CollationElementIterator.java revision 2362
449854c2a07b50ea64d9d6a8b03d18d4afeeee43Ken Stubbings/*
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * Copyright (c) 1996, 2005, Oracle and/or its affiliates. All rights reserved.
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster *
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * This code is free software; you can redistribute it and/or modify it
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * under the terms of the GNU General Public License version 2 only, as
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * published by the Free Software Foundation. Oracle designates this
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * particular file as subject to the "Classpath" exception as provided
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * by Oracle in the LICENSE file that accompanied this code.
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster *
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * This code is distributed in the hope that it will be useful, but WITHOUT
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * version 2 for more details (a copy is included in the LICENSE file that
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * accompanied this code).
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster *
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * You should have received a copy of the GNU General Public License version
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * 2 along with this work; if not, write to the Free Software Foundation,
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster *
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * or visit www.oracle.com if you need additional information or have any
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * questions.
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster */
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster/*
449854c2a07b50ea64d9d6a8b03d18d4afeeee43Ken Stubbings * (C) Copyright Taligent, Inc. 1996, 1997 - All Rights Reserved
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * (C) Copyright IBM Corp. 1996-1998 - All Rights Reserved
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster *
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * The original version of this source code and documentation is copyrighted
449854c2a07b50ea64d9d6a8b03d18d4afeeee43Ken Stubbings * and owned by Taligent, Inc., a wholly-owned subsidiary of IBM. These
449854c2a07b50ea64d9d6a8b03d18d4afeeee43Ken Stubbings * materials are provided under terms of a License Agreement between Taligent
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * and Sun. This technology is protected by multiple US and International
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * patents. This notice and attribution to Taligent may not be removed.
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * Taligent is a registered trademark of Taligent, Inc.
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster *
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster */
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Fosterpackage java.text;
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Fosterimport java.lang.Character;
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Fosterimport java.util.Vector;
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Fosterimport sun.text.CollatorUtilities;
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Fosterimport sun.text.normalizer.NormalizerBase;
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster/**
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * The <code>CollationElementIterator</code> class is used as an iterator
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * to walk through each character of an international string. Use the iterator
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * to return the ordering priority of the positioned character. The ordering
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * priority of a character, which we refer to as a key, defines how a character
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * is collated in the given collation object.
449854c2a07b50ea64d9d6a8b03d18d4afeeee43Ken Stubbings *
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * <p>
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * For example, consider the following in Spanish:
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * <blockquote>
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * <pre>
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * "ca" -> the first key is key('c') and second key is key('a').
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * "cha" -> the first key is key('ch') and second key is key('a').
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * </pre>
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * </blockquote>
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * And in German,
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * <blockquote>
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * <pre>
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * "\u00e4b"-> the first key is key('a'), the second key is key('e'), and
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * the third key is key('b').
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * </pre>
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * </blockquote>
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * The key of a character is an integer composed of primary order(short),
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * secondary order(byte), and tertiary order(byte). Java strictly defines
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * the size and signedness of its primitive data types. Therefore, the static
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * functions <code>primaryOrder</code>, <code>secondaryOrder</code>, and
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * <code>tertiaryOrder</code> return <code>int</code>, <code>short</code>,
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * and <code>short</code> respectively to ensure the correctness of the key
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * value.
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster *
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * <p>
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * Example of the iterator usage,
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * <blockquote>
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * <pre>
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster *
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * String testString = "This is a test";
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * RuleBasedCollator ruleBasedCollator = (RuleBasedCollator)Collator.getInstance();
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * CollationElementIterator collationElementIterator = ruleBasedCollator.getCollationElementIterator(testString);
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * int primaryOrder = CollationElementIterator.primaryOrder(collationElementIterator.next());
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * </pre>
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * </blockquote>
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster *
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * <p>
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * <code>CollationElementIterator.next</code> returns the collation order
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * of the next character. A collation order consists of primary order,
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * secondary order and tertiary order. The data type of the collation
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * order is <strong>int</strong>. The first 16 bits of a collation order
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * is its primary order; the next 8 bits is the secondary order and the
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * last 8 bits is the tertiary order.
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster *
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * @see Collator
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * @see RuleBasedCollator
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * @author Helena Shih, Laura Werner, Richard Gillam
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster */
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Fosterpublic final class CollationElementIterator
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster{
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster /**
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * Null order which indicates the end of string is reached by the
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * cursor.
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster */
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster public final static int NULLORDER = 0xffffffff;
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster /**
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * CollationElementIterator constructor. This takes the source string and
449854c2a07b50ea64d9d6a8b03d18d4afeeee43Ken Stubbings * the collation object. The cursor will walk thru the source string based
449854c2a07b50ea64d9d6a8b03d18d4afeeee43Ken Stubbings * on the predefined collation rules. If the source string is empty,
449854c2a07b50ea64d9d6a8b03d18d4afeeee43Ken Stubbings * NULLORDER will be returned on the calls to next().
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster * @param sourceText the source string.
5bdd6bf9211505ff52afc7e32bdc49cdfacf4879Charles Sparey * @param order the collation object.
5bdd6bf9211505ff52afc7e32bdc49cdfacf4879Charles Sparey */
449854c2a07b50ea64d9d6a8b03d18d4afeeee43Ken Stubbings CollationElementIterator(String sourceText, RuleBasedCollator owner) {
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster this.owner = owner;
a688bcbb4bcff5398fdd29b86f83450257dc0df4Allan Foster ordering = owner.getTables();
if ( sourceText.length() != 0 ) {
NormalizerBase.Mode mode =
CollatorUtilities.toNormalizerMode(owner.getDecomposition());
text = new NormalizerBase(sourceText, mode);
}
}
/**
* CollationElementIterator constructor. This takes the source string and
* the collation object. The cursor will walk thru the source string based
* on the predefined collation rules. If the source string is empty,
* NULLORDER will be returned on the calls to next().
* @param sourceText the source string.
* @param order the collation object.
*/
CollationElementIterator(CharacterIterator sourceText, RuleBasedCollator owner) {
this.owner = owner;
ordering = owner.getTables();
NormalizerBase.Mode mode =
CollatorUtilities.toNormalizerMode(owner.getDecomposition());
text = new NormalizerBase(sourceText, mode);
}
/**
* Resets the cursor to the beginning of the string. The next call
* to next() will return the first collation element in the string.
*/
public void reset()
{
if (text != null) {
text.reset();
NormalizerBase.Mode mode =
CollatorUtilities.toNormalizerMode(owner.getDecomposition());
text.setMode(mode);
}
buffer = null;
expIndex = 0;
swapOrder = 0;
}
/**
* Get the next collation element in the string. <p>This iterator iterates
* over a sequence of collation elements that were built from the string.
* Because there isn't necessarily a one-to-one mapping from characters to
* collation elements, this doesn't mean the same thing as "return the
* collation element [or ordering priority] of the next character in the
* string".</p>
* <p>This function returns the collation element that the iterator is currently
* pointing to and then updates the internal pointer to point to the next element.
* previous() updates the pointer first and then returns the element. This
* means that when you change direction while iterating (i.e., call next() and
* then call previous(), or call previous() and then call next()), you'll get
* back the same element twice.</p>
*/
public int next()
{
if (text == null) {
return NULLORDER;
}
NormalizerBase.Mode textMode = text.getMode();
// convert the owner's mode to something the Normalizer understands
NormalizerBase.Mode ownerMode =
CollatorUtilities.toNormalizerMode(owner.getDecomposition());
if (textMode != ownerMode) {
text.setMode(ownerMode);
}
// if buffer contains any decomposed char values
// return their strength orders before continuing in
// the Normalizer's CharacterIterator.
if (buffer != null) {
if (expIndex < buffer.length) {
return strengthOrder(buffer[expIndex++]);
} else {
buffer = null;
expIndex = 0;
}
} else if (swapOrder != 0) {
if (Character.isSupplementaryCodePoint(swapOrder)) {
char[] chars = Character.toChars(swapOrder);
swapOrder = chars[1];
return chars[0] << 16;
}
int order = swapOrder << 16;
swapOrder = 0;
return order;
}
int ch = text.next();
// are we at the end of Normalizer's text?
if (ch == NormalizerBase.DONE) {
return NULLORDER;
}
int value = ordering.getUnicodeOrder(ch);
if (value == RuleBasedCollator.UNMAPPED) {
swapOrder = ch;
return UNMAPPEDCHARVALUE;
}
else if (value >= RuleBasedCollator.CONTRACTCHARINDEX) {
value = nextContractChar(ch);
}
if (value >= RuleBasedCollator.EXPANDCHARINDEX) {
buffer = ordering.getExpandValueList(value);
expIndex = 0;
value = buffer[expIndex++];
}
if (ordering.isSEAsianSwapping()) {
int consonant;
if (isThaiPreVowel(ch)) {
consonant = text.next();
if (isThaiBaseConsonant(consonant)) {
buffer = makeReorderedBuffer(consonant, value, buffer, true);
value = buffer[0];
expIndex = 1;
} else if (consonant != NormalizerBase.DONE) {
text.previous();
}
}
if (isLaoPreVowel(ch)) {
consonant = text.next();
if (isLaoBaseConsonant(consonant)) {
buffer = makeReorderedBuffer(consonant, value, buffer, true);
value = buffer[0];
expIndex = 1;
} else if (consonant != NormalizerBase.DONE) {
text.previous();
}
}
}
return strengthOrder(value);
}
/**
* Get the previous collation element in the string. <p>This iterator iterates
* over a sequence of collation elements that were built from the string.
* Because there isn't necessarily a one-to-one mapping from characters to
* collation elements, this doesn't mean the same thing as "return the
* collation element [or ordering priority] of the previous character in the
* string".</p>
* <p>This function updates the iterator's internal pointer to point to the
* collation element preceding the one it's currently pointing to and then
* returns that element, while next() returns the current element and then
* updates the pointer. This means that when you change direction while
* iterating (i.e., call next() and then call previous(), or call previous()
* and then call next()), you'll get back the same element twice.</p>
* @since 1.2
*/
public int previous()
{
if (text == null) {
return NULLORDER;
}
NormalizerBase.Mode textMode = text.getMode();
// convert the owner's mode to something the Normalizer understands
NormalizerBase.Mode ownerMode =
CollatorUtilities.toNormalizerMode(owner.getDecomposition());
if (textMode != ownerMode) {
text.setMode(ownerMode);
}
if (buffer != null) {
if (expIndex > 0) {
return strengthOrder(buffer[--expIndex]);
} else {
buffer = null;
expIndex = 0;
}
} else if (swapOrder != 0) {
if (Character.isSupplementaryCodePoint(swapOrder)) {
char[] chars = Character.toChars(swapOrder);
swapOrder = chars[1];
return chars[0] << 16;
}
int order = swapOrder << 16;
swapOrder = 0;
return order;
}
int ch = text.previous();
if (ch == NormalizerBase.DONE) {
return NULLORDER;
}
int value = ordering.getUnicodeOrder(ch);
if (value == RuleBasedCollator.UNMAPPED) {
swapOrder = UNMAPPEDCHARVALUE;
return ch;
} else if (value >= RuleBasedCollator.CONTRACTCHARINDEX) {
value = prevContractChar(ch);
}
if (value >= RuleBasedCollator.EXPANDCHARINDEX) {
buffer = ordering.getExpandValueList(value);
expIndex = buffer.length;
value = buffer[--expIndex];
}
if (ordering.isSEAsianSwapping()) {
int vowel;
if (isThaiBaseConsonant(ch)) {
vowel = text.previous();
if (isThaiPreVowel(vowel)) {
buffer = makeReorderedBuffer(vowel, value, buffer, false);
expIndex = buffer.length - 1;
value = buffer[expIndex];
} else {
text.next();
}
}
if (isLaoBaseConsonant(ch)) {
vowel = text.previous();
if (isLaoPreVowel(vowel)) {
buffer = makeReorderedBuffer(vowel, value, buffer, false);
expIndex = buffer.length - 1;
value = buffer[expIndex];
} else {
text.next();
}
}
}
return strengthOrder(value);
}
/**
* Return the primary component of a collation element.
* @param order the collation element
* @return the element's primary component
*/
public final static int primaryOrder(int order)
{
order &= RBCollationTables.PRIMARYORDERMASK;
return (order >>> RBCollationTables.PRIMARYORDERSHIFT);
}
/**
* Return the secondary component of a collation element.
* @param order the collation element
* @return the element's secondary component
*/
public final static short secondaryOrder(int order)
{
order = order & RBCollationTables.SECONDARYORDERMASK;
return ((short)(order >> RBCollationTables.SECONDARYORDERSHIFT));
}
/**
* Return the tertiary component of a collation element.
* @param order the collation element
* @return the element's tertiary component
*/
public final static short tertiaryOrder(int order)
{
return ((short)(order &= RBCollationTables.TERTIARYORDERMASK));
}
/**
* Get the comparison order in the desired strength. Ignore the other
* differences.
* @param order The order value
*/
final int strengthOrder(int order)
{
int s = owner.getStrength();
if (s == Collator.PRIMARY)
{
order &= RBCollationTables.PRIMARYDIFFERENCEONLY;
} else if (s == Collator.SECONDARY)
{
order &= RBCollationTables.SECONDARYDIFFERENCEONLY;
}
return order;
}
/**
* Sets the iterator to point to the collation element corresponding to
* the specified character (the parameter is a CHARACTER offset in the
* original string, not an offset into its corresponding sequence of
* collation elements). The value returned by the next call to next()
* will be the collation element corresponding to the specified position
* in the text. If that position is in the middle of a contracting
* character sequence, the result of the next call to next() is the
* collation element for that sequence. This means that getOffset()
* is not guaranteed to return the same value as was passed to a preceding
* call to setOffset().
*
* @param newOffset The new character offset into the original text.
* @since 1.2
*/
public void setOffset(int newOffset)
{
if (text != null) {
if (newOffset < text.getBeginIndex()
|| newOffset >= text.getEndIndex()) {
text.setIndexOnly(newOffset);
} else {
int c = text.setIndex(newOffset);
// if the desired character isn't used in a contracting character
// sequence, bypass all the backing-up logic-- we're sitting on
// the right character already
if (ordering.usedInContractSeq(c)) {
// walk backwards through the string until we see a character
// that DOESN'T participate in a contracting character sequence
while (ordering.usedInContractSeq(c)) {
c = text.previous();
}
// now walk forward using this object's next() method until
// we pass the starting point and set our current position
// to the beginning of the last "character" before or at
// our starting position
int last = text.getIndex();
while (text.getIndex() <= newOffset) {
last = text.getIndex();
next();
}
text.setIndexOnly(last);
// we don't need this, since last is the last index
// that is the starting of the contraction which encompass
// newOffset
// text.previous();
}
}
}
buffer = null;
expIndex = 0;
swapOrder = 0;
}
/**
* Returns the character offset in the original text corresponding to the next
* collation element. (That is, getOffset() returns the position in the text
* corresponding to the collation element that will be returned by the next
* call to next().) This value will always be the index of the FIRST character
* corresponding to the collation element (a contracting character sequence is
* when two or more characters all correspond to the same collation element).
* This means if you do setOffset(x) followed immediately by getOffset(), getOffset()
* won't necessarily return x.
*
* @return The character offset in the original text corresponding to the collation
* element that will be returned by the next call to next().
* @since 1.2
*/
public int getOffset()
{
return (text != null) ? text.getIndex() : 0;
}
/**
* Return the maximum length of any expansion sequences that end
* with the specified comparison order.
* @param order a collation order returned by previous or next.
* @return the maximum length of any expansion sequences ending
* with the specified order.
* @since 1.2
*/
public int getMaxExpansion(int order)
{
return ordering.getMaxExpansion(order);
}
/**
* Set a new string over which to iterate.
*
* @param source the new source text
* @since 1.2
*/
public void setText(String source)
{
buffer = null;
swapOrder = 0;
expIndex = 0;
NormalizerBase.Mode mode =
CollatorUtilities.toNormalizerMode(owner.getDecomposition());
if (text == null) {
text = new NormalizerBase(source, mode);
} else {
text.setMode(mode);
text.setText(source);
}
}
/**
* Set a new string over which to iterate.
*
* @param source the new source text.
* @since 1.2
*/
public void setText(CharacterIterator source)
{
buffer = null;
swapOrder = 0;
expIndex = 0;
NormalizerBase.Mode mode =
CollatorUtilities.toNormalizerMode(owner.getDecomposition());
if (text == null) {
text = new NormalizerBase(source, mode);
} else {
text.setMode(mode);
text.setText(source);
}
}
//============================================================
// privates
//============================================================
/**
* Determine if a character is a Thai vowel (which sorts after
* its base consonant).
*/
private final static boolean isThaiPreVowel(int ch) {
return (ch >= 0x0e40) && (ch <= 0x0e44);
}
/**
* Determine if a character is a Thai base consonant
*/
private final static boolean isThaiBaseConsonant(int ch) {
return (ch >= 0x0e01) && (ch <= 0x0e2e);
}
/**
* Determine if a character is a Lao vowel (which sorts after
* its base consonant).
*/
private final static boolean isLaoPreVowel(int ch) {
return (ch >= 0x0ec0) && (ch <= 0x0ec4);
}
/**
* Determine if a character is a Lao base consonant
*/
private final static boolean isLaoBaseConsonant(int ch) {
return (ch >= 0x0e81) && (ch <= 0x0eae);
}
/**
* This method produces a buffer which contains the collation
* elements for the two characters, with colFirst's values preceding
* another character's. Presumably, the other character precedes colFirst
* in logical order (otherwise you wouldn't need this method would you?).
* The assumption is that the other char's value(s) have already been
* computed. If this char has a single element it is passed to this
* method as lastValue, and lastExpansion is null. If it has an
* expansion it is passed in lastExpansion, and colLastValue is ignored.
*/
private int[] makeReorderedBuffer(int colFirst,
int lastValue,
int[] lastExpansion,
boolean forward) {
int[] result;
int firstValue = ordering.getUnicodeOrder(colFirst);
if (firstValue >= RuleBasedCollator.CONTRACTCHARINDEX) {
firstValue = forward? nextContractChar(colFirst) : prevContractChar(colFirst);
}
int[] firstExpansion = null;
if (firstValue >= RuleBasedCollator.EXPANDCHARINDEX) {
firstExpansion = ordering.getExpandValueList(firstValue);
}
if (!forward) {
int temp1 = firstValue;
firstValue = lastValue;
lastValue = temp1;
int[] temp2 = firstExpansion;
firstExpansion = lastExpansion;
lastExpansion = temp2;
}
if (firstExpansion == null && lastExpansion == null) {
result = new int [2];
result[0] = firstValue;
result[1] = lastValue;
}
else {
int firstLength = firstExpansion==null? 1 : firstExpansion.length;
int lastLength = lastExpansion==null? 1 : lastExpansion.length;
result = new int[firstLength + lastLength];
if (firstExpansion == null) {
result[0] = firstValue;
}
else {
System.arraycopy(firstExpansion, 0, result, 0, firstLength);
}
if (lastExpansion == null) {
result[firstLength] = lastValue;
}
else {
System.arraycopy(lastExpansion, 0, result, firstLength, lastLength);
}
}
return result;
}
/**
* Check if a comparison order is ignorable.
* @return true if a character is ignorable, false otherwise.
*/
final static boolean isIgnorable(int order)
{
return ((primaryOrder(order) == 0) ? true : false);
}
/**
* Get the ordering priority of the next contracting character in the
* string.
* @param ch the starting character of a contracting character token
* @return the next contracting character's ordering. Returns NULLORDER
* if the end of string is reached.
*/
private int nextContractChar(int ch)
{
// First get the ordering of this single character,
// which is always the first element in the list
Vector list = ordering.getContractValues(ch);
EntryPair pair = (EntryPair)list.firstElement();
int order = pair.value;
// find out the length of the longest contracting character sequence in the list.
// There's logic in the builder code to make sure the longest sequence is always
// the last.
pair = (EntryPair)list.lastElement();
int maxLength = pair.entryName.length();
// (the Normalizer is cloned here so that the seeking we do in the next loop
// won't affect our real position in the text)
NormalizerBase tempText = (NormalizerBase)text.clone();
// extract the next maxLength characters in the string (we have to do this using the
// Normalizer to ensure that our offsets correspond to those the rest of the
// iterator is using) and store it in "fragment".
tempText.previous();
key.setLength(0);
int c = tempText.next();
while (maxLength > 0 && c != NormalizerBase.DONE) {
if (Character.isSupplementaryCodePoint(c)) {
key.append(Character.toChars(c));
maxLength -= 2;
} else {
key.append((char)c);
--maxLength;
}
c = tempText.next();
}
String fragment = key.toString();
// now that we have that fragment, iterate through this list looking for the
// longest sequence that matches the characters in the actual text. (maxLength
// is used here to keep track of the length of the longest sequence)
// Upon exit from this loop, maxLength will contain the length of the matching
// sequence and order will contain the collation-element value corresponding
// to this sequence
maxLength = 1;
for (int i = list.size() - 1; i > 0; i--) {
pair = (EntryPair)list.elementAt(i);
if (!pair.fwd)
continue;
if (fragment.startsWith(pair.entryName) && pair.entryName.length()
> maxLength) {
maxLength = pair.entryName.length();
order = pair.value;
}
}
// seek our current iteration position to the end of the matching sequence
// and return the appropriate collation-element value (if there was no matching
// sequence, we're already seeked to the right position and order already contains
// the correct collation-element value for the single character)
while (maxLength > 1) {
c = text.next();
maxLength -= Character.charCount(c);
}
return order;
}
/**
* Get the ordering priority of the previous contracting character in the
* string.
* @param ch the starting character of a contracting character token
* @return the next contracting character's ordering. Returns NULLORDER
* if the end of string is reached.
*/
private int prevContractChar(int ch)
{
// This function is identical to nextContractChar(), except that we've
// switched things so that the next() and previous() calls on the Normalizer
// are switched and so that we skip entry pairs with the fwd flag turned on
// rather than off. Notice that we still use append() and startsWith() when
// working on the fragment. This is because the entry pairs that are used
// in reverse iteration have their names reversed already.
Vector list = ordering.getContractValues(ch);
EntryPair pair = (EntryPair)list.firstElement();
int order = pair.value;
pair = (EntryPair)list.lastElement();
int maxLength = pair.entryName.length();
NormalizerBase tempText = (NormalizerBase)text.clone();
tempText.next();
key.setLength(0);
int c = tempText.previous();
while (maxLength > 0 && c != NormalizerBase.DONE) {
if (Character.isSupplementaryCodePoint(c)) {
key.append(Character.toChars(c));
maxLength -= 2;
} else {
key.append((char)c);
--maxLength;
}
c = tempText.previous();
}
String fragment = key.toString();
maxLength = 1;
for (int i = list.size() - 1; i > 0; i--) {
pair = (EntryPair)list.elementAt(i);
if (pair.fwd)
continue;
if (fragment.startsWith(pair.entryName) && pair.entryName.length()
> maxLength) {
maxLength = pair.entryName.length();
order = pair.value;
}
}
while (maxLength > 1) {
c = text.previous();
maxLength -= Character.charCount(c);
}
return order;
}
final static int UNMAPPEDCHARVALUE = 0x7FFF0000;
private NormalizerBase text = null;
private int[] buffer = null;
private int expIndex = 0;
private StringBuffer key = new StringBuffer(5);
private int swapOrder = 0;
private RBCollationTables ordering;
private RuleBasedCollator owner;
}