Home » Modeling » EMF » [Teneo] updates E_Version column
| [Teneo] updates E_Version column [message #501551] |
Wed, 02 December 2009 18:44  |
Eclipse User |
|
|
|
Originally posted by: jandev.gmx.ch
Hi all
When we commit a hibernate transaction, we see hibernate searches for dirty model objects (TypeFactory.findDirty).
Our EMF models, which are contained by another model (containment=true), have a property named "e_container_featurename".
Hibernate is comparing all properties of the cached object vs. the object from the session.
To find out if the property "e_container_featurename" (EContainerFeatureIDHolder.class) has been changed, hibernate uses Teneo's EContainerFeatureIDUserType.equals. The
EContainerFeatureIDUserType.equals does some null checks first and then calls "x.equals(y)" which is Object.equals as EContainerFeatureIDHolder does not implement it.
Unfortunately, the the two EContainerFeatureIDHolder of both objects compared are different objects but having the same values (EClass, EStructuralFeature).
Even if we haven't change any properties, our model gets dirty. This results in an unnecessary update on the table, increasing the E_VERSION and causing a concurrent modification in certain
circumstances.
1) Is it correct to have the property "e_container_featurename" on the models (containment) and in database tables?
2) Is this a bug? We believe that this issue can be fixed, when equals is implemented in EContainerFeatureIDHolder.
We use hibernate 3.2.6-GA, teneo 1.1.0 and EMF 2.5.
Best regards
Jan Dev
|
|
|
| Re: [Teneo] updates E_Version column [message #501554 is a reply to message #501551] |
Wed, 02 December 2009 18:58  |
 Martin Taal
Messages: 5468
Registered: July 2009 |
Senior Member |
|
|
Hi Jan,
Yes this seems to be a bug for sure. Can you enter a bugzilla for this?
The e_container_featurename is needed if you want to use the eContainer member of an EObject. EMF makes use of it in
much of its code, so if you use core EMF (for example in a RCP) then you need this probably. You can disable mapping of
the econtainer using the option:
PersistenceOptions.DISABLE_ECONTAINER_MAPPING
set it to true.
If you disable container mapping then the econtainer of an eobject will only be set if you first read a parent and then
access the children through the parent. If the children are read individually from the database then there econtainer
won't be set.
gr. Martin
Jan Dev wrote:
> Hi all
>
> When we commit a hibernate transaction, we see hibernate searches for
> dirty model objects (TypeFactory.findDirty).
> Our EMF models, which are contained by another model (containment=true),
> have a property named "e_container_featurename".
>
> Hibernate is comparing all properties of the cached object vs. the
> object from the session.
> To find out if the property "e_container_featurename"
> (EContainerFeatureIDHolder.class) has been changed, hibernate uses
> Teneo's EContainerFeatureIDUserType.equals. The
> EContainerFeatureIDUserType.equals does some null checks first and then
> calls "x.equals(y)" which is Object.equals as EContainerFeatureIDHolder
> does not implement it.
>
> Unfortunately, the the two EContainerFeatureIDHolder of both objects
> compared are different objects but having the same values (EClass,
> EStructuralFeature).
>
> Even if we haven't change any properties, our model gets dirty. This
> results in an unnecessary update on the table, increasing the E_VERSION
> and causing a concurrent modification in certain circumstances.
>
>
> 1) Is it correct to have the property "e_container_featurename" on the
> models (containment) and in database tables?
>
> 2) Is this a bug? We believe that this issue can be fixed, when equals
> is implemented in EContainerFeatureIDHolder.
>
>
> We use hibernate 3.2.6-GA, teneo 1.1.0 and EMF 2.5.
>
> Best regards
>
> Jan Dev
--
With Regards, Martin Taal
Springsite/Elver.org
Office: Hardwareweg 4, 3821 BV Amersfoort
Postal: Nassaulaan 7, 3941 EC Doorn
The Netherlands
Cell: +31 (0)6 288 48 943
Tel: +31 (0)84 420 2397
Fax: +31 (0)84 225 9307
Mail: mtaal@springsite.com - mtaal@elver.org
Web: www.springsite.com - www.elver.org
|
|
|
| Re: [Teneo] updates E_Version column [message #501726 is a reply to message #501554] |
Thu, 03 December 2009 15:36 |
| Eclipse User |
|
|
|
Originally posted by: jandev.gmx.ch
Hi Martin
Thanks for the very quick response!
I have filed the bug [296816] and added a patch to fix the issue.
I just implemented equals and hashcode of EContainerFeatureIDHolder and this seems to work.
What workaround do you suggest until this fix has been integrated and release is available?
I can think of:
a) Write my own UserType, which is just a copy of the fixed one and register it instead of the original one.
Can you give me a hint or direction how I can make sure, my new UserType is used instead of the original?
b) Set the PersistenceOptions.DISABLE_ECONTAINER_MAPPING=true
Regards
Jan
[296816]: https://bugs.eclipse.org/bugs/show_bug.cgi?id=296816
Martin Taal wrote:
> Hi Jan,
> Yes this seems to be a bug for sure. Can you enter a bugzilla for this?
>
> The e_container_featurename is needed if you want to use the eContainer
> member of an EObject. EMF makes use of it in much of its code, so if you
> use core EMF (for example in a RCP) then you need this probably. You can
> disable mapping of the econtainer using the option:
> PersistenceOptions.DISABLE_ECONTAINER_MAPPING
> set it to true.
>
> If you disable container mapping then the econtainer of an eobject will
> only be set if you first read a parent and then access the children
> through the parent. If the children are read individually from the
> database then there econtainer won't be set.
>
> gr. Martin
>
> Jan Dev wrote:
>> Hi all
>>
>> When we commit a hibernate transaction, we see hibernate searches for
>> dirty model objects (TypeFactory.findDirty).
>> Our EMF models, which are contained by another model
>> (containment=true), have a property named "e_container_featurename".
>>
>> Hibernate is comparing all properties of the cached object vs. the
>> object from the session.
>> To find out if the property "e_container_featurename"
>> (EContainerFeatureIDHolder.class) has been changed, hibernate uses
>> Teneo's EContainerFeatureIDUserType.equals. The
>> EContainerFeatureIDUserType.equals does some null checks first and
>> then calls "x.equals(y)" which is Object.equals as
>> EContainerFeatureIDHolder does not implement it.
>>
>> Unfortunately, the the two EContainerFeatureIDHolder of both objects
>> compared are different objects but having the same values (EClass,
>> EStructuralFeature).
>>
>> Even if we haven't change any properties, our model gets dirty. This
>> results in an unnecessary update on the table, increasing the
>> E_VERSION and causing a concurrent modification in certain circumstances.
>>
>>
>> 1) Is it correct to have the property "e_container_featurename" on the
>> models (containment) and in database tables?
>>
>> 2) Is this a bug? We believe that this issue can be fixed, when equals
>> is implemented in EContainerFeatureIDHolder.
>>
>>
>> We use hibernate 3.2.6-GA, teneo 1.1.0 and EMF 2.5.
>>
>> Best regards
>>
>> Jan Dev
>
>
|
|
|
| Re: [Teneo] updates E_Version column [message #501749 is a reply to message #501726] |
Thu, 03 December 2009 16:32 |
Martin Taal
Messages: 5468
Registered: July 2009 |
Senior Member |
|
|
Hi Jan,
Thanks, seen it!
I was planning to do a new build anyway the coming days (before monday then). And this change seems fairly small, so I
can incorporate it easily.
Disabling the econtainer persistence will mean that these columns are not updated anymore, so if you enable this later I
think you might get into trouble with your database.
Currently it is a bit hard/impossible to override this with your own implementation I am afraid.
gr. Martin
Jan Dev wrote:
> Hi Martin
>
> Thanks for the very quick response!
>
> I have filed the bug [296816] and added a patch to fix the issue.
> I just implemented equals and hashcode of EContainerFeatureIDHolder and
> this seems to work.
>
> What workaround do you suggest until this fix has been integrated and
> release is available?
>
> I can think of:
> a) Write my own UserType, which is just a copy of the fixed one and
> register it instead of the original one.
> Can you give me a hint or direction how I can make sure, my new UserType
> is used instead of the original?
>
> b) Set the PersistenceOptions.DISABLE_ECONTAINER_MAPPING=true
>
>
> Regards
>
> Jan
>
>
> [296816]: https://bugs.eclipse.org/bugs/show_bug.cgi?id=296816
>
> Martin Taal wrote:
>> Hi Jan,
>> Yes this seems to be a bug for sure. Can you enter a bugzilla for this?
>>
>> The e_container_featurename is needed if you want to use the
>> eContainer member of an EObject. EMF makes use of it in much of its
>> code, so if you use core EMF (for example in a RCP) then you need this
>> probably. You can disable mapping of the econtainer using the option:
>> PersistenceOptions.DISABLE_ECONTAINER_MAPPING
>> set it to true.
>>
>> If you disable container mapping then the econtainer of an eobject
>> will only be set if you first read a parent and then access the
>> children through the parent. If the children are read individually
>> from the database then there econtainer won't be set.
>>
>> gr. Martin
>>
>> Jan Dev wrote:
>>> Hi all
>>>
>>> When we commit a hibernate transaction, we see hibernate searches for
>>> dirty model objects (TypeFactory.findDirty).
>>> Our EMF models, which are contained by another model
>>> (containment=true), have a property named "e_container_featurename".
>>>
>>> Hibernate is comparing all properties of the cached object vs. the
>>> object from the session.
>>> To find out if the property "e_container_featurename"
>>> (EContainerFeatureIDHolder.class) has been changed, hibernate uses
>>> Teneo's EContainerFeatureIDUserType.equals. The
>>> EContainerFeatureIDUserType.equals does some null checks first and
>>> then calls "x.equals(y)" which is Object.equals as
>>> EContainerFeatureIDHolder does not implement it.
>>>
>>> Unfortunately, the the two EContainerFeatureIDHolder of both objects
>>> compared are different objects but having the same values (EClass,
>>> EStructuralFeature).
>>>
>>> Even if we haven't change any properties, our model gets dirty. This
>>> results in an unnecessary update on the table, increasing the
>>> E_VERSION and causing a concurrent modification in certain
>>> circumstances.
>>>
>>>
>>> 1) Is it correct to have the property "e_container_featurename" on
>>> the models (containment) and in database tables?
>>>
>>> 2) Is this a bug? We believe that this issue can be fixed, when
>>> equals is implemented in EContainerFeatureIDHolder.
>>>
>>>
>>> We use hibernate 3.2.6-GA, teneo 1.1.0 and EMF 2.5.
>>>
>>> Best regards
>>>
>>> Jan Dev
>>
>>
--
With Regards, Martin Taal
Springsite/Elver.org
Office: Hardwareweg 4, 3821 BV Amersfoort
Postal: Nassaulaan 7, 3941 EC Doorn
The Netherlands
Cell: +31 (0)6 288 48 943
Tel: +31 (0)84 420 2397
Fax: +31 (0)84 225 9307
Mail: mtaal@springsite.com - mtaal@elver.org
Web: www.springsite.com - www.elver.org
|
|
|
| Re: [Teneo] updates E_Version column [message #501976 is a reply to message #501749] |
Fri, 04 December 2009 15:59 |
| Eclipse User |
|
|
|
Originally posted by: jandev.gmx.ch
Martin
I've seen that you've added the path for teneo 1.2 (bugzilla).
Will you add the patch on the 1.1x branch as well? (We're using EMF 2.5 and Eclipse 3.5)
Regards
Jan
Martin Taal wrote:
> Hi Jan,
> Thanks, seen it!
> I was planning to do a new build anyway the coming days (before monday
> then). And this change seems fairly small, so I can incorporate it easily.
>
> Disabling the econtainer persistence will mean that these columns are
> not updated anymore, so if you enable this later I think you might get
> into trouble with your database.
>
> Currently it is a bit hard/impossible to override this with your own
> implementation I am afraid.
>
> gr. Martin
>
> Jan Dev wrote:
>> Hi Martin
>>
>> Thanks for the very quick response!
>>
>> I have filed the bug [296816] and added a patch to fix the issue.
>> I just implemented equals and hashcode of EContainerFeatureIDHolder
>> and this seems to work.
>>
>> What workaround do you suggest until this fix has been integrated and
>> release is available?
>>
>> I can think of:
>> a) Write my own UserType, which is just a copy of the fixed one and
>> register it instead of the original one.
>> Can you give me a hint or direction how I can make sure, my new
>> UserType is used instead of the original?
>>
>> b) Set the PersistenceOptions.DISABLE_ECONTAINER_MAPPING=true
>>
>>
>> Regards
>>
>> Jan
>>
>>
>> [296816]: https://bugs.eclipse.org/bugs/show_bug.cgi?id=296816
>>
>> Martin Taal wrote:
>>> Hi Jan,
>>> Yes this seems to be a bug for sure. Can you enter a bugzilla for this?
>>>
>>> The e_container_featurename is needed if you want to use the
>>> eContainer member of an EObject. EMF makes use of it in much of its
>>> code, so if you use core EMF (for example in a RCP) then you need
>>> this probably. You can disable mapping of the econtainer using the
>>> option:
>>> PersistenceOptions.DISABLE_ECONTAINER_MAPPING
>>> set it to true.
>>>
>>> If you disable container mapping then the econtainer of an eobject
>>> will only be set if you first read a parent and then access the
>>> children through the parent. If the children are read individually
>>> from the database then there econtainer won't be set.
>>>
>>> gr. Martin
>>>
>>> Jan Dev wrote:
>>>> Hi all
>>>>
>>>> When we commit a hibernate transaction, we see hibernate searches
>>>> for dirty model objects (TypeFactory.findDirty).
>>>> Our EMF models, which are contained by another model
>>>> (containment=true), have a property named "e_container_featurename".
>>>>
>>>> Hibernate is comparing all properties of the cached object vs. the
>>>> object from the session.
>>>> To find out if the property "e_container_featurename"
>>>> (EContainerFeatureIDHolder.class) has been changed, hibernate uses
>>>> Teneo's EContainerFeatureIDUserType.equals. The
>>>> EContainerFeatureIDUserType.equals does some null checks first and
>>>> then calls "x.equals(y)" which is Object.equals as
>>>> EContainerFeatureIDHolder does not implement it.
>>>>
>>>> Unfortunately, the the two EContainerFeatureIDHolder of both objects
>>>> compared are different objects but having the same values (EClass,
>>>> EStructuralFeature).
>>>>
>>>> Even if we haven't change any properties, our model gets dirty. This
>>>> results in an unnecessary update on the table, increasing the
>>>> E_VERSION and causing a concurrent modification in certain
>>>> circumstances.
>>>>
>>>>
>>>> 1) Is it correct to have the property "e_container_featurename" on
>>>> the models (containment) and in database tables?
>>>>
>>>> 2) Is this a bug? We believe that this issue can be fixed, when
>>>> equals is implemented in EContainerFeatureIDHolder.
>>>>
>>>>
>>>> We use hibernate 3.2.6-GA, teneo 1.1.0 and EMF 2.5.
>>>>
>>>> Best regards
>>>>
>>>> Jan Dev
>>>
>>>
>
>
|
|
|
| Re: [Teneo] updates E_Version column [message #501980 is a reply to message #501976] |
Fri, 04 December 2009 16:10 |
Martin Taal
Messages: 5468
Registered: July 2009 |
Senior Member |
|
|
Hi Jan,
Yes my typo. I just now did a 1.1.2 build with the fix and am publishing it right now. Although it seems that something
went wrong in the publish step (some low-level file system permission). In the meantime you can download it directly
here also:
http://modeling.eclipse.org/modeling/emf/downloads/?project= teneo&sortBy=date&hlbuild=0#latest
(take the last 1.1.2 build).
gr. Martin
Jan Dev wrote:
> Martin
>
> I've seen that you've added the path for teneo 1.2 (bugzilla).
> Will you add the patch on the 1.1x branch as well? (We're using EMF 2.5
> and Eclipse 3.5)
>
> Regards
>
> Jan
>
> Martin Taal wrote:
>> Hi Jan,
>> Thanks, seen it!
>> I was planning to do a new build anyway the coming days (before monday
>> then). And this change seems fairly small, so I can incorporate it
>> easily.
>>
>> Disabling the econtainer persistence will mean that these columns are
>> not updated anymore, so if you enable this later I think you might get
>> into trouble with your database.
>>
>> Currently it is a bit hard/impossible to override this with your own
>> implementation I am afraid.
>>
>> gr. Martin
>>
>> Jan Dev wrote:
>>> Hi Martin
>>>
>>> Thanks for the very quick response!
>>>
>>> I have filed the bug [296816] and added a patch to fix the issue.
>>> I just implemented equals and hashcode of EContainerFeatureIDHolder
>>> and this seems to work.
>>>
>>> What workaround do you suggest until this fix has been integrated and
>>> release is available?
>>>
>>> I can think of:
>>> a) Write my own UserType, which is just a copy of the fixed one and
>>> register it instead of the original one.
>>> Can you give me a hint or direction how I can make sure, my new
>>> UserType is used instead of the original?
>>>
>>> b) Set the PersistenceOptions.DISABLE_ECONTAINER_MAPPING=true
>>>
>>>
>>> Regards
>>>
>>> Jan
>>>
>>>
>>> [296816]: https://bugs.eclipse.org/bugs/show_bug.cgi?id=296816
>>>
>>> Martin Taal wrote:
>>>> Hi Jan,
>>>> Yes this seems to be a bug for sure. Can you enter a bugzilla for this?
>>>>
>>>> The e_container_featurename is needed if you want to use the
>>>> eContainer member of an EObject. EMF makes use of it in much of its
>>>> code, so if you use core EMF (for example in a RCP) then you need
>>>> this probably. You can disable mapping of the econtainer using the
>>>> option:
>>>> PersistenceOptions.DISABLE_ECONTAINER_MAPPING
>>>> set it to true.
>>>>
>>>> If you disable container mapping then the econtainer of an eobject
>>>> will only be set if you first read a parent and then access the
>>>> children through the parent. If the children are read individually
>>>> from the database then there econtainer won't be set.
>>>>
>>>> gr. Martin
>>>>
>>>> Jan Dev wrote:
>>>>> Hi all
>>>>>
>>>>> When we commit a hibernate transaction, we see hibernate searches
>>>>> for dirty model objects (TypeFactory.findDirty).
>>>>> Our EMF models, which are contained by another model
>>>>> (containment=true), have a property named "e_container_featurename".
>>>>>
>>>>> Hibernate is comparing all properties of the cached object vs. the
>>>>> object from the session.
>>>>> To find out if the property "e_container_featurename"
>>>>> (EContainerFeatureIDHolder.class) has been changed, hibernate uses
>>>>> Teneo's EContainerFeatureIDUserType.equals. The
>>>>> EContainerFeatureIDUserType.equals does some null checks first and
>>>>> then calls "x.equals(y)" which is Object.equals as
>>>>> EContainerFeatureIDHolder does not implement it.
>>>>>
>>>>> Unfortunately, the the two EContainerFeatureIDHolder of both
>>>>> objects compared are different objects but having the same values
>>>>> (EClass, EStructuralFeature).
>>>>>
>>>>> Even if we haven't change any properties, our model gets dirty.
>>>>> This results in an unnecessary update on the table, increasing the
>>>>> E_VERSION and causing a concurrent modification in certain
>>>>> circumstances.
>>>>>
>>>>>
>>>>> 1) Is it correct to have the property "e_container_featurename" on
>>>>> the models (containment) and in database tables?
>>>>>
>>>>> 2) Is this a bug? We believe that this issue can be fixed, when
>>>>> equals is implemented in EContainerFeatureIDHolder.
>>>>>
>>>>>
>>>>> We use hibernate 3.2.6-GA, teneo 1.1.0 and EMF 2.5.
>>>>>
>>>>> Best regards
>>>>>
>>>>> Jan Dev
>>>>
>>>>
>>
>>
--
With Regards, Martin Taal
Springsite/Elver.org
Office: Hardwareweg 4, 3821 BV Amersfoort
Postal: Nassaulaan 7, 3941 EC Doorn
The Netherlands
Cell: +31 (0)6 288 48 943
Tel: +31 (0)84 420 2397
Fax: +31 (0)84 225 9307
Mail: mtaal@springsite.com - mtaal@elver.org
Web: www.springsite.com - www.elver.org
|
|
|
| Re: [Teneo] updates E_Version column [message #502646 is a reply to message #501980] |
Wed, 09 December 2009 09:32 |
| Eclipse User |
|
|
|
Originally posted by: jandev.gmx.ch
Hi Martin
I've tested the latest teneo 1.1.2 build (M200912041048 (2009/12/04)) and I confirm that bug [296816] has been fixed.
Many thanks and best regards
Jan
Martin Taal wrote:
> Hi Jan,
> Yes my typo. I just now did a 1.1.2 build with the fix and am publishing
> it right now. Although it seems that something went wrong in the publish
> step (some low-level file system permission). In the meantime you can
> download it directly here also:
> http://modeling.eclipse.org/modeling/emf/downloads/?project= teneo&sortBy=date&hlbuild=0#latest
>
> (take the last 1.1.2 build).
>
> gr. Martin
>
> Jan Dev wrote:
>> Martin
>>
>> I've seen that you've added the path for teneo 1.2 (bugzilla).
>> Will you add the patch on the 1.1x branch as well? (We're using EMF
>> 2.5 and Eclipse 3.5)
>>
>> Regards
>>
>> Jan
>>
>> Martin Taal wrote:
>>> Hi Jan,
>>> Thanks, seen it!
>>> I was planning to do a new build anyway the coming days (before
>>> monday then). And this change seems fairly small, so I can
>>> incorporate it easily.
>>>
>>> Disabling the econtainer persistence will mean that these columns are
>>> not updated anymore, so if you enable this later I think you might
>>> get into trouble with your database.
>>>
>>> Currently it is a bit hard/impossible to override this with your own
>>> implementation I am afraid.
>>>
>>> gr. Martin
>>>
>>> Jan Dev wrote:
>>>> Hi Martin
>>>>
>>>> Thanks for the very quick response!
>>>>
>>>> I have filed the bug [296816] and added a patch to fix the issue.
>>>> I just implemented equals and hashcode of EContainerFeatureIDHolder
>>>> and this seems to work.
>>>>
>>>> What workaround do you suggest until this fix has been integrated
>>>> and release is available?
>>>>
>>>> I can think of:
>>>> a) Write my own UserType, which is just a copy of the fixed one and
>>>> register it instead of the original one.
>>>> Can you give me a hint or direction how I can make sure, my new
>>>> UserType is used instead of the original?
>>>>
>>>> b) Set the PersistenceOptions.DISABLE_ECONTAINER_MAPPING=true
>>>>
>>>>
>>>> Regards
>>>>
>>>> Jan
>>>>
>>>>
>>>> [296816]: https://bugs.eclipse.org/bugs/show_bug.cgi?id=296816
>>>>
>>>> Martin Taal wrote:
>>>>> Hi Jan,
>>>>> Yes this seems to be a bug for sure. Can you enter a bugzilla for
>>>>> this?
>>>>>
>>>>> The e_container_featurename is needed if you want to use the
>>>>> eContainer member of an EObject. EMF makes use of it in much of its
>>>>> code, so if you use core EMF (for example in a RCP) then you need
>>>>> this probably. You can disable mapping of the econtainer using the
>>>>> option:
>>>>> PersistenceOptions.DISABLE_ECONTAINER_MAPPING
>>>>> set it to true.
>>>>>
>>>>> If you disable container mapping then the econtainer of an eobject
>>>>> will only be set if you first read a parent and then access the
>>>>> children through the parent. If the children are read individually
>>>>> from the database then there econtainer won't be set.
>>>>>
>>>>> gr. Martin
>>>>>
>>>>> Jan Dev wrote:
>>>>>> Hi all
>>>>>>
>>>>>> When we commit a hibernate transaction, we see hibernate searches
>>>>>> for dirty model objects (TypeFactory.findDirty).
>>>>>> Our EMF models, which are contained by another model
>>>>>> (containment=true), have a property named "e_container_featurename".
>>>>>>
>>>>>> Hibernate is comparing all properties of the cached object vs. the
>>>>>> object from the session.
>>>>>> To find out if the property "e_container_featurename"
>>>>>> (EContainerFeatureIDHolder.class) has been changed, hibernate uses
>>>>>> Teneo's EContainerFeatureIDUserType.equals. The
>>>>>> EContainerFeatureIDUserType.equals does some null checks first and
>>>>>> then calls "x.equals(y)" which is Object.equals as
>>>>>> EContainerFeatureIDHolder does not implement it.
>>>>>>
>>>>>> Unfortunately, the the two EContainerFeatureIDHolder of both
>>>>>> objects compared are different objects but having the same values
>>>>>> (EClass, EStructuralFeature).
>>>>>>
>>>>>> Even if we haven't change any properties, our model gets dirty.
>>>>>> This results in an unnecessary update on the table, increasing the
>>>>>> E_VERSION and causing a concurrent modification in certain
>>>>>> circumstances.
>>>>>>
>>>>>>
>>>>>> 1) Is it correct to have the property "e_container_featurename" on
>>>>>> the models (containment) and in database tables?
>>>>>>
>>>>>> 2) Is this a bug? We believe that this issue can be fixed, when
>>>>>> equals is implemented in EContainerFeatureIDHolder.
>>>>>>
>>>>>>
>>>>>> We use hibernate 3.2.6-GA, teneo 1.1.0 and EMF 2.5.
>>>>>>
>>>>>> Best regards
>>>>>>
>>>>>> Jan Dev
>>>>>
>>>>>
>>>
>>>
>
>
|
|
|
Goto Forum:
Current Time: Tue Apr 16 18:35:23 GMT 2024
Powered by FUDForum. Page generated in 1.57527 seconds |
] 
Search
Help
Register
Login
Home
