all. I
simply did not set up a class-descriptor for BaseClass - just as you
described.

The extent concept came up to my mind when I thought that XDoclet would be a
great time saver to set up the repository. For each attribute in the model
class you declare an ojb.field, ojb.reference etc. Since the common
attributes are in BaseClass I wondered how to set them up via XDoclet for
each model class. That is why I guess U have to use the extent concept of
ojb.   

Any idea?

Dirk

-Ursprüngliche Nachricht-
Von: Peter Wieland [mailto:[EMAIL PROTECTED] 
Gesendet: Sonntag, 29. Februar 2004 16:47
An: [EMAIL PROTECTED]
Betreff: Re: Extent

Hi Dirk,

I'm not sure whether I understood your problem. Maybe I can help anyway...

First, I'm not sure whether you really need to declare the extents for all
your classes. Wouldn't it be sufficient to simply have all model classes
having the same superclass BaseClass. If this is the case, simply model
write a class-descriptor for each model class without any extent
declarations. The BaseClass doesn't even need an entry in the repository
file.

Of course, if you need to do queries using the BaseClass, this approach
won't work. You would then need to declare the BaseClass in the repository
with all it's extents. This would probably look like

class-descriptor class=foo.bar.BaseClass
  extent-class class-ref=foo.bar.ModelClass1/
  extent-class class-ref=foo.bar.ModelClass2/
  ...
  extent-class class-ref=foo.bar.ModelClassN/
  field-descriptor autoincrement=true primarykey=true column=id
jdbc-type=VARCHAR name=id/
  field-descriptor column=version jdbc-type=VARCHAR name=version/
/class-descriptor

In the latter case, I think you would have to redeclare the id and version
fields in each extent class together with the specific attributes for each
extent class:

class-descriptor class=foo.bar.ModelClass1
  field-descriptor autoincrement=true primarykey=true column=id
jdbc-type=VARCHAR name=id/
  field-descriptor column=version jdbc-type=VARCHAR name=version/
  field-descriptor column=fieldAFromClass1 jdbc-type=VARCHAR
name=fieldAFromClass1/
  field-descriptor column=fieldBFromClass1 jdbc-type=VARCHAR
name=fieldBFromClass1/
/class-descriptor

I'm not quite sure, whether it is really necessary to redeclare the common
fields in each extent class, but I'm sure that it works this way as we are
using this approach. The column names for the extent classes do not need to
be the same for the same fields in different extent classes.

Using the extent-approach, you will have one separate table for each model
class, containing all the needed columns, included the inherited ones.

Hope that helps,

Peter


 Hi,

 I am using PB-API and DB2 7 for a few weeks now and have a question 
 regarding the ojb extent concept.

 I have a BaseClass which holds common attributes (timestamp, 
 versionnr) for model objects. Each model class has to extent 
 BaseClass. Database tables hold the common attributes and specific 
 attributes for the appropriate model class.

 In the repository-user.xml I declared the common attributes separately 
 for each model class-descriptor. My question is: Did I miss the 
 ojb-extent concept totally? Could I just declare a model class as an 
 extent-class of BaseClass and delete the common attributes from each 
 (model)class-descriptor because of the extent, ojb would handle these 
 attributes automatically for each extended class?

 thx,
 Dirk


 -
 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]



Re: success deploying ojb in jboss

2004-02-29 Thread Armin Waibel
Hi Michael and Andrew,

I will integrate this approach (howto build an .ear file) in small 
section of the deployment doc. Any suggestions?

regards,
Armin
Clute, Andrew wrote:

Congrats on getting it to work!

Sorry, I forgot to mention the need for your lib's to be in the CLASSPATH in the Manifest file.

Isn't it nice to be able to redeploy also!

-Andrew

-Original Message-
From: Michael Mogley [mailto:[EMAIL PROTECTED]
Sent: Sat 2/28/2004 10:34 PM
To: OJB Users List
Subject: success deploying ojb in jboss
 
Thanks Andrew and Armin.  I followed Andrew's advice and packaged everything
in one ear.  Good news is I am also able to redeploy without having to
restart the app-server (no OJB redeployment issues so far).

Here's how I did it:

/ejb.jar/
...EJBs
...DAOs
...META-INF/
..ejb-jar.xml
..jboss.xml
..MANIFEST.MF [Class-Path: /lib/ojb.jarCRLF]   --- VERY IMPORTANT
/lib/
...ojb.jar
/web-app.war/
...JSP
...WEB-INF/
..web.xml [ejb-refs to EJBs]
/META-INF/
...application.xml
/OJB.properties
/repository.xml
/repository_database.xml
/repository_user.xml
It would have been great to have a template like this before I started.
Hope this makes someone's life easier.
Michael



-
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]


Re: AW: Extent

2004-02-29 Thread Peter Wieland
Hi Dirk,

sorry, no idea how to use this with XDoclet. I pass on the question to the
community.

We use AXgen to generate everything based on an UML metamodel...

Peter

 Hi Peter,

 my first repository-user.xml does not contain the extent concept at all. I
 simply did not set up a class-descriptor for BaseClass - just as you
 described.

 The extent concept came up to my mind when I thought that XDoclet would be
 a
 great time saver to set up the repository. For each attribute in the model
 class you declare an ojb.field, ojb.reference etc. Since the common
 attributes are in BaseClass I wondered how to set them up via XDoclet for
 each model class. That is why I guess U have to use the extent concept of
 ojb.

 Any idea?

 Dirk

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Re: Changing AutoRetrieve on the fly

2004-02-29 Thread Robert S. Sfeir
Armin Waibel wrote:

Hi Robert,

Robert S. Sfeir wrote:

I have auto-retreive set to false in all my descriptors currently.  
Reason I do that is because sometimes I am not interested in getting 
the related objects, I just want to do a select from the table and 
just get the data back.

However, there are times when I do want to fetch the related 
objects.  So with that, I tried the following code below, but it 
seems that it's incorrect since my bean still returns a null for the 
related bean.

final Query query = QueryFactory.newQuery( Category.class, crit );
 broker.beginTransaction();
 final ObjectReferenceDescriptor ord = new 
ObjectReferenceDescriptor( broker.getClassDescriptor( Project.class ) );
 ord.setCascadeRetrieve( true );


OJB couldn't watch your made changes, 'ord' instance was not integrated.
Why you don't use
PB.retrieveAllReferences
PB.retrieveReference
for these objects you want to know the reference objects?
That's the first thing I tried, but it returns null for my objects, I 
must not be using the code Right!

Project p = new ProjectBean(); //TODO: Replace with Factory call.
broker.retrieveAllReferences( p );
final List results = ( List ) broker.getCollectionByQuery( query );
Any help is appreciated.
R
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


Re: Changing AutoRetrieve on the fly

2004-02-29 Thread Armin Waibel
Hi,

Robert S. Sfeir wrote:

...
That's the first thing I tried, but it returns null for my objects, I 
must not be using the code Right!

Project p = new ProjectBean(); //TODO: Replace with Factory call.
broker.retrieveAllReferences( p );
final List results = ( List ) broker.getCollectionByQuery( query );
I don't know your ProjectBean class, but when you create a new object 
you can't find any reference object - or I'm wrong?

regards,
Armin
Any help is appreciated.
R
-
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]


Re: Changing AutoRetrieve on the fly

2004-02-29 Thread Armin Waibel
That's the first thing I tried, but it returns null for my objects, I 
must not be using the code Right!

Project p = new ProjectBean(); //TODO: Replace with Factory call.
broker.retrieveAllReferences( p );
final List results = ( List ) broker.getCollectionByQuery( query );
I don't know your ProjectBean class, but when you create a new object 
you can't find any reference object - or I'm wrong?

ah, I think you do the wrong turn and I misunderstood your example.

Do after retrieve of the Category objects a

final List results = ( List ) broker.getCollectionByQuery( query );
aCategoryObject = ...get from list
broker.retrieveAllReferences( aCategoryObject );
call to get all reference objects for all Category instances you want to 
assign with the appropriate references.

Armin

regards,
Armin
Any help is appreciated.
R
-
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]


Re: Changing AutoRetrieve on the fly

2004-02-29 Thread Robert S. Sfeir
Armin Waibel wrote:

Hi,

Robert S. Sfeir wrote:

...

That's the first thing I tried, but it returns null for my objects, I 
must not be using the code Right!

Project p = new ProjectBean(); //TODO: Replace with Factory call.
broker.retrieveAllReferences( p );
final List results = ( List ) broker.getCollectionByQuery( query );
I don't know your ProjectBean class, but when you create a new object 
you can't find any reference object - or I'm wrong?
I should explain.

I am doing a select on Category.  a Category bean contains a get and set 
for a ProjectBean.  a Category only relates to one Project.  If I do the 
autoretrieve false then the projectBean object in the Category bean is 
null.  If I set it to true, then it correctly contains a ProjectBean 
which the category relates to.

Now sometimes I just want to get a list of categories, and I could care 
less that I get the ProjectBean with it, that's too heavy if I want to 
just display the category id and category name.  Other times I need to 
do the whole thing, like for a category detail view.  In that case I 
want to get all the ProjectBeans along with the category.

Does this makes sense?

So if I understood retrieveAllReferences, it will return all references 
of the object (ProjectBean) which relate to the category based on the 
fact that the Query object is querying Category.class.

R

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


Re: Changing AutoRetrieve on the fly

2004-02-29 Thread Robert S. Sfeir
Armin Waibel wrote:

That's the first thing I tried, but it returns null for my objects, 
I must not be using the code Right!

Project p = new ProjectBean(); //TODO: Replace with Factory call.
broker.retrieveAllReferences( p );
final List results = ( List ) broker.getCollectionByQuery( query );
I don't know your ProjectBean class, but when you create a new object 
you can't find any reference object - or I'm wrong?

ah, I think you do the wrong turn and I misunderstood your example.

Do after retrieve of the Category objects a

final List results = ( List ) broker.getCollectionByQuery( query );
aCategoryObject = ...get from list
broker.retrieveAllReferences( aCategoryObject );
call to get all reference objects for all Category instances you want 
to assign with the appropriate references.
I think I know what you mean, but I'm confused about:

aCategoryObject = ...get from list

aCategoryObject do you mean my CategoryBean?

what do you mean ...get from list?

Are you saying that as I do my for loop to build my view objects, I 
should grab each CategoryBean, and then get the references to it then?

Thanks
R
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


Re: Changing AutoRetrieve on the fly

2004-02-29 Thread Armin Waibel
Hi,

Robert S. Sfeir wrote:

I am doing a select on Category.  a Category bean contains a get and set 
for a ProjectBean.  a Category only relates to one Project.  If I do the 
autoretrieve false then the projectBean object in the Category bean is 
null.  If I set it to true, then it correctly contains a ProjectBean 
which the category relates to.

Now sometimes I just want to get a list of categories, and I could care 
less that I get the ProjectBean with it, that's too heavy if I want to 
just display the category id and category name.  Other times I need to 
do the whole thing, like for a category detail view.  In that case I 
want to get all the ProjectBeans along with the category.

Does this makes sense?
yes of course!

So if I understood retrieveAllReferences, it will return all references 
of the object (ProjectBean) which relate to the category based on the 
fact that the Query object is querying Category.class.

seems my previous mail is lost in space ;-)
Think you do the wrong order of commands. First retrieve the Category 
objects with a query. Then for all Category objects you want to assign 
references, call PB.retrieveAllReferences/retrieveReference.
Or do I misunderstand your post (sorry my bad english)?

regards,
Armin
R

-
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]


Re: Changing AutoRetrieve on the fly

2004-02-29 Thread Armin Waibel
 Are you saying that as I do my for loop to build my view objects, I
 should grab each CategoryBean, and then get the references to it then?
yes, that's how it is work currently.

Maybe we should allow to override the auto-XXX settings on the fly by 
add setAutoXXX methods to query object or something similar in 1.1.
Would be a nice feature and would allow what you expected.

regards,
Armin
Robert S. Sfeir wrote:

Armin Waibel wrote:

That's the first thing I tried, but it returns null for my objects, 
I must not be using the code Right!

Project p = new ProjectBean(); //TODO: Replace with Factory call.
broker.retrieveAllReferences( p );
final List results = ( List ) broker.getCollectionByQuery( query );
I don't know your ProjectBean class, but when you create a new object 
you can't find any reference object - or I'm wrong?

ah, I think you do the wrong turn and I misunderstood your example.

Do after retrieve of the Category objects a

final List results = ( List ) broker.getCollectionByQuery( query );
aCategoryObject = ...get from list
broker.retrieveAllReferences( aCategoryObject );
call to get all reference objects for all Category instances you want 
to assign with the appropriate references.


I think I know what you mean, but I'm confused about:

aCategoryObject = ...get from list

aCategoryObject do you mean my CategoryBean?

what do you mean ...get from list?

Are you saying that as I do my for loop to build my view objects, I 
should grab each CategoryBean, and then get the references to it then?

Thanks
R
-
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]


Re: Changing AutoRetrieve on the fly

2004-02-29 Thread Robert S. Sfeir
Armin Waibel wrote:

 Are you saying that as I do my for loop to build my view objects, I
 should grab each CategoryBean, and then get the references to it then?
yes, that's how it is work currently.

Maybe we should allow to override the auto-XXX settings on the fly by 
add setAutoXXX methods to query object or something similar in 1.1.
Would be a nice feature and would allow what you expected.
Yup. That would be real nice to have it turned on and off.  Think of it 
like AutoCommit.  Sometimes you want to set that to true, and sometimes not.

Ok let me ask a question another way, and perhaps the answer is the same...

What if I turn auto-retrieve to true by default and then turn on some 
setting so that it does do the retreiveAllReferences(), is this the same 
problem?

Thanks
R
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


Re: success deploying ojb in jboss

2004-02-29 Thread Michael Mogley
Hi Armin,

I would have a brief summary of the approach and then just a diagram showing
how all the pieces fit together.  I would list it as a portable alternative
to the sar version.

Michael

- Original Message - 
From: Armin Waibel [EMAIL PROTECTED]
To: OJB Users List [EMAIL PROTECTED]
Sent: Sunday, February 29, 2004 8:37 AM
Subject: Re: success deploying ojb in jboss


 Hi Michael and Andrew,

 I will integrate this approach (howto build an .ear file) in small
 section of the deployment doc. Any suggestions?

 regards,
 Armin

 Clute, Andrew wrote:

  Congrats on getting it to work!
 
  Sorry, I forgot to mention the need for your lib's to be in the
CLASSPATH in the Manifest file.
 
  Isn't it nice to be able to redeploy also!
 
  -Andrew
 
 
  -Original Message-
  From: Michael Mogley [mailto:[EMAIL PROTECTED]
  Sent: Sat 2/28/2004 10:34 PM
  To: OJB Users List
  Subject: success deploying ojb in jboss
 
  Thanks Andrew and Armin.  I followed Andrew's advice and packaged
everything
  in one ear.  Good news is I am also able to redeploy without having to
  restart the app-server (no OJB redeployment issues so far).
 
  Here's how I did it:
 
  /ejb.jar/
  ...EJBs
  ...DAOs
  ...META-INF/
  ..ejb-jar.xml
  ..jboss.xml
  ..MANIFEST.MF [Class-Path: /lib/ojb.jarCRLF]   --- VERY IMPORTANT
  /lib/
  ...ojb.jar
  /web-app.war/
  ...JSP
  ...WEB-INF/
  ..web.xml [ejb-refs to EJBs]
  /META-INF/
  ...application.xml
  /OJB.properties
  /repository.xml
  /repository_database.xml
  /repository_user.xml
 
  It would have been great to have a template like this before I started.
  Hope this makes someone's life easier.
 
  Michael
 
 
 
 
  -
  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]



Re: Non decomposed problem

2004-02-29 Thread olivier.nouguier
Hi all,

In fact I think I found that when cache is NOT the default, the
collection used are the RemovalAwareCollection then the BUG is on the
afterStore callback.


Am I right? Is there another solution?Why did such collection are used
in this case?

Thank 

Have a nice end week end


Le dim 29/02/2004 à 15:23, olivier.nouguier a écrit :
 Hi all
 
 Some news on my problem ... In fact this behavior ( deletion of
 roles
 when removing the cross reference ) only occurs when I'm NOT using the
 ObjectCacheDefaultImpl caching strategie.
 
 It's strange to me ?
 
 
 Le sam 28/02/2004 à 02:32, olivier.nouguier a écrit :
  Hi all,
  I'm really lost with auto-delete feature...
  
  The model:
  
 0,n   0,n
  Actors  Roles
  
  The RDBMS schema:
  
  actors ( id, ... )
  roles ( id, ... )
  actors_roles ( actor_id, role_id )
  
  The problem:
  
  With the attached repository.xml, when I remove a role from an actor,
  the role is also deleted from the table. But when I remove the actor,
  the role is kept ?
  
  All attribute are ( auto-update, auto-delete ) set to false.
  
  ojb-RC5
  
  //
  
  Transaction tx = odmg.newTransaction();
  tx.begin();
  Actor actor = retrieveActor(id);
  tx.lock(actor, Transaction.WRITE);
  actor.removeRole(0); ( delegate actor.roles.remove(0) on a vector )
  
  tx.commit();
  //
  
  
  
  
  
  __
  -
  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]



Re: using single persistence broker instance

2004-02-29 Thread Armin Waibel
Michael Mogley wrote:

EJB.  I currently create a single instance during ejbCreate and keep it
for the duration of
the bean's lifecycle.  This seems like the most efficient way.  Is there
a danger to using
this approach when transactions and distributed caches are involved?
Any other potential pitfalls?
There could be a problem with the used connections when read-only
operations (without tx-demarction) were performed. PB instance obtain a
connection when needed. On PB.commitTransaction/abortTransaction
(internal called in managed environments) and PB.close the connection
was released.


Are you saying the connection must be released for read-only operations?  I
am not understanding.  Is there some danger to keeping a connection
constantly open (so long as serial access is ensured)?
The connection/DataSource handle could get timed out and PB show 
unexpected behaviour. Could be a performance issue, hundered open 
connections (each bean has a bean pool and each bean may have an open 
connection). Your appServer connection pool could exhaust.

regards,
Armin
Michael


On read-only operations (no cm-tx or bm-tx used/required) only on
PB.close the connection was released. In your case the association
between PB instance and connection may not be detached. You can do this
manually by calling PB.serviceConnectionManager().releaseConnection()
for read-only methods, but I don't recommend this when using cm-tx,
because tx declaration in DD may change and I'm not sure if this could
cause side-effects.
regards,
Armin

Just want to make sure I'm not asking for trouble down the road.

Thanks,

Michael
-
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]


Re: [OTM] collection-descriptor issue with latest from head.

2004-02-29 Thread Oleg Nitz
On Sunday 29 February 2004 19:24, David Le Strat wrote:
 When I update the NodeImpl with new nodeProperties (a
 collection), only the NodeImpl object gets updated and
 the nodeProperies are not created.

 I have done this in the past with PB with out any
 issues.  Any suggestions?
With auto-update=true nodeProperties should be creates, whatever you use, 
PB or OTM. I've just done correspondent test with OTM, and all is okay.
So please send me code snippet or just try to do the same with PB.
OTM does nothing special when auto-update=true, it just calls 
broker.store(). If you set auto-update=false and otm-dependent=true, then 
OTM itself will take care about creation and deletion of nodeProperties.

Regards,
 Oleg


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]