* src/gram.h, src/gram.c (pure_parser, glr_parser): Move to...

[bison.git] / src / scan-gram.l

diff --git a/src/scan-gram.l b/src/scan-gram.l
index a1fbac5ea7c65a47803a46bfaff87991d8c9e951..337f9db195872339e1bfb59be0f0b2bfc97c55b1 100644 (file)
--- a/src/scan-gram.l
+++ b/src/scan-gram.l
@@ -1,6 +1,6 @@
 /* Bison Grammar Scanner                             -*- C -*-
 
-   Copyright (C) 2002 Free Software Foundation, Inc.
+   Copyright (C) 2002, 2003 Free Software Foundation, Inc.
 
    This file is part of Bison, the GNU Compiler Compiler.
 
@@ -20,7 +20,7 @@
    02111-1307  USA
 */
 
-%option debug nodefault noyywrap never-interactive
+%option debug nodefault nounput noyywrap never-interactive
 %option prefix="gram_" outfile="lex.yy.c"
 
 %{
@@ -43,6 +43,7 @@
       scanner_cursor.file = current_file;              \
       scanner_cursor.line = 1;                         \
       scanner_cursor.column = 1;                       \
+      code_start = scanner_cursor;                     \
     }                                                  \
   while (0)
 
@@ -101,7 +102,7 @@ static void handle_dollar (int token_type, char *cp, location loc);
 static void handle_at (int token_type, char *cp, location loc);
 static void handle_syncline (char *args);
 static int convert_ucn_to_byte (char const *hex_text);
-static void unexpected_end_of_file (boundary, char const *);
+static void unexpected_eof (boundary, char const *);
 
 %}
 %x SC_COMMENT SC_LINE_COMMENT SC_YACC_COMMENT
@@ -136,14 +137,16 @@ splice     (\\[ \f\t\v]*\n)*
   int token_type IF_LINT (= 0);
 
   /* Location of most recent identifier, when applicable.  */
-  location id_loc IF_LINT (= *loc);
+  location id_loc IF_LINT (= empty_location);
 
-  /* Where containing code started, when applicable.  */
-  boundary code_start IF_LINT (= loc->start);
+  /* Where containing code started, when applicable.  Its initial
+     value is relevant only when yylex is invoked in the SC_EPILOGUE
+     start condition.  */
+  boundary code_start = scanner_cursor;
 
   /* Where containing comment or string or character literal started,
      when applicable.  */
-  boundary token_start IF_LINT (= loc->start);
+  boundary token_start IF_LINT (= scanner_cursor);
 %}
 
 
@@ -195,6 +198,7 @@ splice       (\\[ \f\t\v]*\n)*
   "%name"[-_]"prefix"     return PERCENT_NAME_PREFIX;
   "%no"[-_]"lines"        return PERCENT_NO_LINES;
   "%nonassoc"             return PERCENT_NONASSOC;
+  "%nondeterministic-parser"   return PERCENT_NONDETERMINISTIC_PARSER;
   "%nterm"                return PERCENT_NTERM;
   "%output"               return PERCENT_OUTPUT;
   "%parse-param"         token_type = PERCENT_PARSE_PARAM; BEGIN SC_PRE_CODE;
@@ -270,16 +274,18 @@ splice     (\\[ \f\t\v]*\n)*
   "%%" {
     static int percent_percent_count;
     if (++percent_percent_count == 2)
-      {
-       code_start = loc->start;
-       BEGIN SC_EPILOGUE;
-      }
+      BEGIN SC_EPILOGUE;
     return PERCENT_PERCENT;
   }
 
   . {
     complain_at (*loc, _("invalid character: %s"), quote (yytext));
   }
+
+  <<EOF>> {
+    loc->start = loc->end = scanner_cursor;
+    yyterminate ();
+  }
 }
 
 
@@ -318,7 +324,7 @@ splice       (\\[ \f\t\v]*\n)*
 {
   "*/"     BEGIN context_state;
   .|\n    ;
-  <<EOF>>  unexpected_end_of_file (token_start, "*/");
+  <<EOF>>  unexpected_eof (token_start, "*/"); BEGIN context_state;
 }
 
 
@@ -329,7 +335,7 @@ splice       (\\[ \f\t\v]*\n)*
 <SC_COMMENT>
 {
   "*"{splice}"/"  STRING_GROW; BEGIN context_state;
-  <<EOF>>        unexpected_end_of_file (token_start, "*/");
+  <<EOF>>        unexpected_eof (token_start, "*/"); BEGIN context_state;
 }
 
 
@@ -363,7 +369,7 @@ splice       (\\[ \f\t\v]*\n)*
   }
 
   .|\n     STRING_GROW;
-  <<EOF>>   unexpected_end_of_file (token_start, "\"");
+  <<EOF>>   unexpected_eof (token_start, "\""); BEGIN INITIAL;
 }
 
   /*---------------------------------------------------------------.
@@ -389,7 +395,7 @@ splice       (\\[ \f\t\v]*\n)*
   }
 
   .|\n     STRING_GROW;
-  <<EOF>>   unexpected_end_of_file (token_start, "'");
+  <<EOF>>   unexpected_eof (token_start, "'"); BEGIN INITIAL;
 }
 
 
@@ -451,7 +457,7 @@ splice       (\\[ \f\t\v]*\n)*
 {
   "'"                  STRING_GROW; BEGIN context_state;
   \\{splice}[^$@\[\]]  STRING_GROW;
-  <<EOF>>              unexpected_end_of_file (token_start, "'");
+  <<EOF>>              unexpected_eof (token_start, "'"); BEGIN context_state;
 }
 
 
@@ -464,7 +470,10 @@ splice      (\\[ \f\t\v]*\n)*
 {
   "\""                 STRING_GROW; BEGIN context_state;
   \\{splice}[^$@\[\]]  STRING_GROW;
-  <<EOF>>              unexpected_end_of_file (token_start, "\"");
+  <<EOF>> {
+    unexpected_eof (token_start, "\"");
+    BEGIN context_state;
+  }
 }
 
 
@@ -530,6 +539,8 @@ splice       (\\[ \f\t\v]*\n)*
        return token_type;
       }
   }
+
+  <<EOF>>  unexpected_eof (scanner_cursor, "{}"); BEGIN INITIAL;
 }
 
 
@@ -559,7 +570,7 @@ splice       (\\[ \f\t\v]*\n)*
        should also diagnose other Bison extensions like %yacc.
        Perhaps there should also be a GCC-style --pedantic-errors
        option, so that such warnings are diagnosed as errors.  */
-    if (outer_brace && ! yacc_flag)
+    if (outer_brace && token_type == BRACED_CODE && ! yacc_flag)
       obstack_1grow (&obstack_for_string, ';');
 
     obstack_1grow (&obstack_for_string, '}');
@@ -582,7 +593,7 @@ splice       (\\[ \f\t\v]*\n)*
   "$"("<"{tag}">")?(-?[0-9]+|"$")  handle_dollar (token_type, yytext, *loc);
   "@"(-?[0-9]+|"$")               handle_at (token_type, yytext, *loc);
 
-  <<EOF>>  unexpected_end_of_file (code_start, "}");
+  <<EOF>>  unexpected_eof (code_start, "}"); BEGIN INITIAL;
 }
 
 
@@ -600,7 +611,7 @@ splice       (\\[ \f\t\v]*\n)*
     return PROLOGUE;
   }
 
-  <<EOF>>  unexpected_end_of_file (code_start, "%}");
+  <<EOF>>  unexpected_eof (code_start, "%}"); BEGIN INITIAL;
 }
 
 
@@ -682,14 +693,14 @@ adjust_location (location *loc, char const *token, size_t size)
 static size_t
 no_cr_read (FILE *fp, char *buf, size_t size)
 {
-  size_t s = fread (buf, 1, size, fp);
-  if (s)
+  size_t bytes_read = fread (buf, 1, size, fp);
+  if (bytes_read)
     {
-      char *w = memchr (buf, '\r', s);
+      char *w = memchr (buf, '\r', bytes_read);
       if (w)
        {
          char const *r = ++w;
-         char const *lim = buf + s;
+         char const *lim = buf + bytes_read;
 
          for (;;)
            {
@@ -718,7 +729,7 @@ no_cr_read (FILE *fp, char *buf, size_t size)
        }
     }
 
-  return s;
+  return bytes_read;
 }
 
 
@@ -964,25 +975,15 @@ handle_syncline (char *args)
 /*------------------------------------------------------------------------.
 | Report an unexpected EOF in a token or comment starting at START.       |
 | An end of file was encountered and the expected TOKEN_END was missing.  |
-| After reporting the problem, pretend that TOKEN_END was found.          |
 `------------------------------------------------------------------------*/
 
 static void
-unexpected_end_of_file (boundary start, char const *token_end)
+unexpected_eof (boundary start, char const *token_end)
 {
-  size_t i = strlen (token_end);
-
   location loc;
   loc.start = start;
   loc.end = scanner_cursor;
   complain_at (loc, _("missing `%s' at end of file"), token_end);
-
-  /* Adjust scanner cursor so that any later message does not count
-     the characters about to be inserted.  */
-  scanner_cursor.column -= i;
-
-  while (i != 0)
-    unput (token_end[--i]);
 }