[ 
https://issues.apache.org/jira/browse/GROOVY-7874?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Mike Cargal updated GROOVY-7874:
--------------------------------
    Description: 
The following code segment demonstrates parsing a schema and then serializing 
with xmlUtil.  xmlUtil, in it's attempt to discard necessary namespaces, is 
discarding namespaces that are actually needed.

Save the original schema and the referenced schema in eclipse, and there will 
be no errors.  Save the output schema (as MyRevSchema.xsd for example) and 
you'll have errors as a result of the dropped namespaces.

{code}
import groovy.xml.*

scSrc = '''<?xml version="1.0" encoding="UTF-8"?>
<xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema";
    attributeFormDefault="unqualified" elementFormDefault="qualified"
    targetNamespace="urn:cargal.net/myNS" xmlns="urn:cargal.net/myNS"
    xmlns:myns="urn:cargal.net/myNS">
    <xsd:include schemaLocation="MyRefSchema.xsd" />
    <xsd:complexType name="myCT2">
        <xsd:complexContent>
            <xsd:extension base="myns:myCT">
                <xsd:sequence>
                    <xsd:element name="element1" />
                </xsd:sequence>
            </xsd:extension>
        </xsd:complexContent>
    </xsd:complexType>
    <xsd:element name="test" type="myns:myCT" />
    <xsd:element name="test2" substitutionGroup="subG" type="myCT2" />
</xsd:schema>
'''

schema = new XmlSlurper().parseText(scSrc)
println schema.lookupNamespace('')
println schema.lookupNamespace('myns')
println new XmlUtil().serialize(schema)
{code}

results:
{noformat}
schema.@attributeFormDefault=unqualified
schema.@xmlns=
<?xml version="1.0" encoding="UTF-8"?><xsd:schema 
xmlns:xsd="http://www.w3.org/2001/XMLSchema"; attributeFormDefault="unqualified" 
elementFormDefault="qualified" targetNamespace="urn:cargal.net/myNS">
  <xsd:include schemaLocation="MyRefSchema.xsd"/>
  <xsd:complexType name="myCT2">
    <xsd:complexContent>
      <xsd:extension base="myns:myCT">
        <xsd:sequence>
          <xsd:element name="element1"/>
        </xsd:sequence>
      </xsd:extension>
    </xsd:complexContent>
  </xsd:complexType>
  <xsd:element name="test" type="myns:myCT"/>
  <xsd:element name="test2" substitutionGroup="subG" type="myCT2"/>
</xsd:schema>
{noformat}

For reference:  MyRefSchema.xsd
{noformat}
<?xml version="1.0" encoding="UTF-8"?>
<xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema";
        attributeFormDefault="unqualified" elementFormDefault="qualified"
        targetNamespace="urn:cargal.net/myNS" xmlns="urn:cargal.net/myNS"
        xmlns:myns="urn:cargal.net/myNS">
        <xsd:complexType name="myCT">
                <xsd:sequence>
                        <xsd:element name="element1" />
                </xsd:sequence>
        </xsd:complexType>
        <xsd:element name="subG" type="myCT" />
</xsd:schema>
{noformat}

  was:
When parsing an XML document with XmlSlurper, some attributes are being dropped.
{code}
import groovy.xml.*

scSrc = '''<?xml version="1.0" encoding="UTF8"?><xsd:schema 
xmlns:xsd="http://www.w3.org/2001/XMLSchema"; attributeFormDefault="unqualified" 
elementFormDefault="qualified" targetNamespace="urn:cargal.net/myNS" 
xmlns="urn:cargal.net/myNS">
    <xsd:element name="test" type="xsd:string"/>
</xsd:schema>
'''

schema = new XmlSlurper().parseText(scSrc)
println "schema.@attributeFormDefault=${schema.@attributeFormDefault}"
println "schema.@xmlns=${schema.@xmlns}"
println new XmlUtil().serialize(schema)
{code}

results:
{noformat}
schema.@attributeFormDefault=unqualified
schema.@xmlns=
<?xml version="1.0" encoding="UTF-8"?><xsd:schema 
xmlns:xsd="http://www.w3.org/2001/XMLSchema"; attributeFormDefault="unqualified" 
elementFormDefault="qualified" targetNamespace="urn:cargal.net/myNS">
  <xsd:element name="test" type="xsd:string"/>
</xsd:schema>
{noformat}


        Summary: When serializing a document with xmlUtil, namespaces that are 
actually necessary, are being dropped as "unused"  (was: XmlSlurper is dropping 
root node attributes)

I've revised the description to reflect the problem being in xmlUtil, and that 
it is too aggressive in removing namespaces it believes to be unnecessary, with 
an example of how the  resulting schema is no longer valid (while the input 
schema, without the dropped namespaces, was valid)

> When serializing a document with xmlUtil, namespaces that are actually 
> necessary, are being dropped as "unused"
> ---------------------------------------------------------------------------------------------------------------
>
>                 Key: GROOVY-7874
>                 URL: https://issues.apache.org/jira/browse/GROOVY-7874
>             Project: Groovy
>          Issue Type: Bug
>          Components: XML Processing
>    Affects Versions: 2.4.3, 2.4.6, 2.4.7
>         Environment: MAC OS X, and Windows
>            Reporter: Mike Cargal
>            Priority: Minor
>
> The following code segment demonstrates parsing a schema and then serializing 
> with xmlUtil.  xmlUtil, in it's attempt to discard necessary namespaces, is 
> discarding namespaces that are actually needed.
> Save the original schema and the referenced schema in eclipse, and there will 
> be no errors.  Save the output schema (as MyRevSchema.xsd for example) and 
> you'll have errors as a result of the dropped namespaces.
> {code}
> import groovy.xml.*
> scSrc = '''<?xml version="1.0" encoding="UTF-8"?>
> <xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema";
>     attributeFormDefault="unqualified" elementFormDefault="qualified"
>     targetNamespace="urn:cargal.net/myNS" xmlns="urn:cargal.net/myNS"
>     xmlns:myns="urn:cargal.net/myNS">
>     <xsd:include schemaLocation="MyRefSchema.xsd" />
>     <xsd:complexType name="myCT2">
>         <xsd:complexContent>
>             <xsd:extension base="myns:myCT">
>                 <xsd:sequence>
>                     <xsd:element name="element1" />
>                 </xsd:sequence>
>             </xsd:extension>
>         </xsd:complexContent>
>     </xsd:complexType>
>     <xsd:element name="test" type="myns:myCT" />
>     <xsd:element name="test2" substitutionGroup="subG" type="myCT2" />
> </xsd:schema>
> '''
> schema = new XmlSlurper().parseText(scSrc)
> println schema.lookupNamespace('')
> println schema.lookupNamespace('myns')
> println new XmlUtil().serialize(schema)
> {code}
> results:
> {noformat}
> schema.@attributeFormDefault=unqualified
> schema.@xmlns=
> <?xml version="1.0" encoding="UTF-8"?><xsd:schema 
> xmlns:xsd="http://www.w3.org/2001/XMLSchema"; 
> attributeFormDefault="unqualified" elementFormDefault="qualified" 
> targetNamespace="urn:cargal.net/myNS">
>   <xsd:include schemaLocation="MyRefSchema.xsd"/>
>   <xsd:complexType name="myCT2">
>     <xsd:complexContent>
>       <xsd:extension base="myns:myCT">
>         <xsd:sequence>
>           <xsd:element name="element1"/>
>         </xsd:sequence>
>       </xsd:extension>
>     </xsd:complexContent>
>   </xsd:complexType>
>   <xsd:element name="test" type="myns:myCT"/>
>   <xsd:element name="test2" substitutionGroup="subG" type="myCT2"/>
> </xsd:schema>
> {noformat}
> For reference:  MyRefSchema.xsd
> {noformat}
> <?xml version="1.0" encoding="UTF-8"?>
> <xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema";
>       attributeFormDefault="unqualified" elementFormDefault="qualified"
>       targetNamespace="urn:cargal.net/myNS" xmlns="urn:cargal.net/myNS"
>       xmlns:myns="urn:cargal.net/myNS">
>       <xsd:complexType name="myCT">
>               <xsd:sequence>
>                       <xsd:element name="element1" />
>               </xsd:sequence>
>       </xsd:complexType>
>       <xsd:element name="subG" type="myCT" />
> </xsd:schema>
> {noformat}



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to