Enunciate

StackOverflowError in MapTypeUtil

Details

  • Type: Bug Bug
  • Status: Closed Closed
  • Priority: Major Major
  • Resolution: Fixed
  • Affects Version/s: 1.21
  • Fix Version/s: 1.21.1, 1.26
  • Component/s: None
  • Labels:
    None
  • Number of attachments :
    0

Description

java.lang.StackOverflowError
        at com.sun.tools.javac.code.Type$ClassType.toString(Type.java:574)
        at com.sun.tools.apt.mirror.type.DeclaredTypeImpl.toString(DeclaredTypeImpl.java:92)
        at com.sun.tools.apt.mirror.type.DeclaredTypeImpl.toString(DeclaredTypeImpl.java:42)
        at com.sun.tools.apt.mirror.type.ClassTypeImpl.toString(ClassTypeImpl.java:22)
        at java.lang.String.valueOf(String.java:2826)
        at java.lang.StringBuilder.append(StringBuilder.java:115)
        at com.sun.mirror.type.MirroredTypeException.<init>(MirroredTypeException.java:36)
        at com.sun.tools.apt.mirror.declaration.AnnotationProxyMaker$MirroredTypeExceptionProxy.<init>(AnnotationProxyMaker.java:262)
        at com.sun.tools.apt.mirror.declaration.AnnotationProxyMaker$ValueVisitor.visitClass(AnnotationProxyMaker.java:161)
        at com.sun.tools.javac.code.Attribute$Class.accept(Attribute.java:94)
        at com.sun.tools.apt.mirror.declaration.AnnotationProxyMaker$ValueVisitor.getValue(AnnotationProxyMaker.java:146)
        at com.sun.tools.apt.mirror.declaration.AnnotationProxyMaker.generateValue(AnnotationProxyMaker.java:124)
        at com.sun.tools.apt.mirror.declaration.AnnotationProxyMaker.getAllReflectedValues(AnnotationProxyMaker.java:80)
        at com.sun.tools.apt.mirror.declaration.AnnotationProxyMaker.generateAnnotation(AnnotationProxyMaker.java:65)
        at com.sun.tools.apt.mirror.declaration.AnnotationProxyMaker.generateAnnotation(AnnotationProxyMaker.java:57)
        at com.sun.tools.apt.mirror.declaration.DeclarationImpl.getAnnotation(DeclarationImpl.java:121)
        at com.sun.tools.apt.mirror.declaration.DeclarationImpl.getAnnotation(DeclarationImpl.java:109)
        at org.codehaus.enunciate.contract.jaxb.adapters.AdapterUtil.findAdapterType(AdapterUtil.java:100)
        at org.codehaus.enunciate.contract.jaxb.adapters.AdapterUtil.findAdapterType(AdapterUtil.java:76)
        at org.codehaus.enunciate.util.MapTypeUtil.findMapType(MapTypeUtil.java:80)
        at org.codehaus.enunciate.util.MapTypeUtil.findMapType(MapTypeUtil.java:88)
        at org.codehaus.enunciate.util.MapTypeUtil.findMapType(MapTypeUtil.java:82)
        at org.codehaus.enunciate.util.MapTypeUtil.findMapType(MapTypeUtil.java:88)
        at org.codehaus.enunciate.util.MapTypeUtil.findMapType(MapTypeUtil.java:82)
        at org.codehaus.enunciate.util.MapTypeUtil.findMapType(MapTypeUtil.java:88)
        at org.codehaus.enunciate.util.MapTypeUtil.findMapType(MapTypeUtil.java:82)
        at org.codehaus.enunciate.util.MapTypeUtil.findMapType(MapTypeUtil.java:88)
        at org.codehaus.enunciate.util.MapTypeUtil.findMapType(MapTypeUtil.java:82)
        ....... long series of the last two lines .................       
        at org.codehaus.enunciate.util.MapTypeUtil.findMapType(MapTypeUtil.java:88)
        at org.codehaus.enunciate.util.MapTypeUtil.findMapType(MapTypeUtil.java:82)
        at org.codehaus.enunciate.util.MapTypeUtil.findMapType(MapTypeUtil.java:88)
        at org.codehaus.enunciate.util.MapTypeUtil.findMapType(MapTypeUtil.java:82)
        at org.codehaus.enunciate.util.MapTypeUtil.findMapType(MapTypeUtil.java:88)
        at org.codehaus.enunciate.util.MapTypeUtil.findMapType(MapTypeUtil.java:82)
        at org.codehaus.enunciate.util.MapTypeUtil.findMapType(MapTypeUtil.java:88)
        at org.codehaus.enunciate.util.MapTypeUtil.findMapType(MapTypeUtil.java:82)
        at org.codehaus.enunciate.util.MapTypeUtil.findMapType(MapTypeUtil.java:88)
        at org.codehaus.enunciate.util.MapTypeUtil.findMapType(MapTypeUtil.java:82)

Activity

Ascending order - Click to sort in descending order
Hide
Permalink
Ryan Heaton added a comment -

Happens when an interface adapts to one of its impl classes.

Show
Ryan Heaton added a comment - Happens when an interface adapts to one of its impl classes.
Hide
Permalink
Ryan Heaton added a comment -

Circular adapter. We could put some heuristics in to break out of infinite loop, but I think for now we'll just say "don't do that".

Show
Ryan Heaton added a comment - Circular adapter. We could put some heuristics in to break out of infinite loop, but I think for now we'll just say "don't do that".
Hide
Permalink
Ryan Heaton added a comment -

Planning to cut a 1.21.1 release. This change will be included in that release.

Show
Ryan Heaton added a comment - Planning to cut a 1.21.1 release. This change will be included in that release.
Hide
Permalink
Ryan Heaton added a comment -

closed with release of 1.21.1

Show
Ryan Heaton added a comment - closed with release of 1.21.1
Hide
Permalink
Rob Sonke added a comment -

Could you explain why this is bad practice? The CXF documentation is giving this as a solution as well:
http://cxf.apache.org/docs/a-simple-jax-ws-service.html

It's blocking me now from using enunciate and I cant drop the XmlJavaTypeAdapters.

Thanks for your answer.

Show
Rob Sonke added a comment - Could you explain why this is bad practice? The CXF documentation is giving this as a solution as well: http://cxf.apache.org/docs/a-simple-jax-ws-service.html It's blocking me now from using enunciate and I cant drop the XmlJavaTypeAdapters. Thanks for your answer.
Hide
Permalink
Ryan Heaton added a comment -

It's not that it's bad practice; it's just that I couldn't figure out for the life of me how to support the circular reference.

Are you certain this is your same case? Is the stack trace the same? I looked at the cxf link and Enunciate should be able to support that code.

Show
Ryan Heaton added a comment - It's not that it's bad practice; it's just that I couldn't figure out for the life of me how to support the circular reference. Are you certain this is your same case? Is the stack trace the same? I looked at the cxf link and Enunciate should be able to support that code.
Hide
Permalink
Rob Sonke added a comment -

Thanks, but yes, the stacktrace is the same. I'm pretty sure that it's caused by an Adapter like this:

public class WSCountryAdapter extends XmlAdapter<WSCountryImpl, WSCountry>
{
public WSCountryImpl marshal(WSCountry v) throws Exception

{ if (v instanceof WSCountryImpl) return (WSCountryImpl) v; return new WSCountryImpl(); }

public WSCountry unmarshal(WSCountryImpl v) throws Exception

{ return v; }

}

Where WSCountry is the interface and WSCountryImpl the implementation class.

Show
Rob Sonke added a comment - Thanks, but yes, the stacktrace is the same. I'm pretty sure that it's caused by an Adapter like this: public class WSCountryAdapter extends XmlAdapter<WSCountryImpl, WSCountry> { public WSCountryImpl marshal(WSCountry v) throws Exception { if (v instanceof WSCountryImpl) return (WSCountryImpl) v; return new WSCountryImpl(); } public WSCountry unmarshal(WSCountryImpl v) throws Exception { return v; } } Where WSCountry is the interface and WSCountryImpl the implementation class.
Hide
Permalink
Ryan Heaton added a comment -

Well I'll take another look for 1.26.

Show
Ryan Heaton added a comment - Well I'll take another look for 1.26.
Hide
Permalink
Rob Sonke added a comment -

Thanks, I'll try to have a look myself in the next weeks as well.

Show
Rob Sonke added a comment - Thanks, I'll try to have a look myself in the next weeks as well.
Hide
Permalink
Rob Sonke added a comment -

I was thinking again about this. Isnt it possible to give an origin type as param to findMapType? Which you can use for a check if you're going in circles.

Show
Rob Sonke added a comment - I was thinking again about this. Isnt it possible to give an origin type as param to findMapType? Which you can use for a check if you're going in circles.
Hide
Permalink
Ryan Heaton added a comment -

Oh yeah. That's a good idea. Some kind of mechanism to hack it.

Thanks. I'll try to take a look for 1.26.

Show
Ryan Heaton added a comment - Oh yeah. That's a good idea. Some kind of mechanism to hack it. Thanks. I'll try to take a look for 1.26.
Hide
Permalink
Ryan Heaton added a comment -

fixed at be6e1a8

Show
Ryan Heaton added a comment - fixed at be6e1a8
Hide
Permalink
Rob Sonke added a comment -

Thanks, great. I'll have a look soon.

Show
Rob Sonke added a comment - Thanks, great. I'll have a look soon.
Hide
Permalink
Ryan Heaton added a comment -

Closed with the release of 1.26.

Show
Ryan Heaton added a comment - Closed with the release of 1.26.

People

  • Assignee:
    Ryan Heaton
    Reporter:
    Ryan Heaton
Vote (0)
Watch (2)

Dates

  • Created:
    Updated:
    Resolved: