cpplib: Initial patch towards each token storing its location

[Neil Booth <neil at daikokuya dot demon dot co dot uk>]

The code is not used yet, and the old code still exists.  Next I will
add a

   logical_line -> (physical_file, physical_line)

mapping, and gradually convert existing code to use it rather than the
current method, starting with diagnostic output.

Bootstrapped and made check x86 Linux.

Neil.

	* cpphash.h (struct cpp_reader): New members line, pseudo_newlines.
	* cpplex.c (handle_newline): Update prototype.  Maintain logical
	line number.
	(skip_escaped_newlines, skip_block_comment, parse_string):
	Update accordingly.
	(_cpp_lex_token): Update, and store token position within the token.
	* cpplib.h (struct cpp_token): Add line and column entries.
	* cppmacro.c (replace_args): Position stringified tokens correctly.

Index: cpphash.h
===================================================================
RCS file: /cvs/gcc/gcc/gcc/cpphash.h,v
retrieving revision 1.108
diff -u -p -r1.108 cpphash.h
--- cpphash.h	2001/07/29 22:27:20	1.108
+++ cpphash.h	2001/07/30 18:50:53
@@ -246,6 +246,12 @@ struct cpp_reader
   /* Lexer state.  */
   struct lexer_state state;
 
+  /* Source line tracking.  Subtract pseudo_newlines from the actual
+     line number to get the line number of preprocessed output.  Used
+     for escaped newlines and macro args that cross multiple lines.  */
+  unsigned int line;
+  unsigned int pseudo_newlines;
+
   /* The position of the last lexed token and last lexed directive.  */
   cpp_lexer_pos lexer_pos;
   cpp_lexer_pos directive_pos;
Index: cpplex.c
===================================================================
RCS file: /cvs/gcc/gcc/gcc/cpplex.c,v
retrieving revision 1.147
diff -u -p -r1.147 cpplex.c
--- cpplex.c	2001/07/29 17:27:56	1.147
+++ cpplex.c	2001/07/30 18:51:02
@@ -80,7 +80,7 @@ const struct token_spelling token_spelli
 #define TOKEN_SPELL(token) (token_spellings[(token)->type].category)
 #define TOKEN_NAME(token) (token_spellings[(token)->type].name)
 
-static cppchar_t handle_newline PARAMS ((cpp_buffer *, cppchar_t));
+static cppchar_t handle_newline PARAMS ((cpp_reader *, cppchar_t));
 static cppchar_t skip_escaped_newlines PARAMS ((cpp_buffer *, cppchar_t));
 static cppchar_t get_effective_char PARAMS ((cpp_buffer *));
 
@@ -124,12 +124,17 @@ cpp_ideq (token, string)
 /* Call when meeting a newline.  Returns the character after the newline
    (or carriage-return newline combination), or EOF.  */
 static cppchar_t
-handle_newline (buffer, newline_char)
-     cpp_buffer *buffer;
+handle_newline (pfile, newline_char)
+     cpp_reader *pfile;
      cppchar_t newline_char;
 {
+  cpp_buffer *buffer;
   cppchar_t next = EOF;
 
+  pfile->line++;
+  pfile->pseudo_newlines++;
+
+  buffer = pfile->buffer;
   buffer->col_adjust = 0;
   buffer->lineno++;
   buffer->line_base = buffer->cur;
@@ -264,7 +269,7 @@ skip_escaped_newlines (buffer, next)
 	    cpp_warning (buffer->pfile,
 			 "backslash and newline separated by space");
 
-	  next = handle_newline (buffer, next1);
+	  next = handle_newline (buffer->pfile, next1);
 	  if (next == EOF)
 	    cpp_pedwarn (buffer->pfile, "backslash-newline at end of file");
 	}
@@ -348,7 +353,7 @@ skip_block_comment (pfile)
 	}
       else if (is_vspace (c))
 	{
-	  prevc = c, c = handle_newline (buffer, c);
+	  prevc = c, c = handle_newline (pfile, c);
 	  goto next_char;
 	}
       else if (c == '\t')
@@ -706,7 +711,7 @@ parse_string (pfile, token, terminator)
 	  if (pfile->mlstring_pos.line == 0)
 	    pfile->mlstring_pos = pfile->lexer_pos;
 	      
-	  c = handle_newline (buffer, c);
+	  c = handle_newline (pfile, c);
 	  *dest++ = '\n';
 	  goto have_char;
 	}
@@ -866,6 +871,7 @@ _cpp_lex_token (pfile, result)
   buffer->saved_flags = 0;
  next_char:
   pfile->lexer_pos.line = buffer->lineno;
+  result->line = pfile->line;
  next_char2:
   pfile->lexer_pos.col = CPP_BUF_COLUMN (buffer, buffer->cur);
 
@@ -875,6 +881,7 @@ _cpp_lex_token (pfile, result)
       c = *buffer->cur++;
       pfile->lexer_pos.col++;
     }
+  result->col = pfile->lexer_pos.col;
 
  do_switch:
   buffer->read_ahead = EOF;
@@ -901,7 +908,9 @@ _cpp_lex_token (pfile, result)
     case '\n': case '\r':
       if (!pfile->state.in_directive)
 	{
-	  handle_newline (buffer, c);
+	  handle_newline (pfile, c);
+	  if (!pfile->state.parsing_args)
+	    pfile->pseudo_newlines = 0;
 	  bol = 1;
 	  pfile->lexer_pos.output_line = buffer->lineno;
 	  /* This is a new line, so clear any white space flag.
Index: cpplib.h
===================================================================
RCS file: /cvs/gcc/gcc/gcc/cpplib.h,v
retrieving revision 1.178
diff -u -p -r1.178 cpplib.h
--- cpplib.h	2001/07/29 22:27:20	1.178
+++ cpplib.h	2001/07/30 18:51:02
@@ -172,6 +172,8 @@ struct cpp_string
    occupy 12 bytes on 32-bit hosts and 16 bytes on 64-bit hosts.  */
 struct cpp_token
 {
+  unsigned int line;		/* Logical line of first char of token.  */
+  unsigned short col;		/* Column of first char of token.  */
   ENUM_BITFIELD(cpp_ttype) type : CHAR_BIT;  /* token type */
   unsigned char flags;		/* flags - see above */
 
Index: cppmacro.c
===================================================================
RCS file: /cvs/gcc/gcc/gcc/cppmacro.c,v
retrieving revision 1.55
diff -u -p -r1.55 cppmacro.c
--- cppmacro.c	2001/07/29 17:27:57	1.55
+++ cppmacro.c	2001/07/30 18:51:16
@@ -761,7 +761,12 @@ replace_args (pfile, macro, args, list)
 
 	arg = &args[src->val.arg_no - 1];
 	if (src->flags & STRINGIFY_ARG)
-	  from = arg->stringified, count = 1;
+	  {
+	    from = arg->stringified, count = 1;
+	    /* Ugh.  Maintain position of original argument.  */
+	    arg->stringified->line = src->line;
+	    arg->stringified->col = src->col;
+	  }
 	else if (src->flags & PASTE_LEFT)
 	  count = arg->count, from = arg->first;
 	else if (src > macro->expansion && (src[-1].flags & PASTE_LEFT))
@@ -923,6 +928,7 @@ cpp_get_token (pfile, token)
 	  /* PASTE_LEFT tokens can only appear in macro expansions.  */
 	  if (token->flags & PASTE_LEFT)
 	    {
+	      /* Maintains position of original token.  */
 	      paste_all_tokens (pfile, token);
 	      pfile->buffer->saved_flags = AVOID_LPASTE;
 	    }
@@ -957,6 +963,7 @@ cpp_get_token (pfile, token)
 
 	  if (node->flags & NODE_BUILTIN)
 	    {
+	      /* Maintains position of original token.  */
 	      builtin_macro (pfile, token);
 	      pfile->buffer->saved_flags = AVOID_LPASTE;
 	      break;