cocoon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Giacomo Pati <>
Subject Re: Refactoring cocoon-thread
Date Mon, 10 Sep 2007 11:10:00 GMT
Hash: SHA1

Grzegorz Kossakowski wrote:
> Giacomo Pati pisze:
>> Carsten Ziegeler wrote:
>>> We've already done some incompatible changes in trunk, I would like to
>>> add one more which should not effect most of our users.
>> I was trying to find the other "incompatible changes in trunk" as since I've updated
my repo my app
>> isn't running again. On any request it throws a
>>         ... 46 more
>> Caused by: java.lang.NullPointerException
>>         at java.util.HashMap.putAll(
>>         at org.apache.cocoon.el.impl.objectmodel.ObjectModelImpl.setParent(
>>         at
>> org.apache.cocoon.components.pipeline.spring.PipelineComponentScope.get(
>>         at
>>         at
>>         at
>>         at org.springframework.aop.framework.JdkDynamicAopProxy.invoke(
>>         at $Proxy1.markLocalContext(Unknown Source)
>>         at org.apache.cocoon.components.treeprocessor.InvokeContext.pushMap(
>>         at
>> org.apache.cocoon.components.treeprocessor.AbstractParentProcessingNode.invokeNodes(
>>         at org.apache.cocoon.components.treeprocessor.sitemap.MatchNode.invoke(
>>         ... 79 more
>> Anybody any pointers I could look at?

> Are you sure you are running latest trunk? 

Hmm.. if a 'svn up' isn't enough to get there, tell me how else.

> You are having a trouble with the code handling
> pipelineComponent scope and proper initialization of parent Object Model within it. 

Tell me news, dear ;-)

> I explained the
> problem that this code solves here[1] and idea behind the code here[2]. As Joerg pointed
out[3] this
> code is rather first sketch and must be rewritten to follow best practices in Spring.

Ok, so is it working or not!

> When it comes to your problem, I guess that your problem is caused by the fact that
> PipelineComponentScope tries to set parent Object Model that is null. Such situation
should never
> occur and if parent Object Model is null there should be no setParent() call.

If it cannot happen, why is it happening to me, than?

> Do you use mounting in your sitemaps?


            <map:match pattern="*/**">
                <map:mount uri-prefix="{1}" src="blockcontext:/{1}/" />

Is that causing the troubles?

Thanks and ciao

- --
Giacomo Pati
Otego AG, Switzerland -
Orixo, the XML business alliance -

Version: GnuPG v2.0.6 (GNU/Linux)


View raw message