Runtime
After running the binding compiler the modified classes are ready to be used.
You'll need to include the /lib/jibx-run.jar and
/lib/xpp3.jar jar files from the distribution in your application
classpath (but not the /lib/bcel.jar and /lib/jibx-bind.jar
files, which are only used by the binding compiler). You'll also need to add a
little code at whatever point you want to marshal or unmarshal a document. This uses the
org.jibx.runtime.BindingDirectory class that's included in the
JiBX runtime jar, along with a binding factory class that JiBX generates in the same package
as your code (or in a package you specify in your binding). You don't need to worry about
the details of getting at this generated class, though. Instead, you can access it by passing
any of the classes defined by a global mapping (one that's a child of the root binding
element) in your binding to the BindingDirectory. The code is simple:
IBindingFactory bfact =
BindingDirectory.getFactory(Customer.class);
Here Customer is the name of a class with a global mapping in the binding. If
you've compiled more than one binding into the code, you'll also need to pass the name
of the binding you want to use. Or if you prefer, you can use a different lookup method in the
BindingDirectory class which finds the binding factory using a binding
name and Java package name.
The org.jibx.runtime.IBindingFactory interface that gets
returned provides methods to construct marshalling and unmarshalling contexts,
which in turn allow you to do the actual marshal and unmarshal operations.
Here's an unmarshal example:
IUnmarshallingContext uctx = bfact.createUnmarshallingContext();
Object obj = uctx.unmarshalDocument
(new FileInputStream("filename.xml"), null);
This is just one of several variations of an unmarshal call, in this case to
unmarshal an XML document in the file filename.xml. You can pass a
reader instead of a stream as the source of the document data if you want, and
can also specify an encoding for the document - see the JavaDocs for details. The
returned object will be an instance of one of your classes defined with a global
mapping in the binding - you can either check the type with instanceof
or cast directly to your object type, if you know what it is.
Marshalling is just as easy. Here's an example:
IMarshallingContext mctx = bfact.createMarshallingContext();
mctx.marshalDocument(obj, "UTF-8", null,
new FileOutputStream("filename.xml"));
As with the unmarshal example, this is just one of several variations that
can be used for the marshal call. This marshals the object to an XML document
written to the file filename.xml, with UTF-8 character encoding
(the most common choice for XML). The code as shown writes the output document
with no extra whitespace; you can use the setIndent() method of the
context to add whitespace for readability, if you wish. You can pass a writer
instead of a stream, as well as some other variations - see the following
section for details on character encoding usages, and the JavaDocs for the
different types of marshal calls. The object to be
marshalled must always be an instance of a class defined with a global
mapping in the binding.
Normally XML documents are marshalled as complete units, and to facilitate this
usage the marshalDocument() variations all close the output stream or
writer once the end tag for the XML document has been written. In special cases
you may need to write an XML document without closing the output. You can also do
this with JiBX marshalling, though it's a little more involved than the common
case. Here's the code:
IMarshallingContext mctx = bfact.createMarshallingContext();
mctx.setOutput(new FileOutputStream("filename.xml"), null);
((IMarshallable)obj).marshal(mctx);
mctx.getXmlWriter().flush();
In the above code, the cast to IMarshallable uses an interface added
to each mapped class by the JiBX binding compiler. The call to the marshal()
method of the interface does the actual marshalling of XML, and the last line makes
sure that all the output has been written from internal buffers.
Viewing binding information
The 1.2 release of JiBX adds a PrintInfo tool for accessing a binding factory directly and
viewing compiled binding information. This tool is the default execution target for the
jibx-run.jar, allowing you to invoke it very easily from the command line:
java -jar lib/jibx-run.jar
When executed in this manner it just prints out the JiBX runtime version information. If
you add a -? argument to the command line it'll show usage information for the
tool, including other command line parameters you can use to direct it to a binding factory
which can be loaded from the Java classpath. Here's an example of running PrintInfo for the
starter example, assuming you're doing this from a console open to the examples/starter
directory and you've compiled the classes and the binding for this example (in a single
line, shown split here only for formatting):
java -cp bin:../../lib/jibx-run.jar org.jibx.runtime.PrintInfo
-c org.jibx.starter.Customer
The output tells you the version of the binding compiler used to compile the binding,
the namespaces used by the binding, and the mapping definitions included in the binding.
Character encodings
The Java core classes provides java.io.Writer implementations
that support a wide variety of character encodings. These can wrap simple output
streams, and handle the conversions from Java characters to bytes as appropriate
for the particular encoding used by the writer. This direct conversion of
characters to bytes is not sufficient for use with XML, though. The problem is
that character encodings may not allow for all the legal XML character codes.
Any XML characters that are not supported by the output encoding need to be
converted to character references for output (see the XML recommendation
for details).
Because of this need to use character references, JiBX supports the use of
character escapers for output conversion handling. For the widely-used
UTF-8 and ISO-8859-1 (Western European character set) encodings implementations
are included that handle both stream and writer output formats automatically
(though using a stream will provide the best performance). The US-ASCII 7-bit
format is also handled automatically, though in this case a
java.io.Writer is always used internally.
Other character encodings can be used for output if you supply an appropriate
org.jibx.runtime.ICharacterEscaper instance to be used with the
output stream or writer. Depending on the encoding, you may even be able to use
one of the existing character escaper implementation classes from the
org.jibx.runtime.impl package directly, or at least base your own
escaper code on one of those implementation classes. For most users the standard
encodings are all that will ever be needed, but this approach allows other
alternatives to be used when necessary for special requirements.
StAX support
Support for StAX parser input and StAX writer output has been supported since
JiBX version 1.1. To select which parser you want to use for input, you can set
the system property org.jibx.runtime.impl.parser to the value
org.jibx.runtime.impl.XMLPullReaderFactory to select the XPP3 XMLPull
parser, or the value org.jibx.runtime.impl.StAXReaderFactory to select
the StAX parser. By default, JiBX uses whichever parser implementation it finds
at runtime, with preference given to the XPP3 XMLPull parser if both XPP3 and a
StAX parser are present. If you never want to use the XPP3 parser, simply remove
the /lib/xpp3.jar file from your JiBX installation and runtime
classpath.
Some StAX parsers support schema validation of input. If you wish to make use
of this feature you'll need to substitute an appropriate StAX parser
implementation for the /lib/wstx-asl.jar StAX parser included in the JiBX
distribution, and take whatever action is needed to enable schema validation.
To use StAX output (to a javax.xml.stream.XMLStreamWriter
instance) you'll need to set the XML writer directly on the JiBX marshalling
context, using the special JiBX
org.jibx.runtime.impl.StAXWriter class which wraps the target
javax.xml.stream.XMLStreamWriter instance. Here's a sample of code
for this purpose:
// marshal root object back out to document in memory
IMarshallingContext mctx = bfact.createMarshallingContext();
ByteArrayOutputStream bos = new ByteArrayOutputStream();
try {
XMLOutputFactory ofact = XMLOutputFactory.newInstance();
XMLStreamWriter wrtr = ofact.createXMLStreamWriter(bos, enc);
mctx.setXmlWriter(new StAXWriter(bfact.getNamespaces(), wrtr));
mctx.marshalDocument(obj);
} catch (XMLStreamException e) {
throw new JiBXException("Error creating writer", e);
}