#pragma parsing

[Martin von Loewis]

When parsing

#pragma ident "Hello, world"

a string is accessed as an identifier. This patch fixes that.
Alternatively, token_buffer could be used to get the string
representing the current token.

Regards,
Martin

1998-10-27  Martin von Löwis  <loewis@informatik.hu-berlin.de>

	* lex.c (handle_generic_pragma): Retrieve string value for string
	and integer constants.

Index: lex.c
===================================================================
RCS file: /egcs/carton/cvsfiles/egcs/gcc/cp/lex.c,v
retrieving revision 1.81
diff -c -p -r1.81 lex.c
*** lex.c	1998/10/23 14:53:08	1.81
--- lex.c	1998/10/27 12:17:32
*************** handle_generic_pragma (token)
*** 4976,4985 ****
  	{
  	case IDENTIFIER:
  	case TYPENAME:
- 	case STRING:
- 	case CONSTANT:
  	  handle_pragma_token (IDENTIFIER_POINTER(yylval.ttype), yylval.ttype);
  	  break;
  	case '(':
  	  handle_pragma_token ("(", NULL_TREE);
  	  break;
--- 4976,4990 ----
  	{
  	case IDENTIFIER:
  	case TYPENAME:
  	  handle_pragma_token (IDENTIFIER_POINTER(yylval.ttype), yylval.ttype);
  	  break;
+         case STRING:
+           handle_pragma_token (TREE_STRING_POINTER (yylval.ttype),
+                                yylval.ttype);
+           break;
+         case CONSTANT:
+           handle_pragma_token ("<number>", yylval.ttype);
+           break;
  	case '(':
  	  handle_pragma_token ("(", NULL_TREE);
  	  break;