Home » Modeling » EMF "Technology" (Ecore Tools, EMFatic, etc) » NPE in StoreResource.java:473
| NPE in StoreResource.java:473 [message #58482] |
Fri, 20 October 2006 18:24  |
Eclipse User |
|
|
|
Originally posted by: sprakash.tibco.com
Hello Martin,
I have a simple HibernateResource with a query, which as I see in my
debugger, Hibernate executes correctly and returns a single result.
However, Teneo NPEs while trying to make a list for the Hibernate
resource. Here is the exception (with some Hibernate debug log prepended):
Creating resource with query string:
hibernate://?dsname=matrix&query1=FROM Enterprise ent LEFT JOIN FETCH
ent.resourceConfigurations r WHERE r.id=11
6189 [main] DEBUG org.hibernate.SQL - select
enterprise0_."simpleeditableentity_id" as id1_1_0_,
resourceco2_."simpleeditableentity_id" as id1_1_1_, enterprise0_2_._vn_ as
column2_1_0_, enterprise0_2_."name" as name3_1_0_,
enterprise0_2_."description" as descript4_1_0_, enterprise0_1_."createdby"
as createdby2_4_0_, enterprise0_1_."modifiedby" as modifiedby3_4_0_,
enterprise0_1_."creationdate" as creation4_4_0_,
enterprise0_1_."modificationdate" as modifica5_4_0_,
enterprise0_."admincluster_id" as adminclu2_36_0_, resourceco2_2_._vn_ as
column2_1_1_, resourceco2_2_."name" as name3_1_1_,
resourceco2_2_."description" as descript4_1_1_, resourceco2_1_."createdby"
as createdby2_4_1_, resourceco2_1_."modifiedby" as modifiedby3_4_1_,
resourceco2_1_."creationdate" as creation4_4_1_,
resourceco2_1_."modificationdate" as modifica5_4_1_, resourceco2_."type"
as type2_55_1_, resourceco2_."contents" as contents3_55_1_,
resourceco2_.econtainer_class as econtainer4_55_1_,
resourceco2_.e_container as e5_55_1_, resourceco2_.e_container_featureid
as e6_55_1_, case when resourceco2_3_."resourceconfiguration_id" is not
null then 22 when resourceco2_."simpleeditableentity_id" is not null then
11 end as clazz_1_, resourceco1_.id as id0__, resourceco1_.elt as elt0__,
resourceco1_."enterprise_resourceconfigurations_idx" as enterprise3_0__
from "enterprise" enterprise0_ inner join "simpleeditableentity"
enterprise0_1_ on
enterprise0_."simpleeditableentity_id"=enterprise0_1_."adminentity_id "
inner join "adminentity" enterprise0_2_ on
enterprise0_."simpleeditableentity_id"=enterprise0_2_."id " left outer join
enterprise_resourceconfiguration resourceco1_ on
enterprise0_."simpleeditableentity_id"=resourceco1_.id left outer join
"resourceconfiguration" resourceco2_ on
resourceco1_.elt=resourceco2_."simpleeditableentity_id" left outer join
"simpleeditableentity" resourceco2_1_ on
resourceco2_."simpleeditableentity_id"=resourceco2_1_."adminentity_id "
left outer join "adminentity" resourceco2_2_ on
resourceco2_."simpleeditableentity_id"=resourceco2_2_."id " left outer join
"servicebustransportconfiguration" resourceco2_3_ on
resourceco2_."simpleeditableentity_id"=resourceco2_3_."resourceconfiguration_id "
where resourceco2_."simpleeditableentity_id"=11
6299 [main] DEBUG org.hibernate.SQL - select adminclust0_.e_id as
e1_86_0_, adminclust0_._vn_ as column2_86_0_, adminclust0_."clustername"
as clustern3_86_0_, adminclust0_.econtainer_class as econtainer4_86_0_,
adminclust0_.e_container as e5_86_0_, adminclust0_.e_container_featureid
as e6_86_0_ from "admincluster" adminclust0_ where adminclust0_.e_id=?
Exception in thread "main" java.lang.NullPointerException
at
org.eclipse.emf.teneo.resource.StoreResource.attachedHelper( StoreResource.java:473)
at
org.eclipse.emf.teneo.resource.StoreResource.attached(StoreR esource.java:425)
at
org.eclipse.emf.teneo.resource.StoreResource.load(StoreResou rce.java:252)
The exception seems to occur because StoreResource.attachedHelper is
handed a null object. This in turn happens because
getNonResolvingContent(eObject).basicIterator() returns an iterator which
has null objects in it. What exactly is getNonResolvingContent(eObject)?
eObject in my case has the real object i.e. the only result that Hibernate
returned.
When I put an NPE check in StoreResource.attachedHelper, everything is
fine. Is this a good enough fix (and should I file a bug report), or is
the problem happening in the first place because of some earlier mistake
on my part?
Thanks much,
Sundeep
|
|
| | | | |
| Re: NPE in StoreResource.java:473 [message #59842 is a reply to message #59799] |
Wed, 01 November 2006 02:38  |
| Eclipse User |
|
|
|
Sundeep,
Well afaics I told you that I was not sure that my changes would solve this issue. Also the bugzilla
is not set to fixed (because I was not sure).
This npe occurs when a new eobject is attached to the resource. The iterator returns null
apparently. The iterator is an innerclass of StoreResource which basically does the same as a
standard EMF iterator.
What you can try is see if the emf iterator also results in a npe (I don't expect this but need to
be sure). To test this you can add the following code to the attached method (in the beginning):
for (Iterator tree = new EContentsEList(eObject).basicIterator(); tree.hasNext();) {
System.err.println(tree.next().getClass().getName());
}
If you want to debug some more. The error can be in the hasNext() method in the featureiterator (see
anonymous class decl. in the StoreResource.getFeatureIterator method). The hasNext returns true
while there is no next (guessing here).
The difference with the standard emf implementation is in this piece of code:
if (feature.isMany()) {
Object value = eObject.eGet(feature, resolve());
if (value instanceof PersistableDelegateList
&& !((PersistableDelegateList) value).isLoaded()) {
continue;
}
if (value instanceof PersistableEMap && !((PersistableEMap) value).isLoaded()) {
continue;
}
}
If you have a reproducable test case/model which you provide then you can also send that to me so I
can try it out.
gr. Martin
Sundeep Prakash wrote:
> Martin,
> I was told that the latest build (26th Oct) solved this NPE (Bugzilla
> 162150). But apparently it still does not, in my tests. Do you expect to
> be resolving this bug anytime soon?
>
> Thanks,
>
> Sundeep
>
--
With Regards, Martin Taal
Springsite/Elver.org
Office: Hardwareweg 4, 3821 BV Amersfoort
Postal: Nassaulaan 7, 3941 EC Doorn
The Netherlands
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: NPE in StoreResource.java:473 [message #594521 is a reply to message #58482] |
Fri, 20 October 2006 18:53 |
| Eclipse User |
|
|
|
Sorry, the line numbers are wrong. It should be Line# 467 (that gives the
NPE). Also this is from the build titled I200609141628 (Thu, 14 Sep 2006
-- 16:28 (-0400)). The method doesnt appear to be different even in the
current build:
for (Iterator it = eObject.eClass().getEAllReferences().iterator();
it.hasNext();) {
|
|
|
| Re: NPE in StoreResource.java:473 [message #594530 is a reply to message #58484] |
Sat, 21 October 2006 01:37 |
| Eclipse User |
|
|
|
Hi Sundeep,
When an eobject is attached all its children are also attached to a resource. The basicIterator of
getNonResolvingContent(eObject) ensures that lazily loaded elists are not loaded in this step.
The npe you encounter seems to be a bug in the basicIterator (see the method call getFeatureIterator
in StoreResource). I checked and the local implementation of the FeatureIteratorImpl (created in
getFeatureIterator) seems to be misaligned with the implementation in ecore. I have changed this but
do not know if this solves your npe.
Can you enter a bugzilla for this?
I can do a new build. Do you maybe have time/the possibility to test if this new build solves the npe?
gr. Martin
Sundeep Prakash wrote:
> Sorry, the line numbers are wrong. It should be Line# 467 (that gives
> the NPE). Also this is from the build titled I200609141628 (Thu, 14 Sep
> 2006 -- 16:28 (-0400)). The method doesnt appear to be different even in
> the current build:
>
> for (Iterator it = eObject.eClass().getEAllReferences().iterator();
> it.hasNext();) {
>
--
With Regards, Martin Taal
Springsite/Elver.org
Office: Hardwareweg 4, 3821 BV Amersfoort
Postal: Nassaulaan 7, 3941 EC Doorn
The Netherlands
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: NPE in StoreResource.java:473 [message #595055 is a reply to message #59137] |
Wed, 01 November 2006 01:29 |
| Eclipse User |
|
|
|
Martin,
I was told that the latest build (26th Oct) solved this NPE (Bugzilla
162150). But apparently it still does not, in my tests. Do you expect to
be resolving this bug anytime soon?
Thanks,
Sundeep
|
|
|
| Re: NPE in StoreResource.java:473 [message #595066 is a reply to message #59799] |
Wed, 01 November 2006 02:38 |
| Eclipse User |
|
|
|
Sundeep,
Well afaics I told you that I was not sure that my changes would solve this issue. Also the bugzilla
is not set to fixed (because I was not sure).
This npe occurs when a new eobject is attached to the resource. The iterator returns null
apparently. The iterator is an innerclass of StoreResource which basically does the same as a
standard EMF iterator.
What you can try is see if the emf iterator also results in a npe (I don't expect this but need to
be sure). To test this you can add the following code to the attached method (in the beginning):
for (Iterator tree = new EContentsEList(eObject).basicIterator(); tree.hasNext();) {
System.err.println(tree.next().getClass().getName());
}
If you want to debug some more. The error can be in the hasNext() method in the featureiterator (see
anonymous class decl. in the StoreResource.getFeatureIterator method). The hasNext returns true
while there is no next (guessing here).
The difference with the standard emf implementation is in this piece of code:
if (feature.isMany()) {
Object value = eObject.eGet(feature, resolve());
if (value instanceof PersistableDelegateList
&& !((PersistableDelegateList) value).isLoaded()) {
continue;
}
if (value instanceof PersistableEMap && !((PersistableEMap) value).isLoaded()) {
continue;
}
}
If you have a reproducable test case/model which you provide then you can also send that to me so I
can try it out.
gr. Martin
Sundeep Prakash wrote:
> Martin,
> I was told that the latest build (26th Oct) solved this NPE (Bugzilla
> 162150). But apparently it still does not, in my tests. Do you expect to
> be resolving this bug anytime soon?
>
> Thanks,
>
> Sundeep
>
--
With Regards, Martin Taal
Springsite/Elver.org
Office: Hardwareweg 4, 3821 BV Amersfoort
Postal: Nassaulaan 7, 3941 EC Doorn
The Netherlands
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
|
|
|
Goto Forum:
Current Time: Wed Apr 30 12:20:33 EDT 2025
Powered by FUDForum. Page generated in 0.10198 seconds |
Home