lexer.js

一个类似windows

		      buf.append (File.byteToString (ch));
		      ch = stream.readByte ();
		    }
		}

	      if (ch == #'e' || ch == #'E')
		{
		  buf.append (File.byteToString (ch));
		  ch = stream.readByte ();
		  if (ch == #'+' || ch == #'-')
		    {
		      buf.append (File.byteToString (ch));
		      ch = stream.readByte ();
		    }
		  if (!JSC$lexer_is_decimal_digit (ch))
		    error (JSC$filename + ":" + JSC$linenum.toString ()
			   + ": malformed exponent part in a decimal literal");

		  while (JSC$lexer_is_decimal_digit (ch))
		    {
		      buf.append (File.byteToString (ch));
		      ch = stream.readByte ();
		    }
		}
	    }

	  /* Finally, we put the last character pack to the stream. */
	  stream.ungetByte (ch);

	  if (is_float)
	    {
	      JSC$token_value = parseFloat (buf);
	      return JSC$tFLOAT;
	    }

	  JSC$token_value = parseInt (buf);
	  return JSC$tINTEGER;
	}

      /* Just return the character as-is. */
      else
	return ch;
    }

  /* EOF reached. */
  return JSC$tEOF;
}


/*
 * Help functions.
 */

function JSC$lexer_peek_char (stream)
{
  var ch2 = stream.readByte ();
  stream.ungetByte (ch2);

  return ch2;
}


function JSC$lexer_is_identifier_letter (ch)
{
  return ((#'a' <= ch && ch <= #'z') || (#'A' <= ch && ch <= #'Z')
	  || ch == #'$' || ch == #'_');
}


function JSC$lexer_is_octal_digit (ch)
{
  return (#'0' <= ch && ch <= #'7');
}


function JSC$lexer_is_decimal_digit (ch)
{
  return #'0' <= ch && ch <= #'9';
}


function JSC$lexer_is_hex_digit (ch)
{
  return ((#'0' <= ch && ch <= #'9')
	  || (#'a' <= ch && ch <= #'f')
	  || (#'A' <= ch && ch <= #'F'));
}


function JSC$lexer_is_white_space (ch)
{
  return (ch == #' ' || ch == #'\t' || ch == #'\v' || ch == #'\r'
	  || ch == #'\f' || ch == #'\n');
}


function JSC$lexer_hex_to_dec (ch)
{
  return ((#'0' <= ch && ch <= #'9')
	  ? ch - #'0'
	  : ((#'a' <= ch && ch <= #'f')
	     ? 10 + ch - #'a'
	     : 10 + ch - #'A'));
}


function JSC$lexer_read_backslash_escape (stream, possible_start, name)
{
  var ch = stream.readByte ();

  if (ch == #'n')
    ch = #'\n';
  else if (ch == #'t')
    ch = #'\t';
  else if (ch == #'v')
    ch = #'\v';
  else if (ch == #'b')
    ch = #'\b';
  else if (ch == #'r')
    ch = #'\r';
  else if (ch == #'f')
    ch = #'\f';
  else if (ch == #'a')
    ch = #'\a';
  else if (ch == #'\\')
    ch = #'\\';
  else if (ch == #'?')
    ch = #'?';
  else if (ch == #'\'')
    ch = #'\'';
  else if (ch == #'"')
    ch = #'"';
  else if (ch == #'x')
    {
      /* HexEscapeSequence. */
      var c1, c2;

      c1 = stream.readByte ();
      c2 = stream.readByte ();

      if (c1 == -1 || c2 == -1)
	JSC$lexer_eof_in_constant (possible_start, name);

      if (!JSC$lexer_is_hex_digit (c1) || !JSC$lexer_is_hex_digit (c2))
	error (JSC$filename + ":" + JSC$linenum.toString ()
	       + ": \\x used with no following hex digits");

      ch = (JSC$lexer_hex_to_dec (c1) << 4) + JSC$lexer_hex_to_dec (c2);
    }
  else if (ch == #'u')
    {
      /* UnicodeEscapeSequence. */
      var c1, c2, c3, c4;

      c1 = stream.readByte ();
      c2 = stream.readByte ();
      c3 = stream.readByte ();
      c4 = stream.readByte ();

      if (c1 == -1 || c2 == -1 || c3 == -1 || c4 == -1)
	JSC$lexer_eof_in_constant (possible_start, name);

      if (!JSC$lexer_is_hex_digit (c1) || !JSC$lexer_is_hex_digit (c2)
	  || !JSC$lexer_is_hex_digit (c3) || !JSC$lexer_is_hex_digit (c4))
	error (JSC$filename + ":" + JSC$linenum.toString ()
	       + ": \\u used with no following hex digits");

      ch = ((JSC$lexer_hex_to_dec (c1) << 12)
	    + (JSC$lexer_hex_to_dec (c2) << 8)
	    + (JSC$lexer_hex_to_dec (c3) << 4)
	    + JSC$lexer_hex_to_dec (c4));
    }
  else if (JSC$lexer_is_octal_digit (ch))
    {
      var result = ch - #'0';
      var i = 1;

      if (ch == #'0')
	/* Allow three octal digits after '0'. */
	i = 0;

      ch = stream.readByte ();
      while (i < 3 && JSC$lexer_is_octal_digit (ch))
	{
	  result *= 8;
	  result += ch - #'0';
	  ch = stream.readByte ();
	  i++;
	}
      stream.ungetByte (ch);
      ch = result;
    }
  else
    {
      if (ch == -1)
	error (JSC$filename + ":" + JSC$linenum.toString ()
	       + ": unterminated " + name);

      JSC$warning (JSC$filename + ":" + JSC$linenum.toString ()
		   + ": warning: unknown escape sequence `\\"
		   + File.byteToString (ch) + "'");
    }

  return ch;
}


function JSC$lexer_read_string (stream, name, ender)
{
  var str = new String ("");
  var done = false, ch;
  var possible_start_ln = JSC$linenum;
  var warned_line_terminator = false;

  while (!done)
    {
      ch = stream.readByte ();
      if (ch == #'\n')
	{
	  if (JSC$warn_strict_ecma && !warned_line_terminator)
	    {
	      JSC$warning (JSC$filename + ":" + JSC$linenum.toString ()
			   + ": warning: ECMAScript don't allow line terminators in "
			   + name + " constants");
	      warned_line_terminator = true;
	    }
	  JSC$linenum++;
	}

      if (ch == -1)
	JSC$lexer_eof_in_constant (possible_start_ln, name);

      else if (ch == ender)
	done = true;
      else
	{
	  if (ch == #'\\')
	    {
	      if (JSC$lexer_peek_char (stream) == #'\n')
		{
		  /*
		   * Backslash followed by a newline character.  Ignore
		   * them both.
		   */
		  stream.readByte ();
		  JSC$linenum++;
		  continue;
		}
	      ch = JSC$lexer_read_backslash_escape (stream, possible_start_ln,
						    name);
	    }
	  str.append (ch);
	}
    }

  return str;
}


function JSC$lexer_read_regexp_constant (stream)
{
  /* Regexp literal. */
  var source = JSC$lexer_read_regexp_source (stream);

  /* Check the possible flags. */
  var flags = new String ("");
  while ((ch = JSC$lexer_peek_char (stream)) == #'g' || ch == #'i')
    {
      stream.readByte ();
      flags.append (File.byteToString (ch));
    }

  /* Try to compile it. */
  var msg = false;
  var result;
  try
    {
      result = new RegExp (source, flags);
    }
  catch (msg)
    {
      var start = msg.lastIndexOf (":");
      msg = (JSC$filename + ":" + JSC$token_linenum.toString ()
	     + ": malformed regular expression constant:"
	     + msg.substr (start + 1));
    }
  if (msg)
    error (msg);

  /* Success. */

  return result;
}


function JSC$lexer_read_regexp_source (stream)
{
  var str = new String ("");
  var done = false, ch;
  var possible_start_ln = JSC$linenum;
  var warned_line_terminator = false;
  var name = "regular expression";

  while (!done)
    {
      ch = stream.readByte ();
      if (ch == #'\n')
	{
	  if (JSC$warn_strict_ecma && !warned_line_terminator)
	    {
	      JSC$warning (JSC$filename + ":" + JSC$linenum.toString ()
			   + ": warning: ECMAScript don't allow line "
			   + "terminators in " + name + " constants");
	      warned_line_terminator = true;
	    }
	  JSC$linenum++;
	}

      if (ch == -1)
	JSC$lexer_eof_in_constant (possible_start_ln, name);

      else if (ch == #'/')
	done = true;
      else
	{
	  if (ch == #'\\')
	    {
	      ch = stream.readByte ();
	      if (ch == #'\n')
		{
		  /*
		   * Backslash followed by a newline character.  Ignore
		   * them both.
		   */
		  JSC$linenum++;
		  continue;
		}
	      if (ch == -1)
		JSC$lexer_eof_in_constant (possible_start_ln, name);

	      /* Handle the backslash escapes. */
	      if (ch == #'f')
		ch = #'\f';
	      else if (ch == #'n')
		ch = #'\n';
	      else if (ch == #'r')
		ch = #'\r';
	      else if (ch == #'t')
		ch = #'\t';
	      else if (ch == #'v')
		ch == #'\v';
	      else if (ch == #'c')
		{
		  /* SourceCharacter. */
		  ch = stream.readByte ();
		  if (ch == -1)
		    JSC$lexer_eof_in_constant (possible_start_ln, name);

		  if (ch == #'\n' && JSC$warn_strict_ecma)
		    JSC$warning (JSC$filename + ":" + JSC$linenum.toString ()
				 + ": warning: ECMAScript don't allow line termiantor after \\c in regular expression constants");

		  /*
		   * Append the source-character escape start.  The ch
		   * will be appended later.
		   */
		  str.append ("\\c");
		}
	      else if (ch == #'u' || ch == #'x' || ch == #'0')
		{
		  /* These can be handled with the read_backslash_escape(). */
		  stream.ungetByte (ch);
		  ch = JSC$lexer_read_backslash_escape (stream);
		}
	      else
		{
		  /*
		   * Nothing special.  Leave it to the result as-is.
		   * The regular expression backage will handle it.
		   */
		  stream.ungetByte (ch);
		  ch = #'\\';
		}
	    }
	  str.append (File.byteToString (ch));
	}
    }

  return str;
}


function JSC$lexer_eof_in_constant (possible_start, name)
{
  var msg = (JSC$filename + ":" + JSC$linenum.toString ()
	     + ": unterminated " + name + " constant");

  if (possible_start > 0)
    msg += (System.lineBreakSequence
	    + JSC$filename + ":" + possible_start.toString ()
	    + ": possible real start of unterminated " + name + " constant");

  error (msg);
}


/*
Local variables:
mode: c
End:
*/