Fix #1083

release-notes/CREDITS

@@ -397,3 +397,7 @@ Shumpei Akai (flexfrank@github)
 Hugo Wood (hgwood@github)
  * Contributed #1010: Support for array delegator
  (2.7.0)
+
+Julian Hyde (julianhyde@github)
+ * Reported #1083: Field in base class is not recognized, when using `@JsonType.defaultImpl`
+ (2.7.1)

release-notes/VERSION

@@ -7,6 +7,8 @@ Project: jackson-databind
 2.7.1 (not yet released)
 
 #1079: Add back `TypeFactory.constructType(Type, Class)` as "deprecated" in 2.7.1
+#1083: Field in base class is not recognized, when using `@JsonType.defaultImpl`
+ (reported by Julian H)
 
 2.7.0 (10-Jan-2016)
 

src/main/java/com/fasterxml/jackson/databind/jsontype/impl/TypeDeserializerBase.java

@@ -74,8 +74,8 @@ protected TypeDeserializerBase(JavaType baseType, TypeIdResolver idRes,
  // 22-Dec-2015, tatu: as per [databind#1055], avoid NPE
  _typePropertyName = (typePropertyName == null) ? "" : typePropertyName;
  _typeIdVisible = typeIdVisible;
- // defaults are fine, although concurrency of 4 bit more frugal than 16:
- _deserializers = new ConcurrentHashMap<String, JsonDeserializer<Object>>(16, 0.75f, 4);
+ // defaults are fine, although shouldn't need much concurrency
+ _deserializers = new ConcurrentHashMap<String, JsonDeserializer<Object>>(16, 0.75f, 2);
  if (defaultImpl == null) {
  _defaultImpl = null;
  } else {
@@ -84,6 +84,8 @@ protected TypeDeserializerBase(JavaType baseType, TypeIdResolver idRes,
  * seldom (if ever) base types, may be ok.
  */
  // 01-Nov-2015, tatu: Actually this is still exactly wrong. Should fix.
+ // 15-Jan-2016, tatu: ... as witnessed by [databind#1083], patched, but
+ // fundamentally this call can't be made to work for all cases
  _defaultImpl = baseType.forcedNarrowBy(defaultImpl);
  }
  _property = null;

src/main/java/com/fasterxml/jackson/databind/type/SimpleType.java

@@ -123,7 +123,16 @@ protected JavaType _narrow(Class<?> subclass)
  return this;
  }
  // Should we check that there is a sub-class relationship?
- return new SimpleType(subclass, _bindings, _superClass, _superInterfaces,
+ // 15-Jan-2016, tatu: Almost yes, but there are some complications with
+ // placeholder values, so no.
+ /*
+ if (!_class.isAssignableFrom(subclass)) {
+ throw new IllegalArgumentException("Class "+subclass.getName()+" not sub-type of "
+ +_class.getName());
+ }
+ */
+ // 15-Jan-2015, tatu: Not correct; should really re-resolve...
+ return new SimpleType(subclass, _bindings, this, _superInterfaces,
  _valueHandler, _typeHandler, _asStatic);
  }