regexparser.java

java1.6众多例子参考

/*
 * Copyright 1999-2004 The Apache Software Foundation.
 * 
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 * 
 *      http://www.apache.org/licenses/LICENSE-2.0
 * 
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

package com.sun.org.apache.xerces.internal.impl.xpath.regex;

import java.util.Locale;
import java.util.MissingResourceException;
import java.util.ResourceBundle;
import java.util.Vector;

/**
 * A Regular Expression Parser.
 * 
 * @xerces.internal
 *
 * @version $Id: RegexParser.java,v 1.2.6.1 2005/09/06 11:46:34 neerajbj Exp $
 */
class RegexParser {
    static final int T_CHAR = 0;
    static final int T_EOF = 1;
    static final int T_OR = 2;                  // '|'
    static final int T_STAR = 3;                // '*'
    static final int T_PLUS = 4;                // '+'
    static final int T_QUESTION = 5;            // '?'
    static final int T_LPAREN = 6;              // '('
    static final int T_RPAREN = 7;              // ')'
    static final int T_DOT = 8;                 // '.'
    static final int T_LBRACKET = 9;            // '['
    static final int T_BACKSOLIDUS = 10;        // '\'
    static final int T_CARET = 11;              // '^'
    static final int T_DOLLAR = 12;             // '$'
    static final int T_LPAREN2 = 13;            // '(?:'
    static final int T_LOOKAHEAD = 14;          // '(?='
    static final int T_NEGATIVELOOKAHEAD = 15;  // '(?!'
    static final int T_LOOKBEHIND = 16;         // '(?<='
    static final int T_NEGATIVELOOKBEHIND = 17; // '(?<!'
    static final int T_INDEPENDENT = 18;        // '(?>'
    static final int T_SET_OPERATIONS = 19;     // '(?['
    static final int T_POSIX_CHARCLASS_START = 20; // '[:' in a character class
    static final int T_COMMENT = 21;            // '(?#'
    static final int T_MODIFIERS = 22;          // '(?' [\-,a-z,A-Z]
    static final int T_CONDITION = 23;          // '(?('
    static final int T_XMLSCHEMA_CC_SUBTRACTION = 24; // '-[' in a character class

    static class ReferencePosition {
        int refNumber;
        int position;
        ReferencePosition(int n, int pos) {
            this.refNumber = n;
            this.position = pos;
        }
    }

    int offset;
    String regex;
    int regexlen;
    int options;
    ResourceBundle resources;
    int chardata;
    int nexttoken;
    static protected final int S_NORMAL = 0;
    static protected final int S_INBRACKETS = 1;
    static protected final int S_INXBRACKETS = 2;
    int context = S_NORMAL;
    int parennumber = 1;
    boolean hasBackReferences;
    Vector references = null;

    public RegexParser() {
        this.setLocale(Locale.getDefault());
    }
    public RegexParser(Locale locale) {
        this.setLocale(locale);
    }

    public void setLocale(Locale locale) {
        try {
            this.resources = ResourceBundle.getBundle("com.sun.org.apache.xerces.internal.impl.xpath.regex.message", locale);
        } catch (MissingResourceException mre) {
            throw new RuntimeException("Installation Problem???  Couldn't load messages: "
                                       +mre.getMessage());
        }
    }

    final ParseException ex(String key, int loc) {
        return new ParseException(this.resources.getString(key), loc);
    }

    private final boolean isSet(int flag) {
        return (this.options & flag) == flag;
    }

    synchronized Token parse(String regex, int options) throws ParseException {
        this.options = options;
        this.offset = 0;
        this.setContext(S_NORMAL);
        this.parennumber = 1;
        this.hasBackReferences = false;
        this.regex = regex;
        if (this.isSet(RegularExpression.EXTENDED_COMMENT))
            this.regex = REUtil.stripExtendedComment(this.regex);
        this.regexlen = this.regex.length();


        this.next();
        Token ret = this.parseRegex();
        if (this.offset != this.regexlen)
            throw ex("parser.parse.1", this.offset);
        if (this.references != null) {
            for (int i = 0;  i < this.references.size();  i ++) {
                ReferencePosition position = (ReferencePosition)this.references.elementAt(i);
                if (this.parennumber <= position.refNumber)
                    throw ex("parser.parse.2", position.position);
            }
            this.references.removeAllElements();
        }
        return ret;
    }

    /*
    public RegularExpression createRegex(String regex, int options) throws ParseException {
        Token tok = this.parse(regex, options);
        return new RegularExpression(regex, tok, this.parennumber, this.hasBackReferences, options);
    }
    */

    protected final void setContext(int con) {
        this.context = con;
    }

    final int read() {
        return this.nexttoken;
    }

    final void next() {
        if (this.offset >= this.regexlen) {
            this.chardata = -1;
            this.nexttoken = T_EOF;
            return;
        }

        int ret;
        int ch = this.regex.charAt(this.offset++);
        this.chardata = ch;

        if (this.context == S_INBRACKETS) {
            // In a character class, this.chardata has one character, that is to say,
            // a pair of surrogates is composed and stored to this.chardata.
            switch (ch) {
              case '\\':
                ret = T_BACKSOLIDUS;
                if (this.offset >= this.regexlen)
                    throw ex("parser.next.1", this.offset-1);
                this.chardata = this.regex.charAt(this.offset++);
                break;

              case '-':
                if (this.isSet(RegularExpression.XMLSCHEMA_MODE)
                    && this.offset < this.regexlen && this.regex.charAt(this.offset) == '[') {
                    this.offset++;
                    ret = T_XMLSCHEMA_CC_SUBTRACTION;
                } else
                    ret = T_CHAR;
                break;

              case '[':
                if (!this.isSet(RegularExpression.XMLSCHEMA_MODE)
                    && this.offset < this.regexlen && this.regex.charAt(this.offset) == ':') {
                    this.offset++;
                    ret = T_POSIX_CHARCLASS_START;
                    break;
                } // Through down
              default:
                if (REUtil.isHighSurrogate(ch) && this.offset < this.regexlen) {
                    int low = this.regex.charAt(this.offset);
                    if (REUtil.isLowSurrogate(low)) {
                        this.chardata = REUtil.composeFromSurrogates(ch, low);
                        this.offset ++;
                    }
                }
                ret = T_CHAR;
            }
            this.nexttoken = ret;
            return;
        }

        switch (ch) {
          case '|': ret = T_OR;             break;
          case '*': ret = T_STAR;           break;
          case '+': ret = T_PLUS;           break;
          case '?': ret = T_QUESTION;       break;
          case ')': ret = T_RPAREN;         break;
          case '.': ret = T_DOT;            break;
          case '[': ret = T_LBRACKET;       break;
          case '^': ret = T_CARET;          break;
          case '$': ret = T_DOLLAR;         break;
          case '(':
            ret = T_LPAREN;
            if (this.offset >= this.regexlen)
                break;
            if (this.regex.charAt(this.offset) != '?')
                break;
            if (++this.offset >= this.regexlen)
                throw ex("parser.next.2", this.offset-1);
            ch = this.regex.charAt(this.offset++);
            switch (ch) {
              case ':':  ret = T_LPAREN2;            break;
              case '=':  ret = T_LOOKAHEAD;          break;
              case '!':  ret = T_NEGATIVELOOKAHEAD;  break;
              case '[':  ret = T_SET_OPERATIONS;     break;
              case '>':  ret = T_INDEPENDENT;        break;
              case '<':
                if (this.offset >= this.regexlen)
                    throw ex("parser.next.2", this.offset-3);
                ch = this.regex.charAt(this.offset++);
                if (ch == '=') {
                    ret = T_LOOKBEHIND;
                } else if (ch == '!') {
                    ret = T_NEGATIVELOOKBEHIND;
                } else
                    throw ex("parser.next.3", this.offset-3);
                break;
              case '#':
                while (this.offset < this.regexlen) {
                    ch = this.regex.charAt(this.offset++);
                    if (ch == ')')  break;
                }
                if (ch != ')')
                    throw ex("parser.next.4", this.offset-1);
                ret = T_COMMENT;
                break;
              default:
                if (ch == '-' || 'a' <= ch && ch <= 'z' || 'A' <= ch && ch <= 'Z') {// Options
                    this.offset --;
                    ret = T_MODIFIERS;
                    break;
                } else if (ch == '(') {         // conditional
                    ret = T_CONDITION;          // this.offsets points the next of '('.
                    break;
                }
                throw ex("parser.next.2", this.offset-2);
            }
            break;
            
          case '\\':
            ret = T_BACKSOLIDUS;
            if (this.offset >= this.regexlen)
                throw ex("parser.next.1", this.offset-1);
            this.chardata = this.regex.charAt(this.offset++);
            break;

          default:
            ret = T_CHAR;
        }
        this.nexttoken = ret;
    }

    /**
     * regex ::= term (`|` term)*
     * term ::= factor+
     * factor ::= ('^' | '$' | '\A' | '\Z' | '\z' | '\b' | '\B' | '\<' | '\>'
     *            | atom (('*' | '+' | '?' | minmax ) '?'? )?)
     *            | '(?=' regex ')'  | '(?!' regex ')'  | '(?&lt;=' regex ')'  | '(?&lt;!' regex ')'
     * atom ::= char | '.' | range | '(' regex ')' | '(?:' regex ')' | '\' [0-9]
     *          | '\w' | '\W' | '\d' | '\D' | '\s' | '\S' | category-block 
     */
    Token parseRegex() throws ParseException {
        Token tok = this.parseTerm();
        Token parent = null;
        while (this.read() == T_OR) {
            this.next();                    // '|'
            if (parent == null) {
                parent = Token.createUnion();
                parent.addChild(tok);
                tok = parent;
            }
            tok.addChild(this.parseTerm());
        }
        return tok;
    }

    /**
     * term ::= factor+
     */
    Token parseTerm() throws ParseException {
        int ch = this.read();
        if (ch == T_OR || ch == T_RPAREN || ch == T_EOF) {
            return Token.createEmpty();
        } else {
            Token tok = this.parseFactor();
            Token concat = null;
            while ((ch = this.read()) != T_OR && ch != T_RPAREN && ch != T_EOF) {
                if (concat == null) {
                    concat = Token.createConcat();
                    concat.addChild(tok);
                    tok = concat;
                }
                concat.addChild(this.parseFactor());
                //tok = Token.createConcat(tok, this.parseFactor());
            }
            return tok;
        }
    }

    // ----------------------------------------------------------------

    Token processCaret() throws ParseException {
        this.next();
        return Token.token_linebeginning;
    }
    Token processDollar() throws ParseException {
        this.next();
        return Token.token_lineend;
    }
    Token processLookahead() throws ParseException {
        this.next();
        Token tok = Token.createLook(Token.LOOKAHEAD, this.parseRegex());
        if (this.read() != T_RPAREN)  throw ex("parser.factor.1", this.offset-1);
        this.next();                            // ')'
        return tok;
    }
    Token processNegativelookahead() throws ParseException {
        this.next();
        Token tok = Token.createLook(Token.NEGATIVELOOKAHEAD, this.parseRegex());
        if (this.read() != T_RPAREN)  throw ex("parser.factor.1", this.offset-1);
        this.next();                            // ')'
        return tok;
    }
    Token processLookbehind() throws ParseException {
        this.next();
        Token tok = Token.createLook(Token.LOOKBEHIND, this.parseRegex());
        if (this.read() != T_RPAREN)  throw ex("parser.factor.1", this.offset-1);
        this.next();                            // ')'
        return tok;
    }
    Token processNegativelookbehind() throws ParseException {
        this.next();
        Token tok = Token.createLook(Token.NEGATIVELOOKBEHIND, this.parseRegex());
        if (this.read() != T_RPAREN)  throw ex("parser.factor.1", this.offset-1);
        this.next();                    // ')'
        return tok;
    }
    Token processBacksolidus_A() throws ParseException {
        this.next();
        return Token.token_stringbeginning;
    }
    Token processBacksolidus_Z() throws ParseException {
        this.next();
        return Token.token_stringend2;