re.java

linux下建立JAVA虚拟机的源码KAFFE

		    if (addition.size() == 0) addition = null;
		    result.token = new RETokenOneOf(subIndex,
			options, addition, negative);
		    result.index = index - 1;
		    result.returnAtAndOperator = true;
		    return result;
		}
		// The precedence of the operator "&&" is the lowest.
		// So we postpone adding "&" until other elements
		// are added. And we insert Boolean.FALSE at the
		// beginning of the list of tokens following "&&".
		// So, "&&[a-b][k-m]" will be stored in the Vecter
		// addition in this order:
		//     Boolean.FALSE, [a-b], "|", [k-m], "|", "&"
		if (additionAndAppeared) addition.addElement("&");
		addition.addElement(Boolean.FALSE);
		additionAndAppeared = true;

		// The part on which "&&" operates may be either
		//   (1) explicitly enclosed by []
		//   or
		//   (2) not enclosed by [] and terminated by the
		//       next "&&" or the end of the character list.
	        //  Let the preceding else if block do the case (1).
		//  We must do something in case of (2).
		if ((index + 1 < pLength) && (pattern[index + 1] != '[')) {
		    ParseCharClassResult result = parseCharClass(
			subIndex, pattern, index+1, pLength, cflags, syntax,
			RETURN_AT_AND);
		    addition.addElement(result.token);
		    addition.addElement("|");
		    // If the method returned at the next "&&", it is OK.
		    // Otherwise we have eaten the mark of the end of this
		    // character list "]".  In this case we must give back
		    // the end mark.
		    index = (result.returnAtAndOperator ?
			result.index: result.index - 1);
		}
	  } else {
	    if (lastCharIsSet) options.addElement(new RETokenChar(subIndex,lastChar,insens));
	    lastChar = ch; lastCharIsSet = true;
	  }
	  if (index == pLength) throw new REException(getLocalizedMessage("class.no.end"),REException.REG_EBRACK,index);
	} // while in list
	// Out of list, index is one past ']'
	    
	if (lastCharIsSet) options.addElement(new RETokenChar(subIndex,lastChar,insens));
	   
	ParseCharClassResult result = new ParseCharClassResult(); 
	// Create a new RETokenOneOf
	options.trimToSize();
	if (additionAndAppeared) addition.addElement("&");
	if (addition.size() == 0) addition = null;
	result.token = new RETokenOneOf(subIndex,options, addition, negative);
	result.index = index;
	return result;
  }

  private static int getCharUnit(char[] input, int index, CharUnit unit, boolean quot) throws REException {
    unit.ch = input[index++];
    unit.bk = (unit.ch == '\\'
	       && (!quot || index >= input.length || input[index] == 'E'));
    if (unit.bk)
      if (index < input.length)
	unit.ch = input[index++];
      else throw new REException(getLocalizedMessage("ends.with.backslash"),REException.REG_ESCAPE,index);
    return index;
  }

  private static int parseInt(char[] input, int pos, int len, int radix) {
    int ret = 0;
    for (int i = pos; i < pos + len; i++) {
	ret = ret * radix + Character.digit(input[i], radix);
    }
    return ret;
  }

  /**
   * This class represents various expressions for a character.
   * "a"      : 'a' itself.
   * "\0123"  : Octal char 0123
   * "\x1b"   : Hex char 0x1b
   * "\u1234" : Unicode char \u1234
   */
  private static class CharExpression {
    /** character represented by this expression */
    char ch;
    /** String expression */
    String expr;
    /** length of this expression */
    int len;
    public String toString() { return expr; }
  }

  private static CharExpression getCharExpression(char[] input, int pos, int lim,
        RESyntax syntax) {
    CharExpression ce = new CharExpression();
    char c = input[pos];
    if (c == '\\') {
      if (pos + 1 >= lim) return null;
      c = input[pos + 1];
      switch(c) {
      case 't':
        ce.ch = '\t';
        ce.len = 2;
        break;
      case 'n':
        ce.ch = '\n';
        ce.len = 2;
        break;
      case 'r':
        ce.ch = '\r';
        ce.len = 2;
        break;
      case 'x':
      case 'u':
        if ((c == 'x' && syntax.get(RESyntax.RE_HEX_CHAR)) ||
            (c == 'u' && syntax.get(RESyntax.RE_UNICODE_CHAR))) {
          int l = 0;
          int expectedLength = (c == 'x' ? 2 : 4);
          for (int i = pos + 2; i < pos + 2 + expectedLength; i++) {
            if (i >= lim) break;
            if (!((input[i] >= '0' && input[i] <= '9') ||
                  (input[i] >= 'A' && input[i] <= 'F') ||
                  (input[i] >= 'a' && input[i] <= 'f')))
                break;
	    l++;
          }
          if (l != expectedLength) return null;
          ce.ch = (char)(parseInt(input, pos + 2, l, 16));
	  ce.len = l + 2;
        }
        else {
          ce.ch = c;
          ce.len = 2;
        }
        break;
      case '0':
        if (syntax.get(RESyntax.RE_OCTAL_CHAR)) {
          int l = 0;
          for (int i = pos + 2; i < pos + 2 + 3; i++) {
            if (i >= lim) break;
	    if (input[i] < '0' || input[i] > '7') break;
            l++;
          }
          if (l == 3 && input[pos + 2] > '3') l--;
          if (l <= 0) return null;
          ce.ch = (char)(parseInt(input, pos + 2, l, 8));
          ce.len = l + 2;
        }
        else {
          ce.ch = c;
          ce.len = 2;
        }
        break;
      default:
        ce.ch = c;
        ce.len = 2;
        break;
      }
    }
    else {
      ce.ch = input[pos];
      ce.len = 1;
    }
    ce.expr = new String(input, pos, ce.len);
    return ce;
  }

  /**
   * This class represents a substring in a pattern string expressing
   * a named property.
   * "\pA"      : Property named "A"
   * "\p{prop}" : Property named "prop"
   * "\PA"      : Property named "A" (Negated)
   * "\P{prop}" : Property named "prop" (Negated)
   */
  private static class NamedProperty {
    /** Property name */
    String name;
    /** Negated or not */
    boolean negate;
    /** length of this expression */
    int len;
  }

  private static NamedProperty getNamedProperty(char[] input, int pos, int lim) {
    NamedProperty np = new NamedProperty();
    char c = input[pos];
    if (c == '\\') {
      if (++pos >= lim) return null;
      c = input[pos++];
      switch(c) {
      case 'p':
        np.negate = false;
        break;
      case 'P':
        np.negate = true;
        break;
      default:
	return null;
      }
      c = input[pos++];
      if (c == '{') {
          int p = -1;
	  for (int i = pos; i < lim; i++) {
	      if (input[i] == '}') {
		  p = i;
		  break;
	      }
	  }
	  if (p < 0) return null;
	  int len = p - pos;
          np.name = new String(input, pos, len);
	  np.len = len + 4;
      }
      else {
          np.name = new String(input, pos - 1, 1);
	  np.len = 3;
      }
      return np;
    }
    else return null;
  }

  private static RETokenNamedProperty getRETokenNamedProperty(
      int subIndex, NamedProperty np, boolean insens, int index)
      throws REException {
    try {
	return new RETokenNamedProperty(subIndex, np.name, insens, np.negate);
    }
    catch (REException e) {
	REException ree;
	ree = new REException(e.getMessage(), REException.REG_ESCAPE, index);
	ree.initCause(e);
	throw ree;
    }
  }

  /**
   * Checks if the regular expression matches the input in its entirety.
   *
   * @param input The input text.
   */
  public boolean isMatch(Object input) {
    return isMatch(input,0,0);
  }
  
  /**
   * Checks if the input string, starting from index, is an exact match of
   * this regular expression.
   *
   * @param input The input text.
   * @param index The offset index at which the search should be begin.
   */
  public boolean isMatch(Object input,int index) {
    return isMatch(input,index,0);
  }
  

  /**
   * Checks if the input, starting from index and using the specified
   * execution flags, is an exact match of this regular expression.
   *
   * @param input The input text.
   * @param index The offset index at which the search should be begin.
   * @param eflags The logical OR of any execution flags above.
   */
  public boolean isMatch(Object input,int index,int eflags) {
    return isMatchImpl(makeCharIndexed(input,index),index,eflags);
  }

  private boolean isMatchImpl(CharIndexed input, int index, int eflags) {
    if (firstToken == null)  // Trivial case
      return (input.charAt(0) == CharIndexed.OUT_OF_BOUNDS);
    REMatch m = new REMatch(numSubs, index, eflags);
    if (firstToken.match(input, m)) {
	if (m != null) {
	    if (input.charAt(m.index) == CharIndexed.OUT_OF_BOUNDS) {
		return true;
	    }
	}
    }
    return false;
  }
    
  /**
   * Returns the maximum number of subexpressions in this regular expression.
   * If the expression contains branches, the value returned will be the
   * maximum subexpressions in any of the branches.
   */
  public int getNumSubs() {
    return numSubs;
  }

  // Overrides REToken.setUncle
  void setUncle(REToken uncle) {
      if (lastToken != null) {
	  lastToken.setUncle(uncle);
      } else super.setUncle(uncle); // to deal with empty subexpressions
  }

  // Overrides REToken.chain

  boolean chain(REToken next) {
    super.chain(next);
    setUncle(next);
    return true;
  }

  /**
   * Returns the minimum number of characters that could possibly
   * constitute a match of this regular expression.
   */
  public int getMinimumLength() {
      return minimumLength;
  }

  public int getMaximumLength() {
      return maximumLength;
  }

  /**
   * Returns an array of all matches found in the input.
   *
   * If the regular expression allows the empty string to match, it will
   * substitute matches at all positions except the end of the input.
   *
   * @param input The input text.
   * @return a non-null (but possibly zero-length) array of matches
   */
  public REMatch[] getAllMatches(Object input) {
    return getAllMatches(input,0,0);
  }

  /**
   * Returns an array of all matches found in the input,
   * beginning at the specified index position.
   *
   * If the regular expression allows the empty string to match, it will
   * substitute matches at all positions except the end of the input.
   *
   * @param input The input text.
   * @param index The offset index at which the search should be begin.
   * @return a non-null (but possibly zero-length) array of matches
   */
  public REMatch[] getAllMatches(Object input, int index) {
    return getAllMatches(input,index,0);
  }

  /**
   * Returns an array of all matches found in the input string,
   * beginning at the specified index position and using the specified
   * execution flags.
   *
   * If the regular expression allows the empty string to match, it will
   * substitute matches at all positions except the end of the input.
   *
   * @param input The input text.
   * @param index The offset index at which the search should be begin.
   * @param eflags The logical OR of any execution flags above.
   * @return a non-null (but possibly zero-length) array of matches
   */
  public REMatch[] getAllMatches(Object input, int index, int eflags) {
    return getAllMatchesImpl(makeCharIndexed(input,index),index,eflags);
  }

  // this has been changed since 1.03 to be non-overlapping matches
  private REMatch[] getAllMatchesImpl(CharIndexed input, int index, int eflags) {
    Vector all = new Vector();
    REMatch m = null;
    while ((m = getMatchImpl(input,index,eflags,null)) != null) {
      all.addElement(m);
      index = m.getEndIndex();