Alexandr Klimov created XMLSCHEMA-30:
----------------------------------------

             Summary: Each type from W3C base schema 
(http://www.w3.org/2001/XMLSchema) is equals to each other
                 Key: XMLSCHEMA-30
                 URL: https://issues.apache.org/jira/browse/XMLSCHEMA-30
             Project: XmlSchema
          Issue Type: Bug
    Affects Versions: 2.0.2, 2.0.3, 2.0.4
         Environment: Oracle JDK-1.6
Oracle JDK-1.7
            Reporter: Alexandr Klimov
            Priority: Critical


h2. MAIN DESCIPRTION
At this time, all simple types of the base xml schema _( W3C Schema has 
namesapce http://www.w3.org/2001/XMLSchema )_ *always equlals to each other* 
from point of view *Object@equals* method, but they have different hash codes.
I've attched a simple sample_proj.zip project to be more clean.


Such situation *may lead to wrong behaviour* of some Set or Map implementations.
F.e. ImmutableSet of Guava project doesn't copy all items from 
java.util.HashSet over creation operation _( ImmutableSet#copyOf )_

*Moreover*, such behaviuor isn't in accordance with [Oracle 
rules|http://docs.oracle.com/javase/7/docs/api/java/lang/Object.html#hashCode()]

h2. DETAILS
Comparison of a XmlSchemaType (both simple and complex) is performed by 
*XmlSchemaObject#equals* method.
This method takes in account folowing properties of a XmlSchemaObject:
* lineNumber
* linePosition
* sourceURI

But all base W3C xsd types _(namespace http://www.w3.org/2001/XMLSchema )_ has 
one peculiarity: all the above properties is not set throughout base W3C schema 
creation.

*XmlSchemaCollection#addSimpletype* method is responsible for creation all 
these base schema simple types.

In result, all these "synthatic" types may be distinguished by they QNames only.

h2. SIGGESTION
I suggest to override *equals* method on *XmlSchemaType* class with taking in 
account its *namedDelegate* field comparison result.
 




--
This message was sent by Atlassian JIRA
(v6.1.5#6160)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to