* NEWS: Bison now warns if it finds a stray `$' or `@' in an action.

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

diff --git a/src/scan-gram.l b/src/scan-gram.l
index a2c61fd67e9ea92fb6dcbd59aff19aef47ef5b93..3c7d37c484d5932102804351b097cce301058175 100644 (file)
--- a/src/scan-gram.l
+++ b/src/scan-gram.l
@@ -1,6 +1,6 @@
 /* Bison Grammar Scanner                             -*- C -*-
 
 /* Bison Grammar Scanner                             -*- C -*-
 

-   Copyright (C) 2002 Free Software Foundation, Inc.
+   Copyright (C) 2002, 2003, 2004, 2005 Free Software Foundation, Inc.

 
    This file is part of Bison, the GNU Compiler Compiler.
 
 
    This file is part of Bison, the GNU Compiler Compiler.
 


@@ -16,14 +16,19 @@
 
    You should have received a copy of the GNU General Public License
    along with this program; if not, write to the Free Software
 
    You should have received a copy of the GNU General Public License
    along with this program; if not, write to the Free Software

-   Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
-   02111-1307  USA
+   Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+   02110-1301  USA

 */
 
 */
 

-%option debug nodefault noyywrap never-interactive
+%option debug nodefault nounput noyywrap never-interactive

 %option prefix="gram_" outfile="lex.yy.c"
 
 %{
 %option prefix="gram_" outfile="lex.yy.c"
 
 %{

+/* Work around a bug in flex 2.5.31.  See Debian bug 333231
+   <http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=333231>.  */
+#undef gram_wrap
+#define gram_wrap() 1
+

 #include "system.h"
 
 #include <mbswidth.h>
 #include "system.h"
 
 #include <mbswidth.h>


@@ -34,6 +39,7 @@
 #include "files.h"
 #include "getargs.h"
 #include "gram.h"
 #include "files.h"
 #include "getargs.h"
 #include "gram.h"

+#include "quotearg.h"

 #include "reader.h"
 #include "uniqstr.h"
 
 #include "reader.h"
 #include "uniqstr.h"
 


@@ -43,9 +49,23 @@
       scanner_cursor.file = current_file;              \
       scanner_cursor.line = 1;                         \
       scanner_cursor.column = 1;                       \
       scanner_cursor.file = current_file;              \
       scanner_cursor.line = 1;                         \
       scanner_cursor.column = 1;                       \

+      code_start = scanner_cursor;                     \

     }                                                  \
   while (0)
 
     }                                                  \
   while (0)
 

+/* Pacify "gcc -Wmissing-prototypes" when flex 2.5.31 is used.  */
+int gram_get_lineno (void);
+FILE *gram_get_in (void);
+FILE *gram_get_out (void);
+int gram_get_leng (void);
+char *gram_get_text (void);
+void gram_set_lineno (int);
+void gram_set_in (FILE *);
+void gram_set_out (FILE *);
+int gram_get_debug (void);
+void gram_set_debug (int);
+int gram_lex_destroy (void);
+

 /* Location of scanner cursor.  */
 boundary scanner_cursor;
 
 /* Location of scanner cursor.  */
 boundary scanner_cursor;
 


@@ -100,8 +120,10 @@ static int rule_length;
 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 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 unsigned long int scan_integer (char const *p, int base, location loc);

 static int convert_ucn_to_byte (char const *hex_text);
 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 *);
+static void unexpected_newline (boundary, char const *);

 
 %}
 %x SC_COMMENT SC_LINE_COMMENT SC_YACC_COMMENT
 
 %}
 %x SC_COMMENT SC_LINE_COMMENT SC_YACC_COMMENT


@@ -136,14 +158,16 @@ splice     (\\[ \f\t\v]*\n)*
   int token_type IF_LINT (= 0);
 
   /* Location of most recent identifier, when applicable.  */
   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.  */
 
   /* 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);

 %}
 
 
 %}
 
 


@@ -153,10 +177,9 @@ splice      (\\[ \f\t\v]*\n)*
 
 <INITIAL,SC_AFTER_IDENTIFIER,SC_PRE_CODE>
 {
 
 <INITIAL,SC_AFTER_IDENTIFIER,SC_PRE_CODE>
 {

-  [ \f\n\t\v]  ;
+  /* Comments and white space.  */

   ","         warn_at (*loc, _("stray `,' treated as white space"));
   ","         warn_at (*loc, _("stray `,' treated as white space"));

-
-  /* Comments. */
+  [ \f\n\t\v]  |

   "//".*       ;
   "/*" {
     token_start = loc->start;
   "//".*       ;
   "/*" {
     token_start = loc->start;


@@ -179,28 +202,34 @@ splice     (\\[ \f\t\v]*\n)*
 {
   "%binary"               return PERCENT_NONASSOC;
   "%debug"                return PERCENT_DEBUG;
 {
   "%binary"               return PERCENT_NONASSOC;
   "%debug"                return PERCENT_DEBUG;

+  "%default"[-_]"prec"    return PERCENT_DEFAULT_PREC;

   "%define"               return PERCENT_DEFINE;
   "%defines"              return PERCENT_DEFINES;
   "%destructor"                  token_type = PERCENT_DESTRUCTOR; BEGIN SC_PRE_CODE;
   "%dprec"               return PERCENT_DPREC;
   "%error"[-_]"verbose"   return PERCENT_ERROR_VERBOSE;
   "%expect"               return PERCENT_EXPECT;
   "%define"               return PERCENT_DEFINE;
   "%defines"              return PERCENT_DEFINES;
   "%destructor"                  token_type = PERCENT_DESTRUCTOR; BEGIN SC_PRE_CODE;
   "%dprec"               return PERCENT_DPREC;
   "%error"[-_]"verbose"   return PERCENT_ERROR_VERBOSE;
   "%expect"               return PERCENT_EXPECT;

+  "%expect"[-_]"rr"      return PERCENT_EXPECT_RR;

   "%file-prefix"          return PERCENT_FILE_PREFIX;
   "%fixed"[-_]"output"[-_]"files"   return PERCENT_YACC;
   "%file-prefix"          return PERCENT_FILE_PREFIX;
   "%fixed"[-_]"output"[-_]"files"   return PERCENT_YACC;

+  "%initial-action"       token_type = PERCENT_INITIAL_ACTION; BEGIN SC_PRE_CODE;

   "%glr-parser"           return PERCENT_GLR_PARSER;
   "%left"                 return PERCENT_LEFT;
   "%lex-param"           token_type = PERCENT_LEX_PARAM; BEGIN SC_PRE_CODE;
   "%locations"            return PERCENT_LOCATIONS;
   "%merge"               return PERCENT_MERGE;
   "%name"[-_]"prefix"     return PERCENT_NAME_PREFIX;
   "%glr-parser"           return PERCENT_GLR_PARSER;
   "%left"                 return PERCENT_LEFT;
   "%lex-param"           token_type = PERCENT_LEX_PARAM; BEGIN SC_PRE_CODE;
   "%locations"            return PERCENT_LOCATIONS;
   "%merge"               return PERCENT_MERGE;
   "%name"[-_]"prefix"     return PERCENT_NAME_PREFIX;

+  "%no"[-_]"default"[-_]"prec" return PERCENT_NO_DEFAULT_PREC;

   "%no"[-_]"lines"        return PERCENT_NO_LINES;
   "%nonassoc"             return PERCENT_NONASSOC;
   "%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;
   "%prec"                 rule_length--; return PERCENT_PREC;
   "%printer"              token_type = PERCENT_PRINTER; BEGIN SC_PRE_CODE;
   "%pure"[-_]"parser"     return PERCENT_PURE_PARSER;
   "%nterm"                return PERCENT_NTERM;
   "%output"               return PERCENT_OUTPUT;
   "%parse-param"         token_type = PERCENT_PARSE_PARAM; BEGIN SC_PRE_CODE;
   "%prec"                 rule_length--; return PERCENT_PREC;
   "%printer"              token_type = PERCENT_PRINTER; BEGIN SC_PRE_CODE;
   "%pure"[-_]"parser"     return PERCENT_PURE_PARSER;

+  "%require"              return PERCENT_REQUIRE;

   "%right"                return PERCENT_RIGHT;
   "%skeleton"             return PERCENT_SKELETON;
   "%start"                return PERCENT_START;
   "%right"                return PERCENT_RIGHT;
   "%skeleton"             return PERCENT_SKELETON;
   "%start"                return PERCENT_START;


@@ -228,15 +257,11 @@ splice     (\\[ \f\t\v]*\n)*
   }
 
   {int} {
   }
 
   {int} {

-    unsigned long num;
-    set_errno (0);
-    num = strtoul (yytext, 0, 10);
-    if (INT_MAX < num || get_errno ())
-      {
-       complain_at (*loc, _("integer out of range: %s"), quote (yytext));
-       num = INT_MAX;
-      }
-    val->integer = num;
+    val->integer = scan_integer (yytext, 10, *loc);
+    return INT;
+  }
+  0[xX][0-9abcdefABCDEF]+ {
+    val->integer = scan_integer (yytext, 16, *loc);

     return INT;
   }
 
     return INT;
   }
 


@@ -244,7 +269,7 @@ splice       (\\[ \f\t\v]*\n)*
   "'"        STRING_GROW; token_start = loc->start; BEGIN SC_ESCAPED_CHARACTER;
 
   /* Strings. */
   "'"        STRING_GROW; token_start = loc->start; BEGIN SC_ESCAPED_CHARACTER;
 
   /* Strings. */

-  "\""       STRING_GROW; token_start = loc->start; BEGIN SC_ESCAPED_STRING;
+  "\""       token_start = loc->start; BEGIN SC_ESCAPED_STRING;

 
   /* Prologue. */
   "%{"        code_start = loc->start; BEGIN SC_PROLOGUE;
 
   /* Prologue. */
   "%{"        code_start = loc->start; BEGIN SC_PROLOGUE;


@@ -270,16 +295,18 @@ splice     (\\[ \f\t\v]*\n)*
   "%%" {
     static int percent_percent_count;
     if (++percent_percent_count == 2)
   "%%" {
     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));
   }
     return PERCENT_PERCENT;
   }
 
   . {
     complain_at (*loc, _("invalid character: %s"), quote (yytext));
   }

+
+  <<EOF>> {
+    loc->start = loc->end = scanner_cursor;
+    yyterminate ();
+  }

 }
 
 
 }
 
 


@@ -318,7 +345,7 @@ splice       (\\[ \f\t\v]*\n)*
 {
   "*/"     BEGIN context_state;
   .|\n    ;
 {
   "*/"     BEGIN context_state;
   .|\n    ;

-  <<EOF>>  unexpected_end_of_file (token_start, "*/");
+  <<EOF>>  unexpected_eof (token_start, "*/"); BEGIN context_state;

 }
 
 
 }
 
 


@@ -329,7 +356,7 @@ splice       (\\[ \f\t\v]*\n)*
 <SC_COMMENT>
 {
   "*"{splice}"/"  STRING_GROW; BEGIN context_state;
 <SC_COMMENT>
 {
   "*"{splice}"/"  STRING_GROW; BEGIN context_state;

-  <<EOF>>        unexpected_end_of_file (token_start, "*/");
+  <<EOF>>        unexpected_eof (token_start, "*/"); BEGIN context_state;

 }
 
 
 }
 
 


@@ -345,15 +372,14 @@ splice     (\\[ \f\t\v]*\n)*
 }
 
 
 }
 
 

-  /*----------------------------------------------------------------.
-  | Scanning a C string, including its escapes.  The initial `"' is |
-  | already eaten.                                                  |
-  `----------------------------------------------------------------*/
+  /*------------------------------------------------.
+  | Scanning a Bison string, including its escapes. |
+  | The initial quote is already eaten.             |
+  `------------------------------------------------*/

 
 <SC_ESCAPED_STRING>
 {
   "\"" {
 
 <SC_ESCAPED_STRING>
 {
   "\"" {

-    STRING_GROW;

     STRING_FINISH;
     loc->start = token_start;
     val->chars = last_string;
     STRING_FINISH;
     loc->start = token_start;
     val->chars = last_string;


@@ -361,15 +387,14 @@ splice     (\\[ \f\t\v]*\n)*
     BEGIN INITIAL;
     return STRING;
   }
     BEGIN INITIAL;
     return STRING;
   }

-
-  .|\n     STRING_GROW;
-  <<EOF>>   unexpected_end_of_file (token_start, "\"");
+  \n           unexpected_newline (token_start, "\""); BEGIN INITIAL;
+  <<EOF>>      unexpected_eof (token_start, "\"");     BEGIN INITIAL;

 }
 
 }
 

-  /*---------------------------------------------------------------.
-  | Scanning a C character, decoding its escapes.  The initial "'" |
-  | is already eaten.                                              |
-  `---------------------------------------------------------------*/
+  /*----------------------------------------------------------.
+  | Scanning a Bison character literal, decoding its escapes. |
+  | The initial quote is already eaten.                              |
+  `----------------------------------------------------------*/

 
 <SC_ESCAPED_CHARACTER>
 {
 
 <SC_ESCAPED_CHARACTER>
 {


@@ -378,7 +403,9 @@ splice       (\\[ \f\t\v]*\n)*
     STRING_GROW;
     STRING_FINISH;
     loc->start = token_start;
     STRING_GROW;
     STRING_FINISH;
     loc->start = token_start;

-    val->symbol = symbol_get (last_string, *loc);
+    val->symbol = symbol_get (quotearg_style (escape_quoting_style,
+                                             last_string),
+                             *loc);

     symbol_class_set (val->symbol, token_sym, *loc);
     last_string_1 = last_string[1];
     symbol_user_token_number_set (val->symbol, last_string_1, *loc);
     symbol_class_set (val->symbol, token_sym, *loc);
     last_string_1 = last_string[1];
     symbol_user_token_number_set (val->symbol, last_string_1, *loc);


@@ -387,9 +414,13 @@ splice      (\\[ \f\t\v]*\n)*
     BEGIN INITIAL;
     return ID;
   }
     BEGIN INITIAL;
     return ID;
   }

+  \n           unexpected_newline (token_start, "'");  BEGIN INITIAL;
+  <<EOF>>      unexpected_eof (token_start, "'");      BEGIN INITIAL;
+}

 
 

-  .|\n     STRING_GROW;
-  <<EOF>>   unexpected_end_of_file (token_start, "'");
+<SC_ESCAPED_CHARACTER,SC_ESCAPED_STRING>
+{
+  \0       complain_at (*loc, _("invalid null character"));

 }
 
 
 }
 
 


@@ -400,19 +431,23 @@ splice     (\\[ \f\t\v]*\n)*
 <SC_ESCAPED_STRING,SC_ESCAPED_CHARACTER>
 {
   \\[0-7]{1,3} {
 <SC_ESCAPED_STRING,SC_ESCAPED_CHARACTER>
 {
   \\[0-7]{1,3} {

-    unsigned long c = strtoul (yytext + 1, 0, 8);
+    unsigned long int c = strtoul (yytext + 1, 0, 8);

     if (UCHAR_MAX < c)
       complain_at (*loc, _("invalid escape sequence: %s"), quote (yytext));
     if (UCHAR_MAX < c)
       complain_at (*loc, _("invalid escape sequence: %s"), quote (yytext));

+    else if (! c)
+      complain_at (*loc, _("invalid null character: %s"), quote (yytext));

     else
       obstack_1grow (&obstack_for_string, c);
   }
 
   \\x[0-9abcdefABCDEF]+ {
     else
       obstack_1grow (&obstack_for_string, c);
   }
 
   \\x[0-9abcdefABCDEF]+ {

-    unsigned long c;
+    unsigned long int c;

     set_errno (0);
     c = strtoul (yytext + 2, 0, 16);
     if (UCHAR_MAX < c || get_errno ())
       complain_at (*loc, _("invalid escape sequence: %s"), quote (yytext));
     set_errno (0);
     c = strtoul (yytext + 2, 0, 16);
     if (UCHAR_MAX < c || get_errno ())
       complain_at (*loc, _("invalid escape sequence: %s"), quote (yytext));

+    else if (! c)
+      complain_at (*loc, _("invalid null character: %s"), quote (yytext));

     else
       obstack_1grow (&obstack_for_string, c);
   }
     else
       obstack_1grow (&obstack_for_string, c);
   }


@@ -432,6 +467,8 @@ splice       (\\[ \f\t\v]*\n)*
     int c = convert_ucn_to_byte (yytext);
     if (c < 0)
       complain_at (*loc, _("invalid escape sequence: %s"), quote (yytext));
     int c = convert_ucn_to_byte (yytext);
     if (c < 0)
       complain_at (*loc, _("invalid escape sequence: %s"), quote (yytext));

+    else if (! c)
+      complain_at (*loc, _("invalid null character: %s"), quote (yytext));

     else
       obstack_1grow (&obstack_for_string, c);
   }
     else
       obstack_1grow (&obstack_for_string, c);
   }


@@ -441,30 +478,27 @@ splice     (\\[ \f\t\v]*\n)*
   }
 }
 
   }
 }
 

+  /*--------------------------------------------.
+  | Scanning user-code characters and strings.  |
+  `--------------------------------------------*/

 
 

-  /*----------------------------------------------------------.
-  | Scanning a C character without decoding its escapes.  The |
-  | initial "'" is already eaten.                             |
-  `----------------------------------------------------------*/
+<SC_CHARACTER,SC_STRING>
+{
+  {splice}|\\{splice}[^\n$@\[\]]       STRING_GROW;
+}

 
 <SC_CHARACTER>
 {
 
 <SC_CHARACTER>
 {

-  "'"                  STRING_GROW; BEGIN context_state;
-  \\{splice}[^$@\[\]]  STRING_GROW;
-  <<EOF>>              unexpected_end_of_file (token_start, "'");
+  "'"          STRING_GROW; BEGIN context_state;
+  \n           unexpected_newline (token_start, "'"); BEGIN context_state;
+  <<EOF>>      unexpected_eof (token_start, "'"); BEGIN context_state;

 }
 
 }
 

-
-  /*----------------------------------------------------------------.
-  | Scanning a C string, without decoding its escapes.  The initial |
-  | `"' is already eaten.                                           |
-  `----------------------------------------------------------------*/
-

 <SC_STRING>
 {
 <SC_STRING>
 {

-  "\""                 STRING_GROW; BEGIN context_state;
-  \\{splice}[^$@\[\]]  STRING_GROW;
-  <<EOF>>              unexpected_end_of_file (token_start, "\"");
+  "\""         STRING_GROW; BEGIN context_state;
+  \n           unexpected_newline (token_start, "\""); BEGIN context_state;
+  <<EOF>>      unexpected_eof (token_start, "\""); BEGIN context_state;

 }
 
 
 }
 
 


@@ -521,7 +555,7 @@ splice       (\\[ \f\t\v]*\n)*
       }
     else
       {
       }
     else
       {

-       complain_at (*loc, _("missing `{' in `%s'"),
+       complain_at (*loc, _("missing `{' in %s"),

                     token_name (token_type));
        obstack_sgrow (&obstack_for_string, "{}");
        STRING_FINISH;
                     token_name (token_type));
        obstack_sgrow (&obstack_for_string, "{}");
        STRING_FINISH;


@@ -530,6 +564,8 @@ splice       (\\[ \f\t\v]*\n)*
        return token_type;
       }
   }
        return token_type;
       }
   }

+
+  <<EOF>>  unexpected_eof (scanner_cursor, "{}"); BEGIN INITIAL;

 }
 
 
 }
 
 


@@ -582,7 +618,16 @@ splice      (\\[ \f\t\v]*\n)*
   "$"("<"{tag}">")?(-?[0-9]+|"$")  handle_dollar (token_type, yytext, *loc);
   "@"(-?[0-9]+|"$")               handle_at (token_type, yytext, *loc);
 
   "$"("<"{tag}">")?(-?[0-9]+|"$")  handle_dollar (token_type, yytext, *loc);
   "@"(-?[0-9]+|"$")               handle_at (token_type, yytext, *loc);
 

-  <<EOF>>  unexpected_end_of_file (code_start, "}");
+  "$"  {
+    warn_at (*loc, _("stray `$'"));
+    obstack_sgrow (&obstack_for_string, "$][");
+  }
+  "@"  {
+    warn_at (*loc, _("stray `@'"));
+    obstack_sgrow (&obstack_for_string, "@@");
+  }
+
+  <<EOF>>  unexpected_eof (code_start, "}"); BEGIN INITIAL;

 }
 
 
 }
 
 


@@ -600,7 +645,7 @@ splice       (\\[ \f\t\v]*\n)*
     return PROLOGUE;
   }
 
     return PROLOGUE;
   }
 

-  <<EOF>>  unexpected_end_of_file (code_start, "%}");
+  <<EOF>>  unexpected_eof (code_start, "%}"); BEGIN INITIAL;

 }
 
 
 }
 
 


@@ -621,10 +666,9 @@ splice      (\\[ \f\t\v]*\n)*
 }
 
 
 }
 
 

-  /*----------------------------------------------------------------.
-  | By default, grow the string obstack with the input, escaping M4 |
-  | quoting characters.                                                    |
-  `----------------------------------------------------------------*/
+  /*-----------------------------------------.
+  | Escape M4 quoting characters in C code.  |
+  `-----------------------------------------*/

 
 <SC_COMMENT,SC_LINE_COMMENT,SC_STRING,SC_CHARACTER,SC_BRACED_CODE,SC_PROLOGUE,SC_EPILOGUE>
 {
 
 <SC_COMMENT,SC_LINE_COMMENT,SC_STRING,SC_CHARACTER,SC_BRACED_CODE,SC_PROLOGUE,SC_EPILOGUE>
 {


@@ -632,12 +676,23 @@ splice     (\\[ \f\t\v]*\n)*
   \@   obstack_sgrow (&obstack_for_string, "@@");
   \[   obstack_sgrow (&obstack_for_string, "@{");
   \]   obstack_sgrow (&obstack_for_string, "@}");
   \@   obstack_sgrow (&obstack_for_string, "@@");
   \[   obstack_sgrow (&obstack_for_string, "@{");
   \]   obstack_sgrow (&obstack_for_string, "@}");

-  .|\n  STRING_GROW;

 }
 
 
 }
 
 

+  /*-----------------------------------------------------.
+  | By default, grow the string obstack with the input.  |
+  `-----------------------------------------------------*/
+
+<SC_COMMENT,SC_LINE_COMMENT,SC_BRACED_CODE,SC_PROLOGUE,SC_EPILOGUE,SC_STRING,SC_CHARACTER,SC_ESCAPED_STRING,SC_ESCAPED_CHARACTER>.     |
+<SC_COMMENT,SC_LINE_COMMENT,SC_BRACED_CODE,SC_PROLOGUE,SC_EPILOGUE>\n  STRING_GROW;
+

 %%
 
 %%
 

+/* Keeps track of the maximum number of semantic values to the left of
+   a handle (those referenced by $0, $-1, etc.) are required by the
+   semantic actions of this grammar. */
+int max_left_semantic_context = 0;
+

 /* Set *LOC and adjust scanner cursor to account for token TOKEN of
    size SIZE.  */
 
 /* Set *LOC and adjust scanner cursor to account for token TOKEN of
    size SIZE.  */
 


@@ -682,14 +737,14 @@ adjust_location (location *loc, char const *token, size_t size)
 static size_t
 no_cr_read (FILE *fp, char *buf, 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;
       if (w)
        {
          char const *r = ++w;

-         char const *lim = buf + s;
+         char const *lim = buf + bytes_read;

 
          for (;;)
            {
 
          for (;;)
            {


@@ -718,7 +773,7 @@ no_cr_read (FILE *fp, char *buf, size_t size)
        }
     }
 
        }
     }
 

-  return s;
+  return bytes_read;

 }
 
 
 }
 
 


@@ -763,13 +818,15 @@ handle_action_dollar (char *text, location loc)
     }
   else
     {
     }
   else
     {

-      long num;
+      long int num;

       set_errno (0);
       num = strtol (cp, 0, 10);
 
       if (INT_MIN <= num && num <= rule_length && ! get_errno ())
        {
          int n = num;
       set_errno (0);
       num = strtol (cp, 0, 10);
 
       if (INT_MIN <= num && num <= rule_length && ! get_errno ())
        {
          int n = num;

+         if (1-n > max_left_semantic_context)
+           max_left_semantic_context = 1-n;

          if (!type_name && n > 0)
            type_name = symbol_list_n_type_name_get (current_rule, loc, n);
          if (!type_name && typed)
          if (!type_name && n > 0)
            type_name = symbol_list_n_type_name_get (current_rule, loc, n);
          if (!type_name && typed)


@@ -778,7 +835,7 @@ handle_action_dollar (char *text, location loc)
          if (!type_name)
            type_name = "";
          obstack_fgrow3 (&obstack_for_string,
          if (!type_name)
            type_name = "";
          obstack_fgrow3 (&obstack_for_string,

-                         "]b4_rhs_value([%d], [%d], [%s])[",
+                         "]b4_rhs_value(%d, %d, [%s])[",

                          rule_length, n, type_name);
        }
       else
                          rule_length, n, type_name);
        }
       else


@@ -789,10 +846,10 @@ handle_action_dollar (char *text, location loc)
 }
 
 
 }
 
 

-/*-----------------------------------------------------------------.
-| Dispatch onto handle_action_dollar, or handle_destructor_dollar, |
-| depending upon TOKEN_TYPE.                                       |
-`-----------------------------------------------------------------*/
+/*----------------------------------------------------------------.
+| Map `$?' onto the proper M4 symbol, depending on its TOKEN_TYPE |
+| (are we in an action?).                                         |
+`----------------------------------------------------------------*/

 
 static void
 handle_dollar (int token_type, char *text, location loc)
 
 static void
 handle_dollar (int token_type, char *text, location loc)


@@ -805,6 +862,7 @@ handle_dollar (int token_type, char *text, location loc)
       break;
 
     case PERCENT_DESTRUCTOR:
       break;
 
     case PERCENT_DESTRUCTOR:

+    case PERCENT_INITIAL_ACTION:

     case PERCENT_PRINTER:
       if (text[1] == '$')
        {
     case PERCENT_PRINTER:
       if (text[1] == '$')
        {


@@ -830,7 +888,7 @@ static inline bool
 handle_action_at (char *text, location loc)
 {
   char *cp = text + 1;
 handle_action_at (char *text, location loc)
 {
   char *cp = text + 1;

-  locations_flag = 1;
+  locations_flag = true;

 
   if (! current_rule)
     return false;
 
   if (! current_rule)
     return false;


@@ -839,14 +897,14 @@ handle_action_at (char *text, location loc)
     obstack_sgrow (&obstack_for_string, "]b4_lhs_location[");
   else
     {
     obstack_sgrow (&obstack_for_string, "]b4_lhs_location[");
   else
     {

-      long num;
+      long int num;

       set_errno (0);
       num = strtol (cp, 0, 10);
 
       if (INT_MIN <= num && num <= rule_length && ! get_errno ())
        {
          int n = num;
       set_errno (0);
       num = strtol (cp, 0, 10);
 
       if (INT_MIN <= num && num <= rule_length && ! get_errno ())
        {
          int n = num;

-         obstack_fgrow2 (&obstack_for_string, "]b4_rhs_location([%d], [%d])[",
+         obstack_fgrow2 (&obstack_for_string, "]b4_rhs_location(%d, %d)[",

                          rule_length, n);
        }
       else
                          rule_length, n);
        }
       else


@@ -857,10 +915,10 @@ handle_action_at (char *text, location loc)
 }
 
 
 }
 
 

-/*-------------------------------------------------------------------.
-| Dispatch onto handle_action_at, or handle_destructor_at, depending |
-| upon CODE_KIND.                                                    |
-`-------------------------------------------------------------------*/
+/*----------------------------------------------------------------.
+| Map `@?' onto the proper M4 symbol, depending on its TOKEN_TYPE |
+| (are we in an action?).                                         |
+`----------------------------------------------------------------*/

 
 static void
 handle_at (int token_type, char *text, location loc)
 
 static void
 handle_at (int token_type, char *text, location loc)


@@ -871,6 +929,7 @@ handle_at (int token_type, char *text, location loc)
       handle_action_at (text, loc);
       return;
 
       handle_action_at (text, loc);
       return;
 

+    case PERCENT_INITIAL_ACTION:

     case PERCENT_DESTRUCTOR:
     case PERCENT_PRINTER:
       if (text[1] == '$')
     case PERCENT_DESTRUCTOR:
     case PERCENT_PRINTER:
       if (text[1] == '$')


@@ -888,6 +947,25 @@ handle_at (int token_type, char *text, location loc)
 }
 
 
 }
 
 

+/*------------------------------------------------------.
+| Scan NUMBER for a base-BASE integer at location LOC.  |
+`------------------------------------------------------*/
+
+static unsigned long int
+scan_integer (char const *number, int base, location loc)
+{
+  unsigned long int num;
+  set_errno (0);
+  num = strtoul (number, 0, base);
+  if (INT_MAX < num || get_errno ())
+    {
+      complain_at (loc, _("integer out of range: %s"), quote (number));
+      num = INT_MAX;
+    }
+  return num;
+}
+
+

 /*------------------------------------------------------------------.
 | Convert universal character name UCN to a single-byte character,  |
 | and return that character.  Return -1 if UCN does not correspond  |
 /*------------------------------------------------------------------.
 | Convert universal character name UCN to a single-byte character,  |
 | and return that character.  Return -1 if UCN does not correspond  |


@@ -897,7 +975,7 @@ handle_at (int token_type, char *text, location loc)
 static int
 convert_ucn_to_byte (char const *ucn)
 {
 static int
 convert_ucn_to_byte (char const *ucn)
 {

-  unsigned long code = strtoul (ucn + 2, 0, 16);
+  unsigned long int code = strtoul (ucn + 2, 0, 16);

 
   /* FIXME: Currently we assume Unicode-compatible unibyte characters
      on ASCII hosts (i.e., Latin-1 on hosts with 8-bit bytes).  On
 
   /* FIXME: Currently we assume Unicode-compatible unibyte characters
      on ASCII hosts (i.e., Latin-1 on hosts with 8-bit bytes).  On


@@ -955,34 +1033,48 @@ handle_syncline (char *args)
   const char *file = NULL;
   file = strchr (args, '"') + 1;
   *strchr (file, '"') = 0;
   const char *file = NULL;
   file = strchr (args, '"') + 1;
   *strchr (file, '"') = 0;

-  scanner_cursor.file = current_file = xstrdup (file);
+  scanner_cursor.file = current_file = uniqstr_new (file);

   scanner_cursor.line = lineno;
   scanner_cursor.column = 1;
 }
 
 
   scanner_cursor.line = lineno;
   scanner_cursor.column = 1;
 }
 
 

+/*----------------------------------------------------------------.
+| For a token or comment starting at START, report message MSGID, |
+| which should say that an end marker was found before           |
+| the expected TOKEN_END.                                        |
+`----------------------------------------------------------------*/
+
+static void
+unexpected_end (boundary start, char const *msgid, char const *token_end)
+{
+  location loc;
+  loc.start = start;
+  loc.end = scanner_cursor;
+  complain_at (loc, _(msgid), token_end);
+}
+
+

 /*------------------------------------------------------------------------.
 | 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.  |
 /*------------------------------------------------------------------------.
 | 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
 `------------------------------------------------------------------------*/
 
 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);
+  unexpected_end (start, N_("missing `%s' at end of file"), 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;
+/*----------------------------------------.
+| Likewise, but for unexpected newlines.  |
+`----------------------------------------*/

 
 

-  while (i != 0)
-    unput (token_end[--i]);
+static void
+unexpected_newline (boundary start, char const *token_end)
+{
+  unexpected_end (start, N_("missing `%s' at end of line"), token_end);

 }
 
 
 }