Back to index...

	/*
	* Copyright (c) 2015, Oracle and/or its affiliates. All rights reserved.
	* DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
	*
	* This code is free software; you can redistribute it and/or modify it
	* under the terms of the GNU General Public License version 2 only, as
	* published by the Free Software Foundation. Oracle designates this
	* particular file as subject to the "Classpath" exception as provided
	* by Oracle in the LICENSE file that accompanied this code.
	*
	* This code is distributed in the hope that it will be useful, but WITHOUT
	* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
	* FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
	* version 2 for more details (a copy is included in the LICENSE file that
	* accompanied this code).
	*
	* You should have received a copy of the GNU General Public License version
	* 2 along with this work; if not, write to the Free Software Foundation,
	* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
	*
	* Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
	* or visit www.oracle.com if you need additional information or have any
	* questions.
	*/

	/*
	*******************************************************************************
	* Copyright (C) 2009-2014, International Business Machines Corporation and
	* others. All Rights Reserved.
	*******************************************************************************
	*/

	package sun.text.normalizer;

	import java.io.IOException;
	import java.nio.ByteBuffer;


	/**
	* @author aheninger
	*
	* A read-only Trie2, holding 16 bit data values.
	*
	* A Trie2 is a highly optimized data structure for mapping from Unicode
	* code points (values ranging from 0 to 0x10ffff) to a 16 or 32 bit value.
	*
	* See class Trie2 for descriptions of the API for accessing the contents of a trie.
	*
	* The fundamental data access methods are declared final in this class, with
	* the intent that applications might gain a little extra performance, when compared
	* with calling the same methods via the abstract UTrie2 base class.
	*/
	public final class Trie2_16 extends Trie2 {

	/**
	* Internal constructor, not for general use.
	*/
	Trie2_16() {
	}


	/**
	* Create a Trie2 from its serialized form. Inverse of utrie2_serialize().
	* The serialized format is identical between ICU4C and ICU4J, so this function
	* will work with serialized Trie2s from either.
	*
	* The serialized Trie2 in the bytes may be in either little or big endian byte order.
	* This allows using serialized Tries from ICU4C without needing to consider the
	* byte order of the system that created them.
	*
	* @param bytes a byte buffer to the serialized form of a UTrie2.
	* @return An unserialized Trie2_16, ready for use.
	* @throws IllegalArgumentException if the buffer does not contain a serialized Trie2.
	* @throws IOException if a read error occurs in the buffer.
	* @throws ClassCastException if the bytes contain a serialized Trie2_32
	*/
	public static Trie2_16 createFromSerialized(ByteBuffer bytes) throws IOException {
	return (Trie2_16) Trie2.createFromSerialized(bytes);
	}

	/**
	* Get the value for a code point as stored in the Trie2.
	*
	* @param codePoint the code point
	* @return the value
	*/
	@Override
	public final int get(int codePoint) {
	int value;
	int ix;

	if (codePoint >= 0) {
	if (codePoint < 0x0d800 \|\| (codePoint > 0x0dbff && codePoint <= 0x0ffff)) {
	// Ordinary BMP code point, excluding leading surrogates.
	// BMP uses a single level lookup. BMP index starts at offset 0 in the Trie2 index.
	// 16 bit data is stored in the index array itself.
	ix = index[codePoint >> UTRIE2_SHIFT_2];
	ix = (ix << UTRIE2_INDEX_SHIFT) + (codePoint & UTRIE2_DATA_MASK);
	value = index[ix];
	return value;
	}
	if (codePoint <= 0xffff) {
	// Lead Surrogate Code Point. A Separate index section is stored for
	// lead surrogate code units and code points.
	// The main index has the code unit data.
	// For this function, we need the code point data.
	// Note: this expression could be refactored for slightly improved efficiency, but
	// surrogate code points will be so rare in practice that it's not worth it.
	ix = index[UTRIE2_LSCP_INDEX_2_OFFSET + ((codePoint - 0xd800) >> UTRIE2_SHIFT_2)];
	ix = (ix << UTRIE2_INDEX_SHIFT) + (codePoint & UTRIE2_DATA_MASK);
	value = index[ix];
	return value;
	}
	if (codePoint < highStart) {
	// Supplemental code point, use two-level lookup.
	ix = (UTRIE2_INDEX_1_OFFSET - UTRIE2_OMITTED_BMP_INDEX_1_LENGTH) + (codePoint >> UTRIE2_SHIFT_1);
	ix = index[ix];
	ix += (codePoint >> UTRIE2_SHIFT_2) & UTRIE2_INDEX_2_MASK;
	ix = index[ix];
	ix = (ix << UTRIE2_INDEX_SHIFT) + (codePoint & UTRIE2_DATA_MASK);
	value = index[ix];
	return value;
	}
	if (codePoint <= 0x10ffff) {
	value = index[highValueIndex];
	return value;
	}
	}

	// Fall through. The code point is outside of the legal range of 0..0x10ffff.
	return errorValue;
	}


	/**
	* Get a Trie2 value for a UTF-16 code unit.
	*
	* This function returns the same value as get() if the input
	* character is outside of the lead surrogate range
	*
	* There are two values stored in a Trie2 for inputs in the lead
	* surrogate range. This function returns the alternate value,
	* while Trie2.get() returns the main value.
	*
	* @param codeUnit a 16 bit code unit or lead surrogate value.
	* @return the value
	*/
	@Override
	public int getFromU16SingleLead(char codeUnit) {
	int value;
	int ix;

	// Because the input is a 16 bit char, we can skip the tests for it being in
	// the BMP range. It is.
	ix = index[codeUnit >> UTRIE2_SHIFT_2];
	ix = (ix << UTRIE2_INDEX_SHIFT) + (codeUnit & UTRIE2_DATA_MASK);
	value = index[ix];
	return value;
	}

	/**
	* @return the number of bytes of the serialized trie
	*/
	public int getSerializedLength() {
	return 16+(header.indexLength+dataLength)*2;
	}
	}

Back to index...