001/*
002 * Copyright (c) 2013, 2015, Oracle and/or its affiliates. All rights reserved.
003 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
004 *
005 * This code is free software; you can redistribute it and/or modify it
006 * under the terms of the GNU General Public License version 2 only, as
007 * published by the Free Software Foundation.
008 *
009 * This code is distributed in the hope that it will be useful, but WITHOUT
010 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
011 * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
012 * version 2 for more details (a copy is included in the LICENSE file that
013 * accompanied this code).
014 *
015 * You should have received a copy of the GNU General Public License version
016 * 2 along with this work; if not, write to the Free Software Foundation,
017 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
018 *
019 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
020 * or visit www.oracle.com if you need additional information or have any
021 * questions.
022 */
023package jdk.internal.jvmci.hotspot;
024
025import static jdk.internal.jvmci.common.UnsafeAccess.*;
026import static jdk.internal.jvmci.hotspot.HotSpotJVMCIRuntime.*;
027
028import java.io.*;
029import java.util.*;
030
031/**
032 * Represents a metaspace {@code Symbol}.
033 */
034public class HotSpotSymbol {
035
036    private final long metaspaceSymbol;
037
038    public HotSpotSymbol(long metaspaceSymbol) {
039        assert metaspaceSymbol != 0;
040        this.metaspaceSymbol = metaspaceSymbol;
041    }
042
043    /**
044     * Decodes this {@code Symbol} and returns the symbol string as {@link java.lang.String}.
045     *
046     * @return the decoded string, or null if there was a decoding error
047     */
048    public String asString() {
049        return readModifiedUTF8(asByteArray());
050    }
051
052    /**
053     * Reads the modified UTF-8 string in {@code buf} and converts it to a {@link String}. The
054     * implementation is taken from {@link java.io.DataInputStream#readUTF(DataInput)} and adapted
055     * to operate on a {@code byte} array directly for performance reasons.
056     *
057     * @see java.io.DataInputStream#readUTF(DataInput)
058     */
059    private static String readModifiedUTF8(byte[] buf) {
060        final int utflen = buf.length;
061        byte[] bytearr = null;
062        char[] chararr = new char[utflen];
063
064        int c;
065        int char2;
066        int char3;
067        int count = 0;
068        int chararrCount = 0;
069
070        bytearr = buf;
071
072        while (count < utflen) {
073            c = bytearr[count] & 0xff;
074            if (c > 127) {
075                break;
076            }
077            count++;
078            chararr[chararrCount++] = (char) c;
079        }
080
081        while (count < utflen) {
082            c = bytearr[count] & 0xff;
083            switch (c >> 4) {
084                case 0:
085                case 1:
086                case 2:
087                case 3:
088                case 4:
089                case 5:
090                case 6:
091                case 7:
092                    /* 0xxxxxxx */
093                    count++;
094                    chararr[chararrCount++] = (char) c;
095                    break;
096                case 12:
097                case 13:
098                    /* 110x xxxx 10xx xxxx */
099                    count += 2;
100                    if (count > utflen) {
101                        // malformed input: partial character at end
102                        return null;
103                    }
104                    char2 = bytearr[count - 1];
105                    if ((char2 & 0xC0) != 0x80) {
106                        // malformed input around byte
107                        return null;
108                    }
109                    chararr[chararrCount++] = (char) (((c & 0x1F) << 6) | (char2 & 0x3F));
110                    break;
111                case 14:
112                    /* 1110 xxxx 10xx xxxx 10xx xxxx */
113                    count += 3;
114                    if (count > utflen) {
115                        // malformed input: partial character at end
116                        return null;
117                    }
118                    char2 = bytearr[count - 2];
119                    char3 = bytearr[count - 1];
120                    if (((char2 & 0xC0) != 0x80) || ((char3 & 0xC0) != 0x80)) {
121                        // malformed input around byte
122                        return null;
123                    }
124                    chararr[chararrCount++] = (char) (((c & 0x0F) << 12) | ((char2 & 0x3F) << 6) | ((char3 & 0x3F) << 0));
125                    break;
126                default:
127                    /* 10xx xxxx, 1111 xxxx */
128                    // malformed input around byte
129                    return null;
130            }
131        }
132        // The number of chars produced may be less than utflen
133        char[] value = Arrays.copyOf(chararr, chararrCount);
134        return new String(value);
135    }
136
137    private byte[] asByteArray() {
138        final int length = getLength();
139        byte[] result = new byte[length];
140        for (int index = 0; index < length; index++) {
141            result[index] = getByteAt(index);
142        }
143        return result;
144    }
145
146    private int getLength() {
147        return unsafe.getShort(metaspaceSymbol + runtime().getConfig().symbolLengthOffset);
148    }
149
150    private byte getByteAt(int index) {
151        return unsafe.getByte(metaspaceSymbol + runtime().getConfig().symbolBodyOffset + index);
152    }
153}