perl5matcher.java

jakarta-oro-2.0.8 正则表达式 引擎 源代码

	    tmp = OpCode._isWordCharacter(__previousChar);

	  while(__currentOffset < endOffset) {
	    ch = __input[__currentOffset];
	    if(tmp != OpCode._isWordCharacter(ch)){
	      tmp = !tmp;
	      if(__tryExpression(__currentOffset)) {
		success = true;
		break _mainLoop;
	      }
	    }
	    ++__currentOffset;
	  }

	  if((minLength > 0 || tmp) && 
	     __tryExpression(__currentOffset)) {
	    success = true;
	    break _mainLoop;
	  }
	  break;

	case OpCode._NBOUND:
	  if(minLength > 0) {
	    ++dontTry;
	    --endOffset;
	  }

	  if(__currentOffset != beginOffset) {
	    ch = __input[__currentOffset - 1];
	    tmp = OpCode._isWordCharacter(ch);
	  } else
	    tmp = OpCode._isWordCharacter(__previousChar);

	  while(__currentOffset < endOffset) {
	    ch = __input[__currentOffset];
	    if(tmp != OpCode._isWordCharacter(ch))
	      tmp = !tmp;
	    else if(__tryExpression(__currentOffset)) {
	      success = true;
	      break _mainLoop;
	    }

	    ++__currentOffset;
	  }

	  if((minLength > 0 || !tmp) &&
	     __tryExpression(__currentOffset)) {
	    success = true;
	    break _mainLoop;
	  }
	  break;

	case OpCode._ALNUM:
	  while(__currentOffset < endOffset) {
	    ch = __input[__currentOffset];
	    if(OpCode._isWordCharacter(ch)) {
	      if(tmp && __tryExpression(__currentOffset)) {
		success = true;
		break _mainLoop;
	      } else
		tmp = doEvery;
	    } else
	      tmp = true;
	    ++__currentOffset;
	  }
	  break;

	case OpCode._NALNUM:
	  while(__currentOffset < endOffset) {
	    ch = __input[__currentOffset];
	    if(!OpCode._isWordCharacter(ch)) {
	      if(tmp && __tryExpression(__currentOffset)) {
		success = true;
		break _mainLoop;
	      } else
		tmp = doEvery;
	    } else
	      tmp = true;
	    ++__currentOffset;
	  }
	  break;

	case OpCode._SPACE:
	  while(__currentOffset < endOffset) {
	    if(Character.isWhitespace(__input[__currentOffset])) {
	      if(tmp && __tryExpression(__currentOffset)) {
		success = true;
		break _mainLoop;
	      } else
		tmp = doEvery;
	    } else
	      tmp = true;
	    ++__currentOffset;
	  }
	  break;

	case OpCode._NSPACE:
	  while(__currentOffset < endOffset) {
	    if(!Character.isWhitespace(__input[__currentOffset])) {
	      if(tmp && __tryExpression(__currentOffset)) {
		success = true;
		break _mainLoop;
	      } else
		tmp = doEvery;
	    } else
	      tmp = true;
	    ++__currentOffset;
	  }
	  break;

	case OpCode._DIGIT:
	  while(__currentOffset < endOffset) {
	    if(Character.isDigit(__input[__currentOffset])) {
	      if(tmp && __tryExpression(__currentOffset)) {
		success = true;
		break _mainLoop;
	      } else
		tmp = doEvery;
	    } else
	      tmp = true;
	    ++__currentOffset;
	  }
	  break;


	case OpCode._NDIGIT:
	  while(__currentOffset < endOffset) {
	    if(!Character.isDigit(__input[__currentOffset])) {
	      if(tmp && __tryExpression(__currentOffset)) {
		success = true;
		break _mainLoop;
	      } else
		tmp = doEvery;
	    } else
	      tmp = true;
	    ++__currentOffset;
	  }
	  break;
	} // end switch

      } else {
	if(minLength > 0)
	  dontTry = minLength - 1;
	endOffset-=dontTry;

	do {
	  if(__tryExpression(__currentOffset)) {
	    success = true;
	    break _mainLoop;
	  }
	} while(__currentOffset++ < endOffset);

      }


      break _mainLoop;
    } // end while

    __lastSuccess = success;
    __lastMatchResult = null;

    return success;
  }
  
  private boolean __matchUnicodeClass(char code, char __program[], 
			     int offset ,char opcode)
  {
    boolean isANYOF = ( opcode == OpCode._ANYOFUN );

    while( __program[offset] != OpCode._END ){
      if( __program[offset] == OpCode._RANGE ){
	offset++;
	if((code >= __program[offset]) && (code <= __program[offset+1])){
	  return isANYOF;
	} else {
	  offset+=2;
	}

      } else if(__program[offset] == OpCode._ONECHAR) {
       	offset++;
	if(__program[offset++] == code) return isANYOF;

      } else {
	isANYOF = (__program[offset] == OpCode._OPCODE) 
	  ? isANYOF : !isANYOF;

	offset++;
	switch ( __program[offset++] ) {
	case OpCode._ALNUM:
	  if(OpCode._isWordCharacter(code)) return isANYOF;
	  break;
	case OpCode._NALNUM:
	  if(!OpCode._isWordCharacter(code)) return isANYOF;
	  break;
	case OpCode._SPACE:
	  if(Character.isWhitespace(code)) return isANYOF;
	  break;
	case OpCode._NSPACE:
	  if(!Character.isWhitespace(code)) return isANYOF;
	  break;
	case OpCode._DIGIT:
	  if(Character.isDigit(code)) return isANYOF;
	  break;
	case OpCode._NDIGIT:
	  if(!Character.isDigit(code)) return isANYOF;
	  break;
	case OpCode._ALNUMC:
	  if(Character.isLetterOrDigit(code)) return isANYOF;
	  break;
	case OpCode._ALPHA:
	  if(Character.isLetter(code)) return isANYOF;
	  break;
	case OpCode._BLANK:
	  if(Character.isSpaceChar(code)) return isANYOF;
	  break;
	case OpCode._CNTRL:
	  if(Character.isISOControl(code)) return isANYOF;
	  break;
	case OpCode._LOWER:
	  if(Character.isLowerCase(code)) return isANYOF;
	  // Remove this hack after more efficient case-folding and unicode
	  // character classes are implemented
	  if(__caseInsensitive && Character.isUpperCase(code))
	    return isANYOF;
	  break;
	case OpCode._UPPER:
	  if(Character.isUpperCase(code)) return isANYOF;
	  // Remove this hack after more efficient case-folding and unicode
	  // character classes are implemented
	  if(__caseInsensitive && Character.isLowerCase(code))
	    return isANYOF;
	  break;
	case OpCode._PRINT:
	  if(Character.isSpaceChar(code)) return isANYOF;
          // Fall through to check if the character is alphanumeric,
	  // or a punctuation mark.  Printable characters are either
	  // alphanumeric, punctuation marks, or spaces.
	case OpCode._GRAPH:
	  if(Character.isLetterOrDigit(code)) return isANYOF;
          // Fall through to check if the character is a punctuation mark.
          // Graph characters are either alphanumeric or punctuation.
	case OpCode._PUNCT:
	  switch ( Character.getType(code) ) {
	    case Character.DASH_PUNCTUATION:
	    case Character.START_PUNCTUATION:
	    case Character.END_PUNCTUATION:
	    case Character.CONNECTOR_PUNCTUATION:
	    case Character.OTHER_PUNCTUATION:
            case Character.MATH_SYMBOL:
            case Character.CURRENCY_SYMBOL:
            case Character.MODIFIER_SYMBOL:
	      return isANYOF;
	    default:
	      break;
	    }
	  break;
	case OpCode._XDIGIT:
	  if( (code >= '0' && code <= '9') ||
	      (code >= 'a' && code <= 'f') ||
	      (code >= 'A' && code <= 'F')) return isANYOF;
	  break;
	case OpCode._ASCII:
	  if(code < 0x80)return isANYOF;
	}
      } 
    }
    return !isANYOF;
  }
  
  private boolean __tryExpression(int offset) {
    int count;
    
    __inputOffset = offset;
    __lastParen   = 0;
    __expSize     = 0;

    if(__numParentheses > 0) {
      for(count=0; count <= __numParentheses; count++) {
	__beginMatchOffsets[count] = OpCode._NULL_OFFSET;
	__endMatchOffsets[count]   = OpCode._NULL_OFFSET;
      }
    }

    if(__match(1)){
      __beginMatchOffsets[0] = offset;
      __endMatchOffsets[0]   = __inputOffset;
      return true;
    }

    return false;
  }
    

  private int __repeat(int offset, int max) {
    int scan, eol, operand, ret;
    char ch;
    char op;

    scan = __inputOffset;
    eol  = __eol;

    if(max != Character.MAX_VALUE && max < eol - scan)
      eol = scan + max;

    operand = OpCode._getOperand(offset);

    switch(op = __program[offset]) {

    case OpCode._ANY:
      while(scan < eol && __input[scan] != '\n')
	++scan;
      break;

    case OpCode._SANY:
      scan = eol;
      break;

    case OpCode._EXACTLY:
      ++operand;
      while(scan < eol && __program[operand] == __input[scan])
	++scan;
      break;

    case OpCode._ANYOF:
      if(scan < eol && (ch = __input[scan]) < 256) {
	while((ch < 256  ) && (__program[operand + (ch >> 4)] & (1 << (ch & 0xf))) == 0) {
	  if(++scan < eol)
	    ch = __input[scan];
	  else
	    break;
	}
      }
      break;

    case OpCode._ANYOFUN:
    case OpCode._NANYOFUN:
      if(scan < eol) {
	ch = __input[scan];
	while(__matchUnicodeClass(ch, __program, operand, op)){
	  if(++scan < eol)
	    ch = __input[scan];
	  else
	    break;
	}
      }
      break;

    case OpCode._ALNUM:
      while(scan < eol && OpCode._isWordCharacter(__input[scan]))
	++scan;
      break;

    case OpCode._NALNUM:
      while(scan < eol && !OpCode._isWordCharacter(__input[scan]))
	++scan;
      break;

    case OpCode._SPACE:
      while(scan < eol && Character.isWhitespace(__input[scan]))
	++scan;
      break;

    case OpCode._NSPACE:
      while(scan < eol && !Character.isWhitespace(__input[scan]))
	++scan;
      break;

    case OpCode._DIGIT:
      while(scan < eol && Character.isDigit(__input[scan]))
	++scan;
      break;

    case OpCode._NDIGIT:
      while(scan < eol && !Character.isDigit(__input[scan]))
	++scan;
      break;

    default:
      break;

    }

    ret = scan - __inputOffset;
    __inputOffset = scan;

    return ret;
  }


  private boolean __match(int offset) {
    char nextChar, op;
    int scan, next, input, maxScan, current, line, arg;
    boolean inputRemains = true, minMod = false;
    Perl5Repetition rep;


    input    = __inputOffset;
    inputRemains = (input < __endOffset);
    nextChar = (inputRemains ? __input[input] : __EOS);

    scan     = offset;
    maxScan  = __program.length;

    while(scan < maxScan /*&& scan > 0*/){
      next = OpCode._getNext(__program, scan);

      switch(op = __program[scan]) {

      case OpCode._BOL:
	if(input == __bol ? __previousChar == '\n' :
	   (__multiline && (inputRemains || input < __eol) && 
	    __input[input - 1] == '\n'))
	  break;
	return false;

      case OpCode._MBOL:
	if(input == __bol ? __previousChar == '\n' :
	   ((inputRemains || input < __eol) && __input[input - 1] == '\n'))
	  break;
	return false;

      case OpCode._SBOL:
	if(input == __bol && __previousChar == '\n')
	  break;
	return false;

      case OpCode._GBOL:
	if(input == __bol)
	  break;
	return true;

      case OpCode._EOL :
	if((inputRemains || input < __eol) && nextChar != '\n')
	  return false;
	if(!__multiline && __eol - input > 1)
	  return false;
	break;

      case OpCode._MEOL:
	if((inputRemains || input < __eol) && nextChar != '\n')
	  return false;
	break;

      case OpCode._SEOL:
	if((inputRemains || input < __eol) && nextChar != '\n')
	  return false;
	if(__eol - input > 1)
	  return false;
	break;

      case OpCode._SANY:
	if(!inputRemains && input >= __eol)
	  return false;
	inputRemains = (++input < __endOffset);
	nextChar = (inputRemains ? __input[input] : __EOS);
	break;

      case OpCode._ANY:
	if((!inputRemains && input >= __eol) || nextChar == '\n')
	  return false;
	inputRemains = (++input < __endOffset);
	nextChar = (inputRemains ? __input[input] : __EOS);
	break;

      case OpCode._EXACTLY:
	current = OpCode._getOperand(scan);
	line = __program[current++];

	if(__program[current] != nextChar)
	  return false;
	if(__eol - input < line)
	  return false;

	if(line > 1 && !__compare(__program, current, __input, input, line))
	  return false;