<?xml version="1.0" encoding="iso-8859-1"?>

<html xmlns="http://www.w3.org/1999/xhtml">

<meta http-equiv="content-type" content="text/html; charset=iso-8859-1" />

<title>EPICS pvDataJava</title>

<link rel="stylesheet" type="text/css" href="http://epics-pvdata.sourceforge.net/base.css" />

<link rel="stylesheet" type="text/css" href="http://epics-pvdata.sourceforge.net/epicsv4.css" />

<style type="text/css">

/*<![CDATA[*/

.about { margin-left: 3em; margin-right: 3em; font-size: .83em}

table { margin-left: auto; margin-right: auto }

.diagram { text-align: center; margin: 2.5em 0 }

span.opt { color: grey }

span.nterm { font-style:italic }

span.term { font-family:courier }

span.user { font-family:courier }

span.user:before { content:"<" }

span.user:after { content:">" }

.nonnorm { font-style:italic }

p.ed.priv { display: inline; }

span.ed.priv { display: inline; }

hr { border: none; background-color: #ccc; color: #ccc; height: 1px }

/*]]>*/</style>

<!-- Script that generates the Table of Contents -->

<script type="text/javascript" src="http://epics-pvdata.sourceforge.net/script/tocgen.js"></script>

<div class="head">

<h1>EPICS pvDataJava</h1>

<h2 class="nocount">Release 5.1-DEV - 2017.03.29</h2>

<dl>

<dt>Editors:</dt>

<dd>Marty Kraimer, BNL</dd>

<dd>Dave Hickin, Diamond</dd>

</dl>

<p class="copyright">This product is made available subject to acceptance of the <a

href="http://epics-pvdata.sourceforge.net/LICENSE.html">EPICS open source

license.</a></p>

<h2 class="nocount">Abstract</h2>

<p>pvDataJava is a computer software package for the efficient

storage, access, and communication, of structured data. It is specifically the

Java implementation of pvData, which is one part of the set of related products in the EPICS

V4 control system programming environment:<br>

<a href="http://epics-pvdata.sourceforge.net/relatedDocumentsV4.html">relatedDocumentsV4.html</a>

<h2 class="nocount">Status of this Document</h2>

<p>This is the 29-March-2017 version of the Java implementation of pvData.

<p>RELEASE_NOTES.md provides changes since the last release.

TODO.md describes things to do before the next release.

<div id="toc">

<h2 class="nocount" style="page-break-before: always">Table of Contents</h2>

<div id="contents" class="contents">

<h2>Introduction</h2>

<p>pvData is one of a set of related projects. It describes and implements

data that the other projects support. Thus it is not useful by itself but

understanding pvData is required in order to understand the other projects. The

reader should also become familiar with project pvAccess, which is

located via the same sourceforge site as this project.</p>

<a href="http://epics-pvdata.sourceforge.net/informative/developerGuide/developerGuide.html">developerGuide</a>

briefly describes a set of application programming interfaces (APIs) for EPICS V4.

It should be read before reading this manual.

<p>The Java and C++ implementation of pvData implement the same data model but

differ in implementation because of the differences between Java and C++.</p>

<p>pvData is one of a set of related packages in the EPICS V4 core software. It

describes and implements the data management system to which the the other projects

interface.</p>

<p>pvData (Process Variable Data) defines and implements an efficient way to

store, access, and communicate memory resident data structures.</p>

<dt>definition</dt>

<dd>Package org.epics.pvdata.pv has Java interface definitions that define

pvData. Although defined with Java syntax they also document what is

required for implementation in other languages such as C++.</dd>

<dt>implementation</dt>

<dd>The set of packages provided by this project provide a complete Java

implementation of pvData. Project pvAccess is the network support for

pvData, i.e. it provides a channel access client and server that fully

support pvData.</dd>

<dt>efficient</dt>

<dd>Small memory footprint, low cpu overhead, and concise code base.</dd>

<dt>data storage</dt>

<dd>pvData defines separate introspection and data interfaces. The

introspection interfaces provide access to immutable objects, which

allows introspection instances to be freely shared. The introspection

interface for a process variable can be accessed without requiring access

to the data.</dd>

<dt>data access</dt>

<dd>Client code can access pvData via the introspection and data

interfaces. For "well known" data, e.g. image data, specialized interfaces

can be provided without requiring any changes to the core software.</dd>

<dt>data transfer</dt>

<dd>The separation of introspection and data interfaces allows for

efficient network data transfer. At connection time introspection

information can be passed from server to client. Each side can create a

data instance. The data is transferred between these instances. The data

in the network buffers does not have to be self describing since each

side has the introspection information.</dd>

<dt>memory resident</dt>

<dd>pvData only defines memory resident data.</dd>

<dt>structured data</dt>

<dd>pvData defines data as structures with subfields

where each each subfield has a name and type.

The types are defined next.</dd>

<h2 >Interface Definitions</h2>

<p>This section gives a <b>brief</b> description of the pvData introspection and data interfaces.

In this section the methods are shown but not described. See org.epics.pvdata.pv below for a

description of each method.</p>

<h3>Types</h3>

<p>The following are the type definitions:</p>

<pre>enum Type {

scalar,

scalarArray,

structure,

structureArray,

union_t,

unionArray;

}</pre>

<p>where</p>

<dt>scalar</dt>

<dd>A field that has data with one of the <code>ScalarType</code>s.</dd>

<dt>scalarArray</dt>

<dd>A field that is an array where each element is the same <code>ScalarType</code>.</dd>

<dt>structure</dt>

<dd>A field which has named subfields.</dd>

<dt>structureArray</dt>

<dd>A field that is an array of structures where each element has the same

introspection interface, i.e. each element has the same structure.</dd>

<dt>union</dt>

<dd>A union stores a single subfield with a variable type.

A variant union allows the type to be any type.

A regular union allows only a fixed set of types

</dd>

<dt>unionArray</dt>

<dd>

A union array is an array of union fields where each element

has the same introspection interface.

Different elements can have different types but each must have

a type determined by the union interface.

</dd>

<pre>enum ScalarType {

pvBoolean,

pvByte,pvShort,pvInt,pvLong,

pvUByte,pvUShort,pvUInt,pvULong,

pvFloat,pvDouble,

pvString;

// The following are convenience methods

public boolean isInteger();

public boolean isUInteger();

public boolean isNumeric();

public boolean isPrimitive();

public static ScalarType getScalarType(String type);

public String toString();

}</pre>

<p>where</p>

<dt>boolean</dt>

<dd>true or false</dd>

<dt>byte</dt>

<dd>An 8 bit signed byte</dd>

<dt>short</dt>

<dd>16 bit signed integer</dd>

<dt>int</dt>

<dd>32 bit signed integer</dd>

<dt>long</dt>

<dd>64 bit signed integer</dd>

<dt>ubyte</dt>

<dd>An 8 bit unsigned byte</dd>

<dt>ushort</dt>

<dd>16 bit unsigned integer</dd>

<dt>uint</dt>

<dd>32 bit unsigned integer</dd>

<dt>ulong</dt>

<dd>64 bit unsigned integer</dd>

<dt>float</dt>

<dd>32 bit IEEE float</dd>

<dt>double</dt>

<dd>64 bit IEEE float</dd>

<dt>string</dt>

<dd>An immutable string. The Java implementation is String. For other

implementations the network representation must be the same as for Java.

Note that a string is treated like it is a scaler.</dd>

<p><b>NOTE:</b> Java does not support unsigned integers so the Java implementation

of each unsigned type is implemented as the corresponding signed type.

The only exception is the Convert facility. When it performs a widening

operation, e.g. from ubyte to short, it performs the correct conversion.</p>

<h3>Introspection Interfaces</h3>

<p><code>Field</code> is the base introspection interface. It has only an ID and a type.</p>

<pre>interface Field extends Serializable {

String getId();

Type getType();

void toString(StringBuilder buf);

void toString(StringBuilder buf,int indentLevel);

String toString();

}

interface Scalar extends Field {

ScalarType getScalarType();

}

interface ScalarArray extends Field {

ScalarType getElementType();

}

interface Structure extends Field {

Field getField(String fieldName);

Field getField(int fieldIndex);

&lt;T extends Field&gt; T getField(Class&lt;T&gt; c, String fieldName);

&lt;T extends Field&gt; T getField(Class&lt;T&gt; c, int fieldIndex);

Field[] getFields();

String[] getFieldNames();

int getFieldIndex(String fieldName);

String getFieldName(int fieldIndex);

}

interface StructureArray extends Field {

Structure getStructure();

}

public interface Union extends Field{

public static final String DEFAULT_ID = "union";

public static final String ANY_ID = "any";

Field getField(String fieldName);

Field getField(int fieldIndex);

&lt;T extends Field&gt; T getField(Class&lt;T&gt; c, String fieldName);

&lt;T extends Field&gt; T getField(Class&lt;T&gt; c, int fieldIndex);

Field[] getFields();

String[] getFieldNames();

int getFieldIndex(String fieldName);

String getFieldName(int fieldIndex);

boolean isVariant();

}

public interface UnionArray extends Field{

Union getUnion();

}

<p>The introspection interfaces provide access to immutable objects. This

allows introspection interfaces to be freely shared between data objects. For

example the introspection interface for a timeStamp, which is a structure

containing two fields, can be shared by every record that has a time stamp.</p>

<h3>Data Interfaces</h3>

<p><code>PVField</code> is the base interface for a data field:</p>

<pre>interface PVField extends Serializable {

String getFieldName();

String getFullName();

int getFieldOffset();

int getNextFieldOffset();

int getNumberFields();

boolean isImmutable();

void setImmutable();

Field getField();

PVStructure getParent();

void postPut();

void setPostHandler(PostHandler postHandler);

void toString(StringBuilder buf);

void toString(StringBuilder buf,int indentLevel);

String toString();

// The following should go away.

PVAuxInfo getPVAuxInfo();

}</pre>

<p>Each scalar type has an associated data interface: <code>PVBoolean</code>,

<code>PVByte</code>, <code>PVShort</code>, <code>PVInt</code>, <code>PVLong</code>,

<code>PVUByte</code>, <code>PVUShort</code>, <code>PVUInt</code>, <code>PVULong</code>,

<code>PVFloat</code>, <code>PVDouble</code> and <code>PVString</code>. Each has a <code>get</code> and a

<code>put</code> method. For example:</p>

<pre>interface PVDouble extends PVScalar{

double get();

void put(double value);

}</pre>

<p><code>PVArray</code> is the base class for arrays.</p>

<pre>interface PVArray extends PVField, SerializableArray {

Array getArray();

int getLength();

void setLength(int length);

int getCapacity();

void setCapacity(int length);

boolean isCapacityMutable();

void setCapacityMutable(boolean isMutable);

}

<p><code>PVScalarArray</code> is the base class for scalar arrays.</p>

<pre>interface PVScalarArray extends PVArray {

ScalarArray getScalarArray();

}</pre>

<p>For each scalar type an associated array data interface is defined. Each has

a <code>get</code> and <code>put</code> method. For example: </p>

<pre>public class DoubleArrayData {

public double[] data;

public int offset;

}

interface PVDoubleArray extends PVArray {

int get(int offset, int len, DoubleArrayData data);

int put(int offset,int len, double[] from, int fromOffset);

void shareData(double[] from);

}</pre>

<p><code>PVStructureArray</code> is the interface for an array of structures where each element

has the same introspection interface.</p>

<pre>public class StructureArrayData {

public PVStructure[] data;

public int offset;

}

interface PVStructureArray extends PVArray{

StructureArray getStructureArray();

int get(int offset, int length, StructureArrayData data);

int put(int offset,int length, PVStructure[] from, int fromOffset);

void shareData(PVStructure[] from);

}</pre>

<p><code>PVStructure</code> is the data interface for a structure.</p>

interface PVStructure extends PVField , BitSetSerializable{

Structure getStructure();

PVField[] getPVFields();

PVField getSubField(String fieldName);

PVField getSubField(int fieldOffset);

&lt;T extends PVField&gt; T getSubField(Class&lt;T&gt; c, String fieldName);

&lt;T extends PVField&gt; T getSubField(Class&lt;T&gt; c, int fieldOffset);

// The following are convenience methods

// Note that they are no longer needed because of above generic methods

PVBoolean getBooleanField(String fieldName);

PVByte getByteField(String fieldName);

PVShort getShortField(String fieldName);

PVInt getIntField(String fieldName);

PVLong getLongField(String fieldName);

PVFloat getFloatField(String fieldName);

PVDouble getDoubleField(String fieldName);

PVString getStringField(String fieldName);

PVScalarArray getScalarArrayField(String fieldName);

PVStructureArray getStructureArrayField(String fieldName);

PVStructure getStructureField(String fieldName);

PVArray getArrayField(String fieldName,ScalarType elementType);

PVUnion getUnionField(String fieldName);

PVUnionArray getUnionArrayField(String fieldName);

public boolean checkValid();

}</pre>

<h3>Introspection and Data creation</h3>

<p>The following interface creates introspection instances:</p>

<pre>public interface FieldCreate {

FieldBuilder createFieldBuilder();

Scalar createScalar(ScalarType scalarType);

BoundedString createBoundedString(int maxLength);

ScalarArray createScalarArray(ScalarType elementType);

ScalarArray createFixedScalarArray(ScalarType elementType, int size);

ScalarArray createBoundedScalarArray(ScalarType elementType, int bound);

StructureArray createStructureArray(Structure elementStructure);

UnionArray createUnionArray(Union elementUnion);

UnionArray createVariantUnionArray();

Structure createStructure(String[] fieldNames, Field[] field);

Structure createStructure(String id,String[] fieldNames, Field[] field);

Structure appendField(Structure structure,String fieldName, Field field);

Structure appendFields(Structure structure,String[] fieldNames, Field[] fields);

Structure createStructure(Structure structToClone);

Union createVariantUnion();

Union createUnion(String[] fieldNames, Field[] fields);

Union createUnion(String id, String[] fieldNames, Field[] fields);

Field deserialize(ByteBuffer buffer, DeserializableControl control);

}</pre>

<p>The following is a convenience class for creating introspection objects:</p>

public interface FieldBuilder

{

public FieldBuilder setId(String id);

public FieldBuilder add(String name, ScalarType scalarType);

public FieldBuilder addBoundedString(String name, int maxLength);

public FieldBuilder add(String name, Field field);

public FieldBuilder addArray(String name, ScalarType scalarType);

public FieldBuilder addFixedArray(String name, ScalarType scalarType, int size);

public FieldBuilder addBoundedArray(String name, ScalarType scalarType, int bound);

public FieldBuilder addArray(String name, Field element);

public Structure createStructure();

public Union createUnion();

FieldBuilder addNestedStructure(String name);

FieldBuilder addNestedUnion(String name);

FieldBuilder addNestedStructureArray(String name);

FieldBuilder addNestedUnionArray(String name);

FieldBuilder endNested();

}

<p>The following interface creates data instances:</p>

<pre>public interface PVDataCreate {

PVField createPVField(Field field);

PVField createPVField(PVField fieldToClone);

PVScalar createPVScalar(Scalar scalar);

PVScalar createPVScalar(ScalarType fieldType);

PVScalar createPVScalar(PVScalar scalarToClone);

&lt;T extends PVScalar, TA extends PVScalarArray&gt; T createPVScalar(

PVScalarType&lt;T, TA&gt; scalarType);

PVScalarArray createPVScalarArray(ScalarArray array);

PVScalarArray createPVScalarArray(ScalarType elementType);

PVScalarArray createPVScalarArray(PVScalarArray arrayToClone;

&lt;T extends PVScalar, TA extends PVScalarArray&gt; TA createPVScalarArray(

PVScalarType&lt;T, TA&gt; elementType);

PVStructureArray createPVStructureArray(StructureArray structureArray);

PVUnionArray createPVUnionArray(UnionArray unionArray);

PVUnionArray createPVVariantUnionArray();

PVStructure createPVStructure(Structure structure);

PVStructure createPVStructure(String[] fieldNames,Field[] fields);

PVStructure createPVStructure(PVStructure structToClone);

PVUnion createPVUnion(Union union);

PVUnion createPVVariantUnion();

PVUnion createPVUnion(PVUnion unionToClone);

PVStructureArray createPVStructureArray(Structure structure);

PVUnionArray createPVUnionArray(Union union);

PVField[] flattenPVStructure(PVStructure pvStructure);

}</pre>

<h3>pvData Conversion</h3>

<p>An interface named <code>Convert</code> provides all reasonable conversions to/from

pvData.

The convert utility is also used to implement the <code>toString</code> methods of <code>PVField</code>.

See <code>org.epics.pvdata.pv.Convert</code> for details.</p>

<h2 >Package Summary</h2>

<p>This document describes everything via Java definitions. The initial

implementation is in Java but the functionality is also implemented in C++ (See pvDataCPP for details).</p>

<p>pvData is distributed as a sourceforge mercurial project named pvDataJava.

It consists of the following Java packages:</p>

<h3>org.epics.pvdata.pv</h3>

<p>The Java enum, interface, and class definitions that define pvData. This

section provides a complete definition of what pvData is.</p>

<h3>org.epics.pvdata.factory</h3>

<p>Provides everything required for creating pvData. It provides the following

factories:</p>

<dt>FieldFactory</dt>

<dd>Creates introspection interfaces.</dd>

<dt>PVDataFactory</dt>

<dd>Creates data interfaces for all of the supported data types.</dd>

<dt>ConvertFactory</dt>

<dd>Converts between support data types.</dd>

<dt>StatusFactory</dt>

<dd>Status is a class for communication status between code modules.</dd>

<dt>StandardFieldFactory</dt>

<dd>Creates introspection objects for "well known" data.</dd>

<dt>StandardPVFieldFactory</dt>

<dd>Creates data objects for "well known" data.</dd>

<p>Although pvDataFactory can provide the implementation for all supported data

types, often it is desirable to provide other implementations. To make it easy

to create alternate implementations a set of abstract and base classes are

supplied.</p>

<h3>org.epics.pvdata.property</h3>

<p>Provides a way to associated properties with a field. </p>

<p>The basic idea is to associate properties with any field named "value". All

the fields in the structure that contains the value field are considered

properties of value with the field name being the property name. See that

package overview for details.</p>

<p>This package also provides support for "well known" field definitions like

timeStamp, alarm, display,etc. Code that uses pvData can be simplified by using

this support.</p>

<h3>org.epics.pvdata.misc</h3>

<p>This package provides support that is used by pvData factories and might

also be useful to software that uses pvData.</p>

<h3>org.epics.pvdata.monitor</h3>

<p>Provides the ability to monitor changes to an arbitrary subset of the fields

in a record.</p>

<h2 >Package org.epics.pvdata.pv</h2>

<h3>Overview</h3>

<p>This package has the complete set of enum, interface, and class definitions

that describe pvData. The implementation is provided in package

org.epics.pvdata.factory.</p>

<p>A <code>PVStructure</code> is a field that contains an array of subfields. Each field has

code for accessing the field. The interface for each field is <code>PVField</code> or an

interface that extends <code>PVField</code>. Each field also has an introspection interface,

which is <code>Field</code> or an extension of <code>Field</code>. This package overview describes the

complete set of data and introspection interfaces for pvData.</p>

<p>This package also describes an interface Convert, which provides a rich set

of methods for converting and copying data between field.</p>

<p>The interface <code>FieldCreate</code> creates the introspection interfaces. The

interface <code>PVDataCreate</code> creates the <code>PVField</code> interfaces. Between them they

provide the ability to create every type of <code>Field</code> and <code>PVField</code>, i.e. they

provide a complete implementation of pvData. It is also possible for other code

to provide implementations.</p>

<p>The interface <code>StandardField</code> provides introspection objects for standard fields.

The interface <code>StandardPVField</code> provides data objects for standard fields.</p>

<p>The interface <code>StatusCreate</code> and class <code>Status</code> provide status objects to pass between source

modules.</p>

<h3>Process Variable Reflection</h3>

<p>Given the introspection object it is

possible to introspect a field without requiring access to data. The

reflection and data interfaces are separate because the data may not be

available. For example when a pvAccess client connects to a PV, the client

library can obtain the reflection information without obtaining any data. Only

when a client issues an I/O request will data be available. This separation is

especially important for arrays and structures so that a client can discover

the type without requiring that a large array or structure be transported over

the network.</p>

<h4>Type</h4>

<p>The types are defined by the Java definitions:</p>

enum Type {

scalar,

scalarArray,

structure,

structureArray,

union,

unionArrray;

}

enum ScalarType {

pvBoolean,

pvByte, pvShort, pvInt, pvLong,

pvUByte, pvUShort, pvUInt, pvULong,

pvFloat,pvDouble,

pvString;

//Convenience methods

public boolean isInteger(); // pvByte,...,pvULong

public boolean isUInteger(); // pvUByte,...,pvULong

public boolean isNumeric(); // pvByte,...pvDouble

public boolean isPrimitive(); // pvBoolean,...pvDouble

public static ScalarType getScalarType(String type);

public String toString();

}</pre>

<h4>Serializable</h4>

<p>The following interfaces are called by pvAccess for transporting data over

the network. The abstract and base classes ensure that these methods are

properly implemented. </p>

interface Serializable {

void serialize(ByteBuffer buffer,SerializableControl flusher);

void deserialize(ByteBuffer buffer,DeserializableControl control);

}

where

<dt>serialize</dt>

<dd>Serialize into buffer. flusher is called when buffer is full.</dd>

<dt>deserialize</dt>

<dd>deserialize from byte buffer. control is called when more data is required.</dd>

interface SerializableControl {

void flushSerializeBuffer();

void ensureBuffer(int size);

void alignBuffer(int alignment);

void cachedSerialize(Field field, ByteBuffer buffer)

}

where

<dt>flushSerializeBuffer</dt>

<dd>Code that called serialize must empty the buffer.

The call can block until the buffer is flushed.</dd>

<dt>ensureBuffer</dt>

<dd>Helper method. Ensures specified size of bytes, flushes if necessary.</dd>

<dt>alignBuffer</dt>

<dd>Align buffer.Note that this takes care only current buffer alignment. If streaming protocol is used,

care must be taken that entire stream is aligned.</dd>

<dt>cachedSerialize</dt>

<dd>Serialize <code>Field</code> instance via cache.</dd>

interface DeserializableControl {

void ensureData(int size);

void alignData(int alignment);

Field cachedDeserialize(ByteBuffer buffer)

}

where

<dt>ensureData</dt>

<dd>Helper method. Ensures specified size of bytes, provides it if necessary.</dd>

<dt>alignData</dt>

<dd>Align buffer.Note that this takes care only current buffer alignment. If streaming protocol is used,

care must be taken that entire stream is aligned.</dd>

<dt>cachedDeserialize</dt>

<dd>Deserialize <code>Field</code> instance via cache.</dd>

interface SerializableArray extends Serializable {

void serialize(ByteBuffer buffer, SerializableControl flusher, int offset, int count);

}

where

<dt>serialize</dt>

<dd>Serialize field into given buffer.</dd>

interface BitSetSerializable {

void serialize(ByteBuffer buffer, SerializableControl flusher, BitSet bitSet);

void deserialize(ByteBuffer buffer, DeserializableControl control, BitSet bitSet);

}

where

<dt>serialize</dt>

<dd>Serialize field into given buffer. The <code>BitSet</code> shows the fields to serialize.</dd>

<dt>deserialize</dt>

<dd>Deserialize field into given buffer. The <code>BitSet</code> shows the fields to serialize.</dd>

<h4>Reflection</h4>

<p>This section defines the complete set of Java PV reflection interfaces.

Reflection consists of the following:</p>

<dt>Field</dt>

<dd>A <code>Field</code> has an ID and a type. It can be converted to a string.

The format is the metadata format

described in the overview.</dd>

<dt>Scalar</dt>

<dd>A <code>Scalar</code> has a <code>ScalarType</code></dd>

<dt>ScalarArray</dt>

<dd>A <code>ScalarArray</code> has a <code>ScalarType</code> corresponding to the element type</dd>

<dt>Structure</dt>

<dd>Has fields that can be any of the supported types.

In addition it has a name for each field.</dd>

<dt>StructureArray</dt>

<dd>The field holds <code>Structure[]</code>. Each element has the same <code>Structure</code>

introspection interface. A client can only <code>get</code>/<code>put</code> entire <code>PVStructure</code>

elements NOT subfields of array elements.</dd>

<dt>Union</dt>

<dd>A union field stores a single field. A given union field can either store a

field of any type (variant union) or any of a specified set of types

(regular union). <code>Union</code> describes which of these

applies and the set of types in the case of a regular union. </dd>

<dt>UnionArray</dt>

<dd>The field holds <code>Union[]</code>. Each element has the same

<code>Union</code> introspection interface.</dd>

<dt>FieldCreate</dt>

<dd>This is an interface that provides methods to create introspection

interfaces. A factory is provided to create a <code>FieldCreate</code>.</dd>

<h5>Field</h5>

interface Field extends Serializable {

String getID();

Type getType();

void toString(StringBuilder buf));

void toString(StringBuilder buf,int indentLevel);

String toString();

}

where

<dt>getID</dt>

<dd>Get the identification string.

It can be empty.

For scalar fields the ID is the metadata type, i.e. boolean, byte, etc.

For scalarArray fields the ID is the metadata type, i.e. boolean[],

byte[], etc.

For structure fields the ID is determined by the argument specified

when fieldCreate.createStructure is called.

For structureArray fields the ID is XX[] where XX is the ID of the element structure.

</dd>

<dt>getType</dt>

<dd>Get the field type.</dd>

<dt>toString</dt>

<dd>Create a String that shows the type.

The format is the metadata syntax defined in the overview.</dd>

<h5>Scalar</h5>

interface Scalar extends Field {

ScalarType getScalarType();

}

where

<dt>getScalarType</dt>

<dd>Return the ScalarType.</dd>

<h5>ScalarArray</h5>

interface ScalarArray extends Field{

ScalarType getElementType();

}

where

<dt>getElementType</dt>

<dd>Return the <code>ScalarType</code> for each array element.</dd>

<h5>Structure</h5>

interface Structure extends Field{

Field getField(String fieldName);

Field getField(int fieldIndex);

&lt;T extends Field&gt; T getField(Class&lt;T&gt; c, String fieldName);

&lt;T extends Field&gt; T getField(Class&lt;T&gt; c, int fieldIndex);

Field[] getFields();

String[] getFieldNames();

int getFieldIndex(String fieldName);

String getFieldName(int fieldIndex);

}

where

<dt>getField</dt>

<dd>Return a field of the structure.

The field can be requested by name or by index.</dd>

<dt>getField - generic versions</dt>

<dd>These are the methods that get a field of the structure and convert it to the proper type.</dd>

<dt>getFields</dt>

<dd>Return the array of sub fields.</dd>

<dt>getFieldNames</dt>

<dd>Return the fieldNames for the fields.</dd>

<dt>getFieldName</dt>

<dd>Return the <code>Field</code> for the index.</dd>

<dt>getFieldIndex</dt>

<dd>Return the index of the field with the specified fieldName.</dd>

<h5>StructureArray</h5>

interface StructureArray extends Field{

Structure getStructure();

}

where

<dt>getStructure</dt>

<dd>Return the introspection interface for the array elements.</dd>

<h5>Union</h5>

public interface Union extends Field{

public static final String DEFAULT_ID = "union";

public static final String ANY_ID = "any";

Field getField(String fieldName);

Field getField(int fieldIndex);

&lt;T extends Field&gt; T getField(Class&lt;T&gt; c, String fieldName);

&lt;T extends Field&gt; T getField(Class&lt;T&gt; c, int fieldIndex);

Field[] getFields();

String[] getFieldNames();

int getFieldIndex(String fieldName);

String getFieldName(int fieldIndex);

boolean isVariant();

}

where

<dt>getField</dt>

<dd>Given a name or an index the type is returned.

<b>null</b> is returned if not found.

</dd>

<dt>getFields</dt>

<dd>Get the array of types.</dd>

<dt>getFieldNames</dt>

<dd>Get the array of names.</dd>

<dt>getFieldIndex</dt>

<dd>Get the index for name. -1 is returned if not found.</dd>

<dt>getFieldName</dt>

<dd>Get the name for the specified index.</dd>

<dt>isVariant</dt>

<dd>returns <b>true</b> if this is variant array and <b>false</b> otherwise.</dd>

<h5>UnionArray</h5>

public interface UnionArray extends Field{

Union getUnion();

}

where

<dt>getUnion</dt>

<dd>Get the union interface for each element.</dd>

<h5>FieldCreate</h5>

<p><b>Syntax for fieldName and ID.</b>

A <b>fieldName</b> must begin with a letter and must be a sequence of letters

and digits.

A letter is defined as 'A'-'Z', 'a'-'z','_', or any Unicode character

that denotes a letter in a language. Similarly digits are '0'-'9' and any

Unicode character that denotes a digit in a language.

Note that this is the same as the Java syntax for variable names.

<p>The syntax for <b>ID</b> is the same except the '.' is also allowed

after the initial letter.</p>

<pre>public interface FieldCreate {

FieldBuilder createFieldBuilder();

Scalar createScalar(ScalarType scalarType);

BoundedString createBoundedString(int maxLength);

ScalarArray createScalarArray(ScalarType elementType);

ScalarArray createFixedScalarArray(ScalarType elementType, int size);

ScalarArray createBoundedScalarArray(ScalarType elementType, int bound);

StructureArray createStructureArray(Structure elementStructure);

UnionArray createUnionArray(Union elementUnion);

UnionArray createVariantUnionArray();

Structure createStructure(String[] fieldNames, Field[] field);

Structure createStructure(String id,String[] fieldNames, Field[] field);

Structure appendField(Structure structure,String fieldName, Field field);

Structure appendFields(Structure structure,String[] fieldNames, Field[] fields);

Structure createStructure(Structure structToClone);

Union createVariantUnion();

Union createUnion(String[] fieldNames, Field[] fields);

Union createUnion(String id, String[] fieldNames, Field[] fields);

Field deserialize(ByteBuffer buffer, DeserializableControl control);

}</pre>

where

<dt>createFieldBuilder</dt>

<dd>Create an instance of <code>FieldBuilder</code>, which is described next.</dd>

<dt>createScalar</dt>

<dd>Return the <code>Scalar</code> with the specified <code>ScalarType</code>.

Note that the implementation creates a single instance for

each <code>ScalarType</code>.</dd>

<dt>createBoundedString</dt>

<dd>Return a scalar of type <code>pvString</code> that has a bounded size for a string.</dd>

<dt>createScalarArray</dt>

<dd>Return the <code>ScalarArray</code> with the specified <code>elementType</code>.

Note that the implementation creates a single instance for

each <code>elementType</code>.</dd>

<dt>createFixedScalarArray</dt>

<dd>Return the <code>ScalarArray</code> with the specified <code>elementType</code> and specifies that size of

element size of any object that has this introspection interface.</dd>

<dt>createBoundedScalarArray</dt>

<dd>Return the <code>ScalarArray</code> with the specified <code> elementType</code> and specifies that maximum

element size of any object that has this introspection interface.</dd>

<dt>createStructureArray</dt>

<dd>Return a <code>StructureArray</code> with the specified introspection

interface for each array element.</dd>

<dt>createUnionArray</dt>

<dd>Return a <code>UnionArray</code> with the specified introspection interface.</dd>

<dt>createVariantUnionArray</dt>

<dd>Return a <code>UnionArray</code> of variant union elements.</dd>

<dt>createStructure</dt>

<dd>Return a <code>Structure</code>. There are two methods.

The first creates a structure with an empty ID which results in an ID of <b>structure</b>.

</dd>

<dt>appendField</dt>

<dd>Append a field to a structure.</dd>

<dt>appendFields</dt>

<dd>Append an array of fields to a structure.</dd>

<dt>createVariantUnion</dt>

<dd>Create a variant union, i.e. a union where the subfield can be any valid type.</dd>

<dt>createUnion</dt>

<dd>Create a union where the subfield can have any of the types an names defined

by the arguments and the associated names.

The default ID is <b>union</b> for regular unions and <b>any</b> for variant unions.

</dd>

<dt>deserialize</dt>

<dd>Deserialize a field from the the buffer.</dd>

<h5>FieldBuilder</h5>

<p><code>FieldBuilder</code> is a convenience class for creating introspection interfaces.

An example is:</p>

FieldCreate fieldCreate = FieldFactory.getFieldCreate();

Structure s = fieldCreate.createFieldBuilder().

add("double", ScalarType.pvDouble).

addNestedStructure("nested").

add("short", ScalarType.pvShort).

add("long", ScalarType.pvLong).

endNested().

addArray("intArray", ScalarType.pvInt).

createStructure();

System.out.println(s);

This produces:

structure

double double

structure nested

short short

long long

int[] intArray

<p>The Java interface is:</p>

public interface FieldBuilder

{

public FieldBuilder setId(String id);

public FieldBuilder add(String name, ScalarType scalarType);

public FieldBuilder addBoundedString(String name, int maxLength);

public FieldBuilder add(String name, Field field);

public FieldBuilder addArray(String name, ScalarType scalarType);

public FieldBuilder addFixedArray(String name, ScalarType scalarType, int size);

public FieldBuilder addBoundedArray(String name, ScalarType scalarType, int bound);

public FieldBuilder addArray(String name, Field element);

public Structure createStructure();

public Union createUnion();

FieldBuilder addNestedStructure(String name);

FieldBuilder addNestedUnion(String name);

FieldBuilder addNestedStructureArray(String name);

FieldBuilder addNestedUnionArray(String name);

FieldBuilder endNested();

}

<dt>setId</dt>

<dd>

Specify the ID for the next <code>createStructure</code> or <code>createUnion call</code>.

</dd>

<dt>add</dt>

<dd>

Add a name and field for then next call to <code>createStructure</code> or <code>createUnion</code>.

There are two forms: one for an existing field and the other for a scalar field.

</dd>

<dt>addBoundedString</dt>

<dd>Add a bounded string field.</dd>

<dt>addArray</dt>

<dd>

Add a name and array field for then next call to <code>createStructure</code> or <code>createUnion</code>.

There are two forms: one for an existing field and the other for a scalar field.

Note that for an existing field this specifies the element type.

</dd>

<dt>addFixedArray</dt>

<dd>Add a scalar fixed array field.</dd>

<dt>addBoundedArray</dt>

<dd>Add a scalar bounded array field.</dd>

<dt>createStructure</dt>

<dd>

Create a <code>Structure</code> from the previous <code>add</code> and <code>addArray</code> methods.

</dd>

<dt>createUnion</dt>

<dd>

Create a <code>Union</code> from the previous <code>add</code> and <code>addArray</code> methods.

</dd>