JAXB behaviour when unmarshalling documents with an encoding declaration

Question

The following is against JWSDP-1.5, 1.4.2_07-b05 on XP Pro SP 2...
When I unmarshall a document using JAXB 1.0, and obtain Strings from the resulting objects, I would expect the Strings to be encoded with the character encoding specified by the text declaration. Is this a correct assumption?
Unfortunately, from the sample code (given below) I am not experiencing this behaviour. I've included the code, sample output, and the RELAX NG schema used.
Any help very welcome.
ian
import java.io.ByteArrayInputStream;
import java.io.ByteArrayOutputStream;
import java.io.FileNotFoundException;
import java.io.FileOutputStream;
import java.util.List;
import javax.xml.bind.JAXBContext;
import javax.xml.bind.JAXBException;
import javax.xml.bind.Marshaller;
import org.xml.sax.InputSource;
import com.chellomedia.transcoding.Category;
import com.chellomedia.transcoding.ObjectFactory;
import com.chellomedia.transcoding.Schedule;
public class TranscoderExemplar {
     private static final String CATEGORY_NAME = "\u00bfMa\u00f1ana?";
     public static void main(String[] args) throws Exception {
          try {
               ByteArrayOutputStream os = generateSampleDocument("US-ASCII");
               inspectDocument("US-ASCII", os);
               os = generateSampleDocument("ISO-8859-1");
               inspectDocument("ISO-8859-1", os);
               os = generateSampleDocument("UTF-8");
               inspectDocument("UTF-8", os);
          } catch (FileNotFoundException e) {
               e.printStackTrace();
          } catch (JAXBException e) {
               e.printStackTrace();
          } catch (Exception e) {
               e.printStackTrace();
     private static void inspectDocument(String encoding, ByteArrayOutputStream os) throws Exception {
          byte[] b = os.toByteArray();
          System.out.println(encoding + " document byte length: " + b.length);
          InputSource is = new InputSource(new ByteArrayInputStream(b));
          is.setEncoding(encoding);
          JAXBContext context = JAXBContext.newInstance("com.chellomedia.transcoding");
          Schedule schedule = (Schedule)context.createUnmarshaller().unmarshal(is);
          context.createValidator().validate(schedule);
          List categories = schedule.getCategories();
          for (int i = 0, n = categories.size(); i < n; i++) {
               Category c = (Category)categories.get(i);
               String name = c.getName();
               System.out.println(encoding + " name string length: " + name.length());
               System.out.println(encoding + " name byte length: " + name.getBytes().length);
     private static ByteArrayOutputStream generateSampleDocument(String encoding) throws JAXBException,
          FileNotFoundException
        ObjectFactory of = new ObjectFactory();
          Schedule schedule = of.createSchedule();
          List categories = schedule.getCategories();
          Category category = of.createCategory();
          category.setId(1);
          category.setName(CATEGORY_NAME);
          categories.add(category);
          Marshaller m = of.createMarshaller();
          m.setProperty("jaxb.encoding", encoding);
        m.setProperty(Marshaller.JAXB_FORMATTED_OUTPUT, Boolean.TRUE);
          ByteArrayOutputStream os = new ByteArrayOutputStream();
          m.marshal(schedule, os);
          FileOutputStream fos = new FileOutputStream(encoding + "-representation.xml");
          m.marshal(schedule, fos);
          return os;
}...and the schema
<grammar xmlns="http://relaxng.org/ns/structure/1.0"
         datatypeLibrary="http://www.w3.org/2001/XMLSchema-datatypes"
         xmlns:jaxb="http://java.sun.com/xml/ns/jaxb"
         xmlns:xjc="http://java.sun.com/xml/ns/jaxb/xjc"
         jaxb:extensionBindingPrefixes="xjc"
         jaxb:version="1.0">
  <jaxb:schemaBindings>
    <jaxb:package name="com.chellomedia.transcoding"/>
  </jaxb:schemaBindings>
  <start>
    <ref name="Schedule"/>
  </start>
  <define name="Schedule">
    <element name="schedule">
      <ref name="Categories"/>
      <zeroOrMore>
        <ref name="Event"/>
      </zeroOrMore>
    </element>
  </define>
  <define name="Categories">
    <element name="categories">
      <zeroOrMore>
        <ref name="Category"/>
      </zeroOrMore>
    </element>
  </define>
  <define name="Category">
    <element name="category">
      <attribute name="id"><data type="int"/></attribute>
      <attribute name="name"/>
      <zeroOrMore>
        <ref name="Subcategory"/>
      </zeroOrMore>
    </element>
  </define>
  <define name="Subcategory">
    <element name="subcategory">
      <attribute name="id"/>
      <attribute name="name"/>
    </element>
  </define>
  <define name="Event">
    <element name="event">
      <attribute name="id"/>
      <attribute name="title"/>
      <attribute name="description"/>
      <attribute name="category"/>
      <attribute name="subcategory"/>
    </element>
  </define>
</grammar>...with sample output
US-ASCII document byte length: 171
US-ASCII name string length: 8
US-ASCII name byte length: 8
ISO-8859-1 document byte length: 163
ISO-8859-1 name string length: 8
ISO-8859-1 name byte length: 8
UTF-8 document byte length: 160
UTF-8 name string length: 8
UTF-8 name byte length: 8

Answer 1

The following is against JWSDP-1.5, 1.4.2_07-b05 on XP Pro SP 2...
When I unmarshall a document using JAXB 1.0, and obtain Strings from the resulting objects, I would expect the Strings to be encoded with the character encoding specified by the text declaration. Is this a correct assumption?
Unfortunately, from the sample code (given below) I am not experiencing this behaviour. I've included the code, sample output, and the RELAX NG schema used.
Any help very welcome.
ian
import java.io.ByteArrayInputStream;
import java.io.ByteArrayOutputStream;
import java.io.FileNotFoundException;
import java.io.FileOutputStream;
import java.util.List;
import javax.xml.bind.JAXBContext;
import javax.xml.bind.JAXBException;
import javax.xml.bind.Marshaller;
import org.xml.sax.InputSource;
import com.chellomedia.transcoding.Category;
import com.chellomedia.transcoding.ObjectFactory;
import com.chellomedia.transcoding.Schedule;
public class TranscoderExemplar {
     private static final String CATEGORY_NAME = "\u00bfMa\u00f1ana?";
     public static void main(String[] args) throws Exception {
          try {
               ByteArrayOutputStream os = generateSampleDocument("US-ASCII");
               inspectDocument("US-ASCII", os);
               os = generateSampleDocument("ISO-8859-1");
               inspectDocument("ISO-8859-1", os);
               os = generateSampleDocument("UTF-8");
               inspectDocument("UTF-8", os);
          } catch (FileNotFoundException e) {
               e.printStackTrace();
          } catch (JAXBException e) {
               e.printStackTrace();
          } catch (Exception e) {
               e.printStackTrace();
     private static void inspectDocument(String encoding, ByteArrayOutputStream os) throws Exception {
          byte[] b = os.toByteArray();
          System.out.println(encoding + " document byte length: " + b.length);
          InputSource is = new InputSource(new ByteArrayInputStream(b));
          is.setEncoding(encoding);
          JAXBContext context = JAXBContext.newInstance("com.chellomedia.transcoding");
          Schedule schedule = (Schedule)context.createUnmarshaller().unmarshal(is);
          context.createValidator().validate(schedule);
          List categories = schedule.getCategories();
          for (int i = 0, n = categories.size(); i < n; i++) {
               Category c = (Category)categories.get(i);
               String name = c.getName();
               System.out.println(encoding + " name string length: " + name.length());
               System.out.println(encoding + " name byte length: " + name.getBytes().length);
     private static ByteArrayOutputStream generateSampleDocument(String encoding) throws JAXBException,
          FileNotFoundException
        ObjectFactory of = new ObjectFactory();
          Schedule schedule = of.createSchedule();
          List categories = schedule.getCategories();
          Category category = of.createCategory();
          category.setId(1);
          category.setName(CATEGORY_NAME);
          categories.add(category);
          Marshaller m = of.createMarshaller();
          m.setProperty("jaxb.encoding", encoding);
        m.setProperty(Marshaller.JAXB_FORMATTED_OUTPUT, Boolean.TRUE);
          ByteArrayOutputStream os = new ByteArrayOutputStream();
          m.marshal(schedule, os);
          FileOutputStream fos = new FileOutputStream(encoding + "-representation.xml");
          m.marshal(schedule, fos);
          return os;
}...and the schema
<grammar xmlns="http://relaxng.org/ns/structure/1.0"
         datatypeLibrary="http://www.w3.org/2001/XMLSchema-datatypes"
         xmlns:jaxb="http://java.sun.com/xml/ns/jaxb"
         xmlns:xjc="http://java.sun.com/xml/ns/jaxb/xjc"
         jaxb:extensionBindingPrefixes="xjc"
         jaxb:version="1.0">
  <jaxb:schemaBindings>
    <jaxb:package name="com.chellomedia.transcoding"/>
  </jaxb:schemaBindings>
  <start>
    <ref name="Schedule"/>
  </start>
  <define name="Schedule">
    <element name="schedule">
      <ref name="Categories"/>
      <zeroOrMore>
        <ref name="Event"/>
      </zeroOrMore>
    </element>
  </define>
  <define name="Categories">
    <element name="categories">
      <zeroOrMore>
        <ref name="Category"/>
      </zeroOrMore>
    </element>
  </define>
  <define name="Category">
    <element name="category">
      <attribute name="id"><data type="int"/></attribute>
      <attribute name="name"/>
      <zeroOrMore>
        <ref name="Subcategory"/>
      </zeroOrMore>
    </element>
  </define>
  <define name="Subcategory">
    <element name="subcategory">
      <attribute name="id"/>
      <attribute name="name"/>
    </element>
  </define>
  <define name="Event">
    <element name="event">
      <attribute name="id"/>
      <attribute name="title"/>
      <attribute name="description"/>
      <attribute name="category"/>
      <attribute name="subcategory"/>
    </element>
  </define>
</grammar>...with sample output
US-ASCII document byte length: 171
US-ASCII name string length: 8
US-ASCII name byte length: 8
ISO-8859-1 document byte length: 163
ISO-8859-1 name string length: 8
ISO-8859-1 name byte length: 8
UTF-8 document byte length: 160
UTF-8 name string length: 8
UTF-8 name byte length: 8