1.1 --- a/src/com/sun/org/apache/regexp/internal/REProgram.java Sat Oct 24 16:18:47 2020 +0800
1.2 +++ /dev/null Thu Jan 01 00:00:00 1970 +0000
1.3 @@ -1,158 +0,0 @@
1.4 -/*
1.5 - * reserved comment block
1.6 - * DO NOT REMOVE OR ALTER!
1.7 - */
1.8 -/*
1.9 - * Copyright 1999-2004 The Apache Software Foundation.
1.10 - *
1.11 - * Licensed under the Apache License, Version 2.0 (the "License");
1.12 - * you may not use this file except in compliance with the License.
1.13 - * You may obtain a copy of the License at
1.14 - *
1.15 - * http://www.apache.org/licenses/LICENSE-2.0
1.16 - *
1.17 - * Unless required by applicable law or agreed to in writing, software
1.18 - * distributed under the License is distributed on an "AS IS" BASIS,
1.19 - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
1.20 - * See the License for the specific language governing permissions and
1.21 - * limitations under the License.
1.22 - */
1.23 -
1.24 -package com.sun.org.apache.regexp.internal;
1.25 -
1.26 -import java.io.Serializable;
1.27 -
1.28 -/**
1.29 - * A class that holds compiled regular expressions. This is exposed mainly
1.30 - * for use by the recompile utility (which helps you produce precompiled
1.31 - * REProgram objects). You should not otherwise need to work directly with
1.32 - * this class.
1.33 -*
1.34 - * @see RE
1.35 - * @see RECompiler
1.36 - *
1.37 - * @author <a href="mailto:jonl@muppetlabs.com">Jonathan Locke</a>
1.38 - */
1.39 -public class REProgram implements Serializable
1.40 -{
1.41 - static final int OPT_HASBACKREFS = 1;
1.42 -
1.43 - char[] instruction; // The compiled regular expression 'program'
1.44 - int lenInstruction; // The amount of the instruction buffer in use
1.45 - char[] prefix; // Prefix string optimization
1.46 - int flags; // Optimization flags (REProgram.OPT_*)
1.47 - int maxParens = -1;
1.48 -
1.49 - /**
1.50 - * Constructs a program object from a character array
1.51 - * @param instruction Character array with RE opcode instructions in it
1.52 - */
1.53 - public REProgram(char[] instruction)
1.54 - {
1.55 - this(instruction, instruction.length);
1.56 - }
1.57 -
1.58 - /**
1.59 - * Constructs a program object from a character array
1.60 - * @param parens Count of parens in the program
1.61 - * @param instruction Character array with RE opcode instructions in it
1.62 - */
1.63 - public REProgram(int parens, char[] instruction)
1.64 - {
1.65 - this(instruction, instruction.length);
1.66 - this.maxParens = parens;
1.67 - }
1.68 -
1.69 - /**
1.70 - * Constructs a program object from a character array
1.71 - * @param instruction Character array with RE opcode instructions in it
1.72 - * @param lenInstruction Amount of instruction array in use
1.73 - */
1.74 - public REProgram(char[] instruction, int lenInstruction)
1.75 - {
1.76 - setInstructions(instruction, lenInstruction);
1.77 - }
1.78 -
1.79 - /**
1.80 - * Returns a copy of the current regular expression program in a character
1.81 - * array that is exactly the right length to hold the program. If there is
1.82 - * no program compiled yet, getInstructions() will return null.
1.83 - * @return A copy of the current compiled RE program
1.84 - */
1.85 - public char[] getInstructions()
1.86 - {
1.87 - // Ensure program has been compiled!
1.88 - if (lenInstruction != 0)
1.89 - {
1.90 - // Return copy of program
1.91 - char[] ret = new char[lenInstruction];
1.92 - System.arraycopy(instruction, 0, ret, 0, lenInstruction);
1.93 - return ret;
1.94 - }
1.95 - return null;
1.96 - }
1.97 -
1.98 - /**
1.99 - * Sets a new regular expression program to run. It is this method which
1.100 - * performs any special compile-time search optimizations. Currently only
1.101 - * two optimizations are in place - one which checks for backreferences
1.102 - * (so that they can be lazily allocated) and another which attempts to
1.103 - * find an prefix anchor string so that substantial amounts of input can
1.104 - * potentially be skipped without running the actual program.
1.105 - * @param instruction Program instruction buffer
1.106 - * @param lenInstruction Length of instruction buffer in use
1.107 - */
1.108 - public void setInstructions(char[] instruction, int lenInstruction)
1.109 - {
1.110 - // Save reference to instruction array
1.111 - this.instruction = instruction;
1.112 - this.lenInstruction = lenInstruction;
1.113 -
1.114 - // Initialize other program-related variables
1.115 - flags = 0;
1.116 - prefix = null;
1.117 -
1.118 - // Try various compile-time optimizations if there's a program
1.119 - if (instruction != null && lenInstruction != 0)
1.120 - {
1.121 - // If the first node is a branch
1.122 - if (lenInstruction >= RE.nodeSize && instruction[0 + RE.offsetOpcode] == RE.OP_BRANCH)
1.123 - {
1.124 - // to the end node
1.125 - int next = instruction[0 + RE.offsetNext];
1.126 - if (instruction[next + RE.offsetOpcode] == RE.OP_END)
1.127 - {
1.128 - // and the branch starts with an atom
1.129 - if (lenInstruction >= (RE.nodeSize * 2) && instruction[RE.nodeSize + RE.offsetOpcode] == RE.OP_ATOM)
1.130 - {
1.131 - // then get that atom as an prefix because there's no other choice
1.132 - int lenAtom = instruction[RE.nodeSize + RE.offsetOpdata];
1.133 - prefix = new char[lenAtom];
1.134 - System.arraycopy(instruction, RE.nodeSize * 2, prefix, 0, lenAtom);
1.135 - }
1.136 - }
1.137 - }
1.138 -
1.139 - BackrefScanLoop:
1.140 -
1.141 - // Check for backreferences
1.142 - for (int i = 0; i < lenInstruction; i += RE.nodeSize)
1.143 - {
1.144 - switch (instruction[i + RE.offsetOpcode])
1.145 - {
1.146 - case RE.OP_ANYOF:
1.147 - i += (instruction[i + RE.offsetOpdata] * 2);
1.148 - break;
1.149 -
1.150 - case RE.OP_ATOM:
1.151 - i += instruction[i + RE.offsetOpdata];
1.152 - break;
1.153 -
1.154 - case RE.OP_BACKREF:
1.155 - flags |= OPT_HASBACKREFS;
1.156 - break BackrefScanLoop;
1.157 - }
1.158 - }
1.159 - }
1.160 - }
1.161 -}
