com.ibm.as400.data.RfmlRecordFormat Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of jt400 Show documentation
Show all versions of jt400 Show documentation
The Open Source version of the IBM Toolbox for Java
The newest version!
///////////////////////////////////////////////////////////////////////////////
//
// JTOpen (IBM Toolbox for Java - OSS version)
//
// Filename: RfmlRecordFormat.java
//
// The source code contained herein is licensed under the IBM Public License
// Version 1.0, which has been approved by the Open Source Initiative.
// Copyright (C) 1997-2002 International Business Machines Corporation and
// others. All rights reserved.
//
///////////////////////////////////////////////////////////////////////////////
package com.ibm.as400.data;
import java.io.IOException;
import java.io.ObjectOutputStream;
import java.io.ByteArrayOutputStream;
import java.math.BigDecimal;
import java.math.BigInteger;
import java.util.Enumeration;
import java.util.Hashtable;
import java.util.StringTokenizer;
import java.util.TimeZone;
import java.util.Vector;
import com.ibm.as400.access.*;
import com.ibm.as400.access.Record;
/**
**/
class RfmlRecordFormat extends PcmlDocNode
{
static final long serialVersionUID = 5L;
private static final String RECORD_FORMAT_ATTRIBUTES[] = {
"name",
"description",
};
private String m_Description;
private static final int DESCRIPTION_MAX_LENGTH = 50; // maximum length of "description" attribute value.
private static final int MAX_DDS_NAME_LENGTH = 10; // maximum length of "DDS name" for a FieldDescription.
private static final int MAX_FIELD_TEXT_LENGTH = 50; // maximum length of "TEXT" (description) for a FieldDescription.
/**
Constructor
* @param attrs
**/
public RfmlRecordFormat(PcmlAttributeList attrs)
{
super(attrs);
setNodeType(PcmlNodeType.RECORDFORMAT);
// **********************************
// Set attribute values
//
// The following code extracts the attribute values
// from the parsed document node and
// stores the values in private data members.
// **********************************
// Note: The superclass does the setName().
setDescription(getAttributeValue("description"));
}
/**
Creates and returns a copy of this object.
@return A copy of this object.
**/
public Object clone()
{
RfmlRecordFormat node = (RfmlRecordFormat) super.clone();
// Cloning does not include 'live' data, only the interface
// definitions described by the PCML tags.
// Null out the 'semi-transient' data values.
return node;
}
/**
Composes a list of FieldDescription objects representing the specified node and its children.
**/
private static void addFieldDescriptions(PcmlDocNode node, Vector fieldDescriptions, Vector keyFields, Vector namesAlreadyUsed, String preferredFieldName, TimeZone timeZone) throws XmlException // @A1c added arg
{
boolean typeIsStruct = false; // We set this to true if current node is or . // @A1a
String fieldName = null; // @A1c
// If this is a STRUCT node, and we are being passed a parent name, then set up to pass the name on to the child. // @A1a - This entire 'if' block.
if (node.getNodeType() == PcmlNodeType.STRUCT) {
typeIsStruct = true;
if (preferredFieldName != null) fieldName = preferredFieldName;
else fieldName = node.getName();
}
// If this is a DATA node, then set up the appropriate field description object and add it to the descriptions list.
else if (node.getNodeType() == PcmlNodeType.DATA) // It's a node. @A1c
{
RfmlData dNode = (RfmlData)node;
PcmlDimensions noDimensions = new PcmlDimensions();
int fieldLength;
try { fieldLength = dNode.getLength(noDimensions); }
catch (PcmlException e) { throw new XmlException(e); }
if (preferredFieldName != null) fieldName = preferredFieldName; // @A1a
else fieldName = dNode.getName(); // @A1c
FieldDescription fieldDesc = null;
String initValue = dNode.getInit();
int precision = dNode.getPrecision();
switch (dNode.getDataType())
{
// Node is .
case (PcmlData.STRUCT):
{
// Do nothing, a doesn't convert directly to a FieldDescription. Skip it and go on to resolve any child nodes.
typeIsStruct = true; // @A1a
break;
}
// Node is .
case (PcmlData.CHAR) :
{
// See if the ccsid attribute is set.
int ccsid = dNode.getCcsid();
AS400Text convAS400Text;
if (ccsid == 0) { // ccsid was not specified
convAS400Text = new AS400Text(fieldLength);
}
else {
convAS400Text = new AS400Text(fieldLength, ccsid);
}
fieldDesc = new CharacterFieldDescription(convAS400Text, fieldName);
// Note: Unless we do the explicit setCCSID(), the FieldDescription doesn't report the correct CCSID. @A1a - This entire 'if' block.
if (ccsid != 0) {
((CharacterFieldDescription)fieldDesc).setCCSID(Integer.toString(ccsid));
}
if (initValue != null) {
((CharacterFieldDescription)fieldDesc).setDFT(initValue);
}
break;
}
// Node is .
case (PcmlData.INT) :
{
switch (fieldLength)
{
case 2: // Note: All DDS 'binary' fields occupy at least 2 bytes on server.
if (precision == 16) { // Unsigned. @A1c - Swapped the if/else.
fieldDesc = new BinaryFieldDescription(new AS400UnsignedBin2(), fieldName);
if (initValue != null) {
((BinaryFieldDescription)fieldDesc).setDFT(Integer.valueOf(initValue));
}
}
else { // Signed. @A1c
fieldDesc = new BinaryFieldDescription(new AS400Bin2(), fieldName);
if (initValue != null) {
((BinaryFieldDescription)fieldDesc).setDFT(Short.valueOf(initValue));
}
}
break;
case 4:
if (precision == 32) { // Unsigned. @A1c - Swapped the if/else.
fieldDesc = new BinaryFieldDescription(new AS400UnsignedBin4(), fieldName);
if (initValue != null) {
((BinaryFieldDescription)fieldDesc).setDFT(Long.valueOf(initValue));
}
}
else { // Signed. @A1c
fieldDesc = new BinaryFieldDescription(new AS400Bin4(), fieldName);
if (initValue != null) {
((BinaryFieldDescription)fieldDesc).setDFT(Integer.valueOf(initValue));
}
}
break;
case 8:
if (precision == 64) { // Unsigned.
fieldDesc = new BinaryFieldDescription(new AS400UnsignedBin8(), fieldName);
if (initValue != null) {
((BinaryFieldDescription)fieldDesc).setDFT(new BigInteger(initValue));
}
}
else { // Signed.
fieldDesc = new BinaryFieldDescription(new AS400Bin8(), fieldName);
if (initValue != null) {
((BinaryFieldDescription)fieldDesc).setDFT(Long.valueOf(initValue));
}
}
break;
default:
Trace.log(Trace.ERROR, "Invalid field length for type=int: " + fieldLength);
throw new InternalErrorException(InternalErrorException.UNKNOWN);
}
break;
}
// Node is .
case (PcmlData.ZONED) :
{
AS400ZonedDecimal convZoned = new AS400ZonedDecimal(fieldLength, precision);
fieldDesc = new ZonedDecimalFieldDescription(convZoned, fieldName);
if (initValue != null) {
BigDecimal bigDec = new BigDecimal(initValue);
if (bigDec.scale() != precision) {
bigDec = bigDec.setScale(precision,BigDecimal.ROUND_HALF_EVEN);
}
((ZonedDecimalFieldDescription)fieldDesc).setDFT(bigDec);
}
break;
}
// Node is .
case (PcmlData.PACKED) :
{
AS400PackedDecimal convPacked = new AS400PackedDecimal(fieldLength, precision);
fieldDesc = new PackedDecimalFieldDescription(convPacked, fieldName);
if (initValue != null) {
BigDecimal bigDec = new BigDecimal(initValue);
if (bigDec.scale() != precision) {
bigDec = bigDec.setScale(precision,BigDecimal.ROUND_HALF_EVEN);
}
((PackedDecimalFieldDescription)fieldDesc).setDFT(new BigDecimal(initValue));
}
break;
}
// Node is .
case (PcmlData.FLOAT) :
{
switch (fieldLength)
{
case 4:
fieldDesc = new FloatFieldDescription(new AS400Float4(), fieldName);
break;
case 8:
fieldDesc = new FloatFieldDescription(new AS400Float8(), fieldName);
break;
default:
Trace.log(Trace.ERROR, "Invalid field length for type=float: " + fieldLength);
throw new InternalErrorException(InternalErrorException.UNKNOWN);
}
if (initValue != null)
{
if (fieldLength == 4) {
((FloatFieldDescription)fieldDesc).setDFT(Float.valueOf(initValue));
}
else { // length==8
((FloatFieldDescription)fieldDesc).setDFT(Double.valueOf(initValue));
}
}
((FloatFieldDescription)fieldDesc).setLength(fieldLength);
// Note: The "precision" attribute is not allowed for type=float.
break;
}
// Node is .
case (PcmlData.BYTE) :
{
AS400ByteArray convByte = new AS400ByteArray(fieldLength);
fieldDesc = new HexFieldDescription(convByte, fieldName);
if (initValue != null) {
Object convertedValue = PcmlDataValues.convertValue(initValue, PcmlData.BYTE, fieldLength, 0, dNode.getNameForException(), timeZone); // @A1a
((HexFieldDescription)fieldDesc).setDFT((byte[])convertedValue); // @A1a
// Note: We could alternatively use Arrays.fill(). However, java.util.Arrays is new in Java2.
}
break;
}
// Node is .
case (PcmlData.DATE) :
{
String separatorName = dNode.getDateSeparator();
String format = dNode.getDateFormat();
AS400Date convDate;
if (format == null) convDate = new AS400Date(timeZone);
else {
int formatInt = AS400Date.toFormat(format);
if (separatorName == null) convDate = new AS400Date(timeZone, formatInt);
else convDate = new AS400Date(timeZone, formatInt, separatorAsChar(separatorName));
}
fieldDesc = new DateFieldDescription(convDate, fieldName);
if (initValue != null) {
// We require the 'init=' value to be specified in standard XML Schema 'date' format.
// Normalize it to match the field's specified DDS format.
String initValueNormalized = convDate.toString(AS400Date.parseXsdString(initValue, timeZone));
((DateFieldDescription)fieldDesc).setDFT(initValueNormalized);
}
if (format != null) {
((DateFieldDescription)fieldDesc).setDATFMT(format);
}
if (separatorName != null) {
((DateFieldDescription)fieldDesc).setDATSEP(separatorAsChar(separatorName).toString());
}
break;
}
// Node is .
case (PcmlData.TIME) :
{
String separatorName = dNode.getTimeSeparator();
String format = dNode.getTimeFormat();
AS400Time convTime;
if (format == null) convTime = new AS400Time(timeZone);
else {
int formatInt = AS400Time.toFormat(format);
if (separatorName == null) convTime = new AS400Time(timeZone, formatInt);
else convTime = new AS400Time(timeZone, formatInt, separatorAsChar(separatorName));
}
fieldDesc = new TimeFieldDescription(convTime, fieldName);
if (initValue != null) {
// We require the 'init=' value to be specified in standard XML Schema 'time' format.
// Normalize it to match the field's specified DDS format.
String initValueNormalized = convTime.toString(AS400Time.parseXsdString(initValue, timeZone));
((TimeFieldDescription)fieldDesc).setDFT(initValueNormalized);
}
if (format != null) {
((TimeFieldDescription)fieldDesc).setTIMFMT(format);
}
if (separatorName != null) {
((TimeFieldDescription)fieldDesc).setTIMSEP(separatorAsChar(separatorName).toString());
}
break;
}
// Node is .
case (PcmlData.TIMESTAMP) :
{
AS400Timestamp convTimestamp = new AS400Timestamp(timeZone);
fieldDesc = new TimestampFieldDescription(convTimestamp, fieldName);
if (initValue != null) {
// We require the 'init=' value to be specified in standard XML Schema 'timestamp' format.
// Normalize it to match the field's expected DDS format.
String initValueNormalized = convTimestamp.toString(AS400Timestamp.parseXsdString(initValue, timeZone));
((TimestampFieldDescription)fieldDesc).setDFT(initValueNormalized);
}
break;
}
default:
Trace.log(Trace.ERROR, "Invalid data field type: " + dNode.getDataType());
throw new InternalErrorException(InternalErrorException.UNKNOWN);
} // ... switch
if (fieldDesc != null) {
// Store the fully-qualified node name in the field text description.
String qualifiedName = dNode.getQualifiedName();
if (qualifiedName.length() <= MAX_FIELD_TEXT_LENGTH) {
fieldDesc.setTEXT(qualifiedName);
}
else {
// Truncate from the end of the qualified name, and prepend "..".
int startPos = qualifiedName.length() - MAX_FIELD_TEXT_LENGTH + 2;
String truncatedName = ".." + qualifiedName.substring(startPos);
fieldDesc.setTEXT(truncatedName);
}
// Derive a reasonably unique "DDS name" (limit is 10 characters).
String ddsName = generateUniqueName(fieldName, namesAlreadyUsed);
fieldDesc.setDDSName(ddsName); // Note: setDDSName() uppercases its argument.
namesAlreadyUsed.addElement(ddsName.toUpperCase()); // Avoid re-using the same name for another field within this RecordFormat (case-insensitive).
// Print a warning if a count was specified.
try {
int count = dNode.getCount(noDimensions);
if (count != 1 && count != 0) {
Trace.log(Trace.WARNING, "Ignoring attribute 'count' ("+count+") for field " + qualifiedName);
}
}
catch (PcmlException e) { throw new XmlException(e); }
// Add this field description to the list.
fieldDescriptions.addElement(fieldDesc);
// If this field is a "key field", add its relative field index to keyFields list.
if (dNode.isKeyField()) {
int fieldIndex = fieldDescriptions.size() - 1; // most-recently-added field
keyFields.addElement(Integer.valueOf(fieldIndex));
}
}
} // ... getNodeType() == DATA
else { // Neither nor . @A1c
// Only nodes get converted directly to FieldDescriptions.
// Do nothing with this node, and proceed to its child nodes.
}
// If this node had child nodes, generate field descriptions for them.
if (node.hasChildren())
{
// @A1a - This entire if-else block.
// If the current node is a or , and has exactly one child, then assume that the struct is being used as an "alias", and pass the parent's name to the child (for composing the child's FieldDescription).
String nameForChild;
if (typeIsStruct && (node.getNbrChildren() == 1)) {
nameForChild = fieldName;
}
else nameForChild = null; // This will force the child to use own name.
// Generate FieldDescriptions for each child node.
Enumeration children = node.getChildren();
while (children.hasMoreElements())
{
PcmlDocNode child = (PcmlDocNode) children.nextElement();
addFieldDescriptions(child, fieldDescriptions, keyFields, namesAlreadyUsed, nameForChild, timeZone); // @A1c
}
}
}
/**
Generates a unique "DDS name" for a field, based on the field name.
**/
private static String generateUniqueName(String initialName, Vector namesInUse)
{
if (initialName.length() <= MAX_DDS_NAME_LENGTH &&
!namesInUse.contains(initialName)) return initialName;
else
{
// Truncate down to the final (MAX_DDS_NAME_LENGTH) characters.
int startPos = Math.max(0, initialName.length()-MAX_DDS_NAME_LENGTH);
String baseName = initialName.substring(startPos);
if (!namesInUse.contains(baseName)) return baseName;
int baseLength = baseName.length();
// Try appending digits to the end.
String suffix = Integer.toString(0);
StringBuffer newName = new StringBuffer(baseName);
for (int i=0; suffix.length() nodes get converted directly to FieldDescriptions.
// Do nothing with this node, and proceed to its child nodes.
}
else // It's a node.
{
RfmlData dNode = (RfmlData)node;
PcmlDimensions noDimensions = new PcmlDimensions();
int fieldLength;
try { fieldLength = dNode.getLength(noDimensions); }
catch (PcmlException e) { throw new XmlException(e); }
String fieldName = dNode.getName();
switch (dNode.getDataType())
{
// Node is .
case (PcmlData.STRUCT):
// Do nothing, a doesn't convert directly to a FieldDescription. Skip it and go on to resolve any child nodes.
// If this node had child nodes, generate field names for them.
break;
// Node is .
default :
// Add this field description to the list.
dataNodes.addElement(dNode);
break;
}
} // ... else node.
// If this node had child nodes, generate field names for them.
if (node.hasChildren())
{
// Generate FieldDescriptions for each child node.
Enumeration children = node.getChildren();
while (children.hasMoreElements())
{
PcmlDocNode child = (PcmlDocNode) children.nextElement();
getDataNodes(child, dataNodes);
}
}
}
/**
Returns a string containing the recordformat description.
Returns null if description is not set.
**/
String getDescription()
{
return m_Description;
}
/**
Assigns the value of the current node and its children, by parsing the input bytes.
**/
int parseBytes(byte[] bytes)
throws XmlException
{
PcmlDimensions noDimensions = new PcmlDimensions();
// Stack of offsets used by RfmlData.parseBytes() and RfmlStruct.parseBytes()
Hashtable offsetStack = new Hashtable();
int offsetIntoBuffer = 0;
// Convert all fields from Java objects to IBM i data.
Enumeration children = getChildren(); // children of this node.
while (children.hasMoreElements())
{
PcmlDocNode child = (PcmlDocNode) children.nextElement();
// Create a FieldDescription for the node and convert the Java objects to IBM i data.
if (child.getNodeType() == PcmlNodeType.DATA) {
try {
int bytesConsumed = ((RfmlData) child).parseBytes(bytes, offsetIntoBuffer, offsetStack, noDimensions);
offsetIntoBuffer += bytesConsumed;
}
catch (ArrayIndexOutOfBoundsException e) {
Trace.log(Trace.ERROR, e);
throw new XmlException(DAMRI.INSUFFICIENT_INPUT_DATA, new Object[] {"(unknown)", Integer.toString(bytes.length), "", this.getNameForException()} );
}
}
else {
throw new XmlException(DAMRI.BAD_NODE_TYPE, new Object[] {Integer.valueOf(child.getNodeType()) , child.getNameForException()} );
}
}
return offsetIntoBuffer; // total number of bytes consumed
}
/**
Sets the values of the current node and its children, from the values of the corresponding fields in the specified Record.
* @param record
* @throws java.io.UnsupportedEncodingException
* @throws XmlException
**/
public void setValues(Record record)
throws java.io.UnsupportedEncodingException,
XmlException
{
Vector dataNodes = new Vector(); // This is where we will accumulate the data nodes of the record
getDataNodes(this, dataNodes);
for (int fieldNum=0; fieldNum declarations.
break;
case PcmlNodeType.DATA:
RfmlData dataNode = (RfmlData) child;
{
{
int bytesWritten = dataNode.toBytes(bytes, offsetForNextField, noDimensions);
offsetForNextField += bytesWritten;
}
}
break;
default:
throw new XmlException(DAMRI.BAD_NODE_TYPE, new Object[] {Integer.valueOf(child.getNodeType()) , child.getNameForException()} );
}
}
try { bytes.flush(); } catch (IOException e) { throw new XmlException(e); }
return bytes.toByteArray();
}
/**
Creates a RecordFormat object whose fields correspond to the current node and its children.
* @return RecordFormat
* @throws XmlException
**/
public RecordFormat toRecordFormat() throws XmlException
{
Vector fieldDescriptions = new Vector(); // This is where we will accumulate the generated FieldDescription objects.
Vector namesAlreadyUsed = new Vector(); // "DDS names" for fields must be unique within the RecordFormat. This Vector will accumulate the (uppercased) DDS names assigned so far, so we can generate a new unique name for each field.
Vector keyFields = new Vector(); // Indexes of key fields (within fieldDescriptions).
// Note: We ignore the "count" attribute. Regardless of what value is specified in , we will generate a single FieldDescription for the node.
// Recursively compose FieldDescription objects representing this node and its child nodes.
addFieldDescriptions(this, fieldDescriptions, keyFields, namesAlreadyUsed, null, AS400.getDefaultTimeZone(getAs400())); // @A1c
RecordFormat recordFormat = new RecordFormat(getName());
for (int i=0; i < fieldDescriptions.size(); ++i)
{
recordFormat.addFieldDescription((FieldDescription)fieldDescriptions.elementAt(i));
}
for (int i=0; i < keyFields.size(); ++i)
{
recordFormat.addKeyFieldDescription(((Integer)keyFields.elementAt(i)).intValue());
}
return recordFormat;
}
/**
Custom serialization.
**/
private void writeObject(ObjectOutputStream out) throws IOException
{
synchronized (this)
{
// Perform default serialization
out.defaultWriteObject();
}
}
}
© 2015 - 2025 Weber Informatics LLC | Privacy Policy