Various DJGGP specific issues in /djgpp.

[bison.git] / src / parse-gram.y

diff --git a/src/parse-gram.y b/src/parse-gram.y
index 02608bb0476352b7139973e54616c30b08dbb73b..11cd489e36f578999f3d6e5c5ddc9b040385d23f 100644 (file)
--- a/src/parse-gram.y
+++ b/src/parse-gram.y
@@ -1,6 +1,6 @@
-/* Bison Grammar Parser                             -*- C -*-
+%{/* Bison Grammar Parser                             -*- C -*-

 
 

-   Copyright (C) 2002, 2003 Free Software Foundation, Inc.
+   Copyright (C) 2002, 2003, 2004, 2005, 2006 Free Software Foundation, Inc.

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


@@ -16,20 +16,11 @@
 
    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

 */
 
 */
 

-
-%debug
-%defines
-%locations
-%pure-parser
-// %error-verbose
-%defines
-%name-prefix="gram_"
-
-%{
+#include <config.h>

 #include "system.h"
 
 #include "complain.h"
 #include "system.h"
 
 #include "complain.h"


@@ -39,36 +30,65 @@
 #include "gram.h"
 #include "muscle_tab.h"
 #include "output.h"
 #include "gram.h"
 #include "muscle_tab.h"
 #include "output.h"

+#include "quotearg.h"

 #include "reader.h"
 #include "symlist.h"
 #include "reader.h"
 #include "symlist.h"

-
-/* Produce verbose syntax errors.  */
-#define YYERROR_VERBOSE 1
+#include "strverscmp.h"

 
 #define YYLLOC_DEFAULT(Current, Rhs, N)  (Current) = lloc_default (Rhs, N)
 static YYLTYPE lloc_default (YYLTYPE const *, int);
 
 
 #define YYLLOC_DEFAULT(Current, Rhs, N)  (Current) = lloc_default (Rhs, N)
 static YYLTYPE lloc_default (YYLTYPE const *, int);
 

+#define YY_LOCATION_PRINT(File, Loc) \
+         location_print (File, Loc)
+
+static void version_check (location const *loc, char const *version);
+

 /* Request detailed syntax error messages, and pass them to GRAM_ERROR.
    FIXME: depends on the undocumented availability of YYLLOC.  */
 #undef  yyerror
 #define yyerror(Msg) \
 /* Request detailed syntax error messages, and pass them to GRAM_ERROR.
    FIXME: depends on the undocumented availability of YYLLOC.  */
 #undef  yyerror
 #define yyerror(Msg) \

-        gram_error (&yylloc, Msg)
+       gram_error (&yylloc, Msg)

 static void gram_error (location const *, char const *);
 
 static void gram_error (location const *, char const *);
 

-#define YYPRINT(File, Type, Value) \
-       print_token_value (File, Type, &Value)
-static void print_token_value (FILE *, int, YYSTYPE const *);
-

 static void add_param (char const *, char *, location);
 
 static void add_param (char const *, char *, location);
 

-symbol_class current_class = unknown_sym;
-uniqstr current_type = 0;
-symbol *current_lhs;
-location current_lhs_location;
-assoc current_assoc;
-int current_prec = 0;
+static symbol_class current_class = unknown_sym;
+static uniqstr current_type = 0;
+static symbol *current_lhs;
+static location current_lhs_location;
+static int current_prec = 0;
+
+#ifdef UINT_FAST8_MAX
+# define YYTYPE_UINT8 uint_fast8_t
+#endif
+#ifdef INT_FAST8_MAX
+# define YYTYPE_INT8 int_fast8_t
+#endif
+#ifdef UINT_FAST16_MAX
+# define YYTYPE_UINT16 uint_fast16_t
+#endif
+#ifdef INT_FAST16_MAX
+# define YYTYPE_INT16 int_fast16_t
+#endif

 %}
 
 %}
 

+%debug
+%verbose
+%defines
+%locations
+%pure-parser
+%error-verbose
+%defines
+%name-prefix="gram_"
+
+%initial-action
+{
+  /* Bison's grammar can initial empty locations, hence a default
+     location is needed. */
+  @$.start.file   = @$.end.file   = current_file;
+  @$.start.line   = @$.end.line   = 1;
+  @$.start.column = @$.end.column = 0;
+}

 
 /* Only NUMBERS have a value.  */
 %union
 
 /* Only NUMBERS have a value.  */
 %union


@@ -109,26 +129,32 @@ int current_prec = 0;
 `----------------------*/
 
 %token
 `----------------------*/
 
 %token

-  PERCENT_DEBUG         "%debug"
-  PERCENT_DEFINE        "%define"
-  PERCENT_DEFINES       "%defines"
-  PERCENT_ERROR_VERBOSE "%error-verbose"
-  PERCENT_EXPECT        "%expect"
-  PERCENT_FILE_PREFIX   "%file-prefix"
-  PERCENT_GLR_PARSER    "%glr-parser"
-  PERCENT_LEX_PARAM     "%lex-param {...}"
-  PERCENT_LOCATIONS     "%locations"
-  PERCENT_NAME_PREFIX   "%name-prefix"
-  PERCENT_NO_LINES      "%no-lines"
-  PERCENT_NONDETERMINISTIC_PARSER    "%nondeterministic-parser"
-  PERCENT_OUTPUT        "%output"
-  PERCENT_PARSE_PARAM   "%parse-param {...}"
-  PERCENT_PURE_PARSER   "%pure-parser"
-  PERCENT_SKELETON      "%skeleton"
-  PERCENT_START         "%start"
-  PERCENT_TOKEN_TABLE   "%token-table"
-  PERCENT_VERBOSE       "%verbose"
-  PERCENT_YACC          "%yacc"
+  PERCENT_DEBUG           "%debug"
+  PERCENT_DEFAULT_PREC    "%default-prec"
+  PERCENT_DEFINE          "%define"
+  PERCENT_DEFINES         "%defines"
+  PERCENT_ERROR_VERBOSE   "%error-verbose"
+  PERCENT_EXPECT          "%expect"
+  PERCENT_EXPECT_RR      "%expect-rr"
+  PERCENT_FILE_PREFIX     "%file-prefix"
+  PERCENT_GLR_PARSER      "%glr-parser"
+  PERCENT_INITIAL_ACTION  "%initial-action {...}"
+  PERCENT_LEX_PARAM       "%lex-param {...}"
+  PERCENT_LOCATIONS       "%locations"
+  PERCENT_NAME_PREFIX     "%name-prefix"
+  PERCENT_NO_DEFAULT_PREC "%no-default-prec"
+  PERCENT_NO_LINES        "%no-lines"
+  PERCENT_NONDETERMINISTIC_PARSER
+                         "%nondeterministic-parser"
+  PERCENT_OUTPUT          "%output"
+  PERCENT_PARSE_PARAM     "%parse-param {...}"
+  PERCENT_PURE_PARSER     "%pure-parser"
+  PERCENT_REQUIRE        "%require"
+  PERCENT_SKELETON        "%skeleton"
+  PERCENT_START           "%start"
+  PERCENT_TOKEN_TABLE     "%token-table"
+  PERCENT_VERBOSE         "%verbose"
+  PERCENT_YACC            "%yacc"

 ;
 
 %token TYPE            "type"
 ;
 
 %token TYPE            "type"


@@ -145,15 +171,30 @@ int current_prec = 0;
 
 %type <chars> STRING string_content
              "%destructor {...}"
 
 %type <chars> STRING string_content
              "%destructor {...}"

+             "%initial-action {...}"
+             "%lex-param {...}"
+             "%parse-param {...}"
+             "%printer {...}"
+             "%union {...}"
+             PROLOGUE EPILOGUE
+%printer { fprintf (stderr, "\"%s\"", $$); }
+             STRING string_content
+%printer { fprintf (stderr, "{\n%s\n}", $$); }
+             "%destructor {...}"
+             "%initial-action {...}"

              "%lex-param {...}"
              "%parse-param {...}"
              "%printer {...}"
              "%union {...}"
              "%lex-param {...}"
              "%parse-param {...}"
              "%printer {...}"
              "%union {...}"

-             BRACED_CODE action

              PROLOGUE EPILOGUE
 %type <uniqstr> TYPE
              PROLOGUE EPILOGUE
 %type <uniqstr> TYPE

+%printer { fprintf (stderr, "<%s>", $$); } TYPE

 %type <integer> INT
 %type <integer> INT

-%type <symbol> ID ID_COLON symbol string_as_id
+%printer { fprintf (stderr, "%d", $$); } INT
+%type <symbol> ID symbol string_as_id
+%printer { fprintf (stderr, "%s", $$->tag); } ID symbol string_as_id
+%type <symbol> ID_COLON
+%printer { fprintf (stderr, "%s:", $$->tag); } ID_COLON

 %type <assoc> precedence_declarator
 %type <list>  symbols.1
 %%
 %type <assoc> precedence_declarator
 %type <list>  symbols.1
 %%


@@ -175,27 +216,41 @@ declarations:
 declaration:
   grammar_declaration
 | PROLOGUE                                 { prologue_augment ($1, @1); }
 declaration:
   grammar_declaration
 | PROLOGUE                                 { prologue_augment ($1, @1); }

-| "%debug"                                 { debug_flag = 1; }
+| "%debug"                                 { debug_flag = true; }
+| "%define" string_content
+    {
+      static char one[] = "1";
+      muscle_insert ($2, one);
+    }

 | "%define" string_content string_content  { muscle_insert ($2, $3); }
 | "%define" string_content string_content  { muscle_insert ($2, $3); }

-| "%defines"                               { defines_flag = 1; }
-| "%error-verbose"                         { error_verbose = 1; }
-| "%expect" INT                            { expected_conflicts = $2; }
+| "%defines"                               { defines_flag = true; }
+| "%error-verbose"                         { error_verbose = true; }
+| "%expect" INT                            { expected_sr_conflicts = $2; }
+| "%expect-rr" INT                        { expected_rr_conflicts = $2; }

 | "%file-prefix" "=" string_content        { spec_file_prefix = $3; }
 | "%file-prefix" "=" string_content        { spec_file_prefix = $3; }

-| "%glr-parser"                           { nondeterministic_parser = true;
-                                             glr_parser = true; }
+| "%glr-parser"
+    {
+      nondeterministic_parser = true;
+      glr_parser = true;
+    }
+| "%initial-action {...}"
+    {
+      muscle_code_grow ("initial_action", $1, @1);
+    }

 | "%lex-param {...}"                      { add_param ("lex_param", $1, @1); }
 | "%lex-param {...}"                      { add_param ("lex_param", $1, @1); }

-| "%locations"                             { locations_flag = 1; }
+| "%locations"                             { locations_flag = true; }

 | "%name-prefix" "=" string_content        { spec_name_prefix = $3; }
 | "%name-prefix" "=" string_content        { spec_name_prefix = $3; }

-| "%no-lines"                              { no_lines_flag = 1; }
-| "%nondeterministic-parser"              { nondeterministic_parser = true; }
+| "%no-lines"                              { no_lines_flag = true; }
+| "%nondeterministic-parser"              { nondeterministic_parser = true; }

 | "%output" "=" string_content             { spec_outfile = $3; }
 | "%output" "=" string_content             { spec_outfile = $3; }

-| "%parse-param {...}"                  { add_param ("parse_param", $1, @1); }
+| "%parse-param {...}"                    { add_param ("parse_param", $1, @1); }

 | "%pure-parser"                           { pure_parser = true; }
 | "%pure-parser"                           { pure_parser = true; }

+| "%require" string_content                { version_check (&@2, $2); }

 | "%skeleton" string_content               { skeleton = $2; }
 | "%skeleton" string_content               { skeleton = $2; }

-| "%token-table"                           { token_table_flag = 1; }
+| "%token-table"                           { token_table_flag = true; }

 | "%verbose"                               { report_flag = report_states; }
 | "%verbose"                               { report_flag = report_states; }

-| "%yacc"                                  { yacc_flag = 1; }
-| ";"
+| "%yacc"                                  { yacc_flag = true; }
+| /*FIXME: Err?  What is this horror doing here? */ ";"

 ;
 
 grammar_declaration:
 ;
 
 grammar_declaration:


@@ -207,9 +262,19 @@ grammar_declaration:
     }
 | "%union {...}"
     {
     }
 | "%union {...}"
     {

-      typed = 1;
-      MUSCLE_INSERT_INT ("stype_line", @1.start.line);
-      muscle_insert ("stype", $1);
+      char const *body = $1;
+
+      if (typed)
+       {
+         /* Concatenate the union bodies, turning the first one's
+            trailing '}' into '\n', and omitting the second one's '{'.  */
+         char *code = muscle_find ("stype");
+         code[strlen (code) - 1] = '\n';
+         body++;
+       }
+
+      typed = true;
+      muscle_code_grow ("stype", body, @1);

     }
 | "%destructor {...}" symbols.1
     {
     }
 | "%destructor {...}" symbols.1
     {


@@ -222,9 +287,17 @@ grammar_declaration:
     {
       symbol_list *list;
       for (list = $2; list; list = list->next)
     {
       symbol_list *list;
       for (list = $2; list; list = list->next)

-       symbol_printer_set (list->sym, $1, list->location);
+       symbol_printer_set (list->sym, $1, @1);

       symbol_list_free ($2);
     }
       symbol_list_free ($2);
     }

+| "%default-prec"
+    {
+      default_prec = true;
+    }
+| "%no-default-prec"
+    {
+      default_prec = false;
+    }

 ;
 
 symbol_declaration:
 ;
 
 symbol_declaration:


@@ -288,24 +361,24 @@ symbol_def:
      }
 | ID
      {
      }
 | ID
      {

-       symbol_class_set ($1, current_class, @1);
+       symbol_class_set ($1, current_class, @1, true);

        symbol_type_set ($1, current_type, @1);
      }
 | ID INT
     {
        symbol_type_set ($1, current_type, @1);
      }
 | ID INT
     {

-      symbol_class_set ($1, current_class, @1);
+      symbol_class_set ($1, current_class, @1, true);

       symbol_type_set ($1, current_type, @1);
       symbol_user_token_number_set ($1, $2, @2);
     }
 | ID string_as_id
     {
       symbol_type_set ($1, current_type, @1);
       symbol_user_token_number_set ($1, $2, @2);
     }
 | ID string_as_id
     {

-      symbol_class_set ($1, current_class, @1);
+      symbol_class_set ($1, current_class, @1, true);

       symbol_type_set ($1, current_type, @1);
       symbol_make_alias ($1, $2, @$);
     }
 | ID INT string_as_id
     {
       symbol_type_set ($1, current_type, @1);
       symbol_make_alias ($1, $2, @$);
     }
 | ID INT string_as_id
     {

-      symbol_class_set ($1, current_class, @1);
+      symbol_class_set ($1, current_class, @1, true);

       symbol_type_set ($1, current_type, @1);
       symbol_user_token_number_set ($1, $2, @2);
       symbol_make_alias ($1, $3, @$);
       symbol_type_set ($1, current_type, @1);
       symbol_user_token_number_set ($1, $2, @2);
       symbol_make_alias ($1, $3, @$);


@@ -333,10 +406,6 @@ grammar:
 rules_or_grammar_declaration:
   rules
 | grammar_declaration ";"
 rules_or_grammar_declaration:
   rules
 | grammar_declaration ";"

-    {
-      if (yacc_flag)
-       complain_at (@$, _("POSIX forbids declarations in the grammar"));
-    }

 | error ";"
     {
       yyerrok;
 | error ";"
     {
       yyerrok;


@@ -348,18 +417,17 @@ rules:
 ;
 
 rhses.1:
 ;
 
 rhses.1:

-  rhs                { grammar_rule_end (@1); }
-| rhses.1 "|" rhs    { grammar_rule_end (@3); }
+  rhs                { grammar_current_rule_end (@1); }
+| rhses.1 "|" rhs    { grammar_current_rule_end (@3); }

 | rhses.1 ";"
 ;
 
 rhs:
   /* Nothing.  */
 | rhses.1 ";"
 ;
 
 rhs:
   /* Nothing.  */

-    { grammar_rule_begin (current_lhs, current_lhs_location); }
+    { grammar_current_rule_begin (current_lhs, current_lhs_location); }

 | rhs symbol
     { grammar_current_rule_symbol_append ($2, @2); }
 | rhs action
 | rhs symbol
     { grammar_current_rule_symbol_append ($2, @2); }
 | rhs action

-    { grammar_current_rule_action_append ($2, @2); }

 | rhs "%prec" symbol
     { grammar_current_rule_prec_set ($3, @3); }
 | rhs "%dprec" INT
 | rhs "%prec" symbol
     { grammar_current_rule_prec_set ($3, @3); }
 | rhs "%dprec" INT


@@ -373,34 +441,44 @@ symbol:
 | string_as_id    { $$ = $1; }
 ;
 
 | string_as_id    { $$ = $1; }
 ;
 

+/* Handle the semantics of an action specially, with a mid-rule
+   action, so that grammar_current_rule_action_append is invoked
+   immediately after the braced code is read by the scanner.
+
+   This implementation relies on the LALR(1) parsing algorithm.
+   If grammar_current_rule_action_append were executed in a normal
+   action for this rule, then when the input grammar contains two
+   successive actions, the scanner would have to read both actions
+   before reducing this rule.  That wouldn't work, since the scanner
+   relies on all preceding input actions being processed by
+   grammar_current_rule_action_append before it scans the next
+   action.  */

 action:
 action:

+    { grammar_current_rule_action_append (last_string, last_braced_code_loc); }

   BRACED_CODE
   BRACED_CODE

-   { $$ = $1; }

 ;
 
 ;
 

-/* A string used as an ID: we have to keep the quotes. */
+/* A string used as an ID: quote it.  */

 string_as_id:
   STRING
     {
 string_as_id:
   STRING
     {

-      $$ = symbol_get ($1, @1);
-      symbol_class_set ($$, token_sym, @1);
+      $$ = symbol_get (quotearg_style (c_quoting_style, $1), @1);
+      symbol_class_set ($$, token_sym, @1, false);

     }
 ;
 
     }
 ;
 

-/* A string used for its contents.  Strip the quotes. */
+/* A string used for its contents.  Don't quote it.  */

 string_content:
   STRING
 string_content:
   STRING

-    {
-      $$ = $1 + 1;
-      $$[strlen ($$) - 1] = '\0';
-    };
+    { $$ = $1; }
+;

 
 
 epilogue.opt:
   /* Nothing.  */
 | "%%" EPILOGUE
     {
 
 
 epilogue.opt:
   /* Nothing.  */
 | "%%" EPILOGUE
     {

-      epilogue_augment ($2, @2);
+      muscle_code_grow ("epilogue", $2, @2);

       scanner_last_string_free ();
     }
 ;
       scanner_last_string_free ();
     }
 ;


@@ -418,7 +496,11 @@ lloc_default (YYLTYPE const *rhs, int n)
 {
   int i;
   YYLTYPE loc;
 {
   int i;
   YYLTYPE loc;

-  loc.start = loc.end = rhs[n].end;
+
+  /* SGI MIPSpro 7.4.1m miscompiles "loc.start = loc.end = rhs[n].end;".
+     The bug is fixed in 7.4.2m, but play it safe for now.  */
+  loc.start = rhs[n].end;
+  loc.end = rhs[n].end;

 
   /* Ignore empty nonterminals the start of the the right-hand side.
      Do not bother to ignore them at the end of the right-hand side,
 
   /* Ignore empty nonterminals the start of the the right-hand side.
      Do not bother to ignore them at the end of the right-hand side,


@@ -440,22 +522,28 @@ lloc_default (YYLTYPE const *rhs, int n)
 static void
 add_param (char const *type, char *decl, location loc)
 {
 static void
 add_param (char const *type, char *decl, location loc)
 {

-  static char const alphanum[] =
-    "0123456789"
+  static char const alphanum[26 + 26 + 1 + 10] =

     "abcdefghijklmnopqrstuvwxyz"
     "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
     "abcdefghijklmnopqrstuvwxyz"
     "ABCDEFGHIJKLMNOPQRSTUVWXYZ"

-    "_";
-  char const *alpha = alphanum + 10;
+    "_"
+    "0123456789";

   char const *name_start = NULL;
   char *p;
 
   char const *name_start = NULL;
   char *p;
 

-  for (p = decl; *p; p++)
-    if ((p == decl || ! strchr (alphanum, p[-1])) && strchr (alpha, p[0]))
+  /* Stop on last actual character.  */
+  for (p = decl; p[1]; p++)
+    if ((p == decl
+        || ! memchr (alphanum, p[-1], sizeof alphanum))
+       && memchr (alphanum, p[0], sizeof alphanum - 10))

       name_start = p;
 
       name_start = p;
 

-  /* Strip the surrounding '{' and '}'.  */
-  decl++;
-  p[-1] = '\0';
+  /* Strip the surrounding '{' and '}', and any blanks just inside
+     the braces.  */
+  while (*--p == ' ' || *p == '\t')
+    continue;
+  p[1] = '\0';
+  while (*++decl == ' ' || *decl == '\t')
+    continue;

 
   if (! name_start)
     complain_at (loc, _("missing identifier in parameter declaration"));
 
   if (! name_start)
     complain_at (loc, _("missing identifier in parameter declaration"));


@@ -465,7 +553,7 @@ add_param (char const *type, char *decl, location loc)
       size_t name_len;
 
       for (name_len = 1;
       size_t name_len;
 
       for (name_len = 1;

-          name_start[name_len] && strchr (alphanum, name_start[name_len]);
+          memchr (alphanum, name_start[name_len], sizeof alphanum);

           name_len++)
        continue;
 
           name_len++)
        continue;
 


@@ -479,46 +567,14 @@ add_param (char const *type, char *decl, location loc)
   scanner_last_string_free ();
 }
 
   scanner_last_string_free ();
 }
 

-/*----------------------------------------------------.
-| When debugging the parser, display tokens' values.  |
-`----------------------------------------------------*/
-

 static void
 static void

-print_token_value (FILE *file, int type, YYSTYPE const *value)
+version_check (location const *loc, char const *version)

 {
 {

-  fputc (' ', file);
-  switch (type)
+  if (strverscmp (version, PACKAGE_VERSION) > 0)

     {
     {

-    case ID:
-      fprintf (file, " = %s", value->symbol->tag);
-      break;
-
-    case INT:
-      fprintf (file, " = %d", value->integer);
-      break;
-
-    case STRING:
-      fprintf (file, " = \"%s\"", value->chars);
-      break;
-
-    case TYPE:
-      fprintf (file, " = <%s>", value->uniqstr);
-      break;
-
-    case BRACED_CODE:
-    case PERCENT_DESTRUCTOR:
-    case PERCENT_LEX_PARAM:
-    case PERCENT_PARSE_PARAM:
-    case PERCENT_PRINTER:
-    case PERCENT_UNION:
-    case PROLOGUE:
-    case EPILOGUE:
-      fprintf (file, " = {{ %s }}", value->chars);
-      break;
-
-    default:
-      fprintf (file, "unknown token type");
-      break;
+      complain_at (*loc, "require bison %s, but have %s",
+                  version, PACKAGE_VERSION);
+      exit (63);

     }
 }
 
     }
 }
 


@@ -531,5 +587,5 @@ gram_error (location const *loc, char const *msg)
 char const *
 token_name (int type)
 {
 char const *
 token_name (int type)
 {

-  return yytname[type];
+  return yytname[YYTRANSLATE (type)];

 }
 }