version 2.5.91

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

diff --git a/src/parse-gram.y b/src/parse-gram.y
index a8a1335ba1ede24f5ce22fc51de42b05ece2fe19..eb034b0823a18240fa325f12a3bb6f421ba04c35 100644 (file)
--- a/src/parse-gram.y
+++ b/src/parse-gram.y
@@ -1,13 +1,12 @@
 %{/* Bison Grammar Parser                             -*- C -*-
 
 %{/* Bison Grammar Parser                             -*- C -*-
 

-   Copyright (C) 2002, 2003, 2004, 2005, 2006, 2007 Free Software Foundation,
-   Inc.
+   Copyright (C) 2002-2012 Free Software Foundation, Inc.

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

-   This program is free software; you can redistribute it and/or modify
+   This program is free software: you can redistribute it and/or modify

    it under the terms of the GNU General Public License as published by
    it under the terms of the GNU General Public License as published by

-   the Free Software Foundation; either version 2 of the License, or
+   the Free Software Foundation, either version 3 of the License, or

    (at your option) any later version.
 
    This program is distributed in the hope that it will be useful,
    (at your option) any later version.
 
    This program is distributed in the hope that it will be useful,


@@ -16,27 +15,24 @@
    GNU General Public License for more details.
 
    You should have received a copy of the GNU General Public License
    GNU General Public License for more details.
 
    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., 51 Franklin Street, Fifth Floor, Boston, MA
-   02110-1301  USA
-*/
+   along with this program.  If not, see <http://www.gnu.org/licenses/>.  */

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

-#include <strverscmp.h>
-

 #include "complain.h"
 #include "conflicts.h"
 #include "files.h"
 #include "getargs.h"
 #include "gram.h"
 #include "complain.h"
 #include "conflicts.h"
 #include "files.h"
 #include "getargs.h"
 #include "gram.h"

-#include "muscle_tab.h"
+#include "muscle-tab.h"
+#include "named-ref.h"

 #include "quotearg.h"
 #include "reader.h"
 #include "symlist.h"
 #include "scan-gram.h"
 #include "scan-code.h"
 #include "quotearg.h"
 #include "reader.h"
 #include "symlist.h"
 #include "scan-gram.h"
 #include "scan-code.h"

+#include "xmemdup0.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);


@@ -66,10 +62,30 @@ static void add_param (char const *type, char *decl, location loc);
 
 static symbol_class current_class = unknown_sym;
 static uniqstr current_type = NULL;
 
 static symbol_class current_class = unknown_sym;
 static uniqstr current_type = NULL;

-static symbol *current_lhs;
+static symbol *current_lhs_symbol;

 static location current_lhs_location;
 static location current_lhs_location;

+static named_ref *current_lhs_named_ref;

 static int current_prec = 0;
 
 static int current_prec = 0;
 

+/** Set the new current left-hand side symbol, possibly common
+ * to several right-hand side parts of rule.
+ */
+static
+void
+current_lhs(symbol *sym, location loc, named_ref *ref)
+{
+  current_lhs_symbol = sym;
+  current_lhs_location = loc;
+  /* In order to simplify memory management, named references for lhs
+     are always assigned by deep copy into the current symbol_list
+     node.  This is because a single named-ref in the grammar may
+     result in several uses when the user factors lhs between several
+     rules using "|".  Therefore free the parser's original copy.  */
+  free (current_lhs_named_ref);
+  current_lhs_named_ref = ref;
+}
+
+

 #define YYTYPE_INT16 int_fast16_t
 #define YYTYPE_INT8 int_fast8_t
 #define YYTYPE_UINT16 uint_fast16_t
 #define YYTYPE_INT16 int_fast16_t
 #define YYTYPE_INT8 int_fast8_t
 #define YYTYPE_UINT16 uint_fast16_t


@@ -82,7 +98,7 @@ static int current_prec = 0;
 %locations
 %pure-parser
 %error-verbose
 %locations
 %pure-parser
 %error-verbose

-%defines
+%define parse.lac full

 %name-prefix="gram_"
 %expect 0
 
 %name-prefix="gram_"
 %expect 0
 


@@ -104,6 +120,7 @@ static int current_prec = 0;
   assoc assoc;
   uniqstr uniqstr;
   unsigned char character;
   assoc assoc;
   uniqstr uniqstr;
   unsigned char character;

+  named_ref *named_ref;

 };
 
 /* Define the tokens together with their human representation.  */
 };
 
 /* Define the tokens together with their human representation.  */


@@ -154,9 +171,6 @@ static int current_prec = 0;
   PERCENT_OUTPUT          "%output"
   PERCENT_PARSE_PARAM     "%parse-param"
   PERCENT_PURE_PARSER     "%pure-parser"
   PERCENT_OUTPUT          "%output"
   PERCENT_PARSE_PARAM     "%parse-param"
   PERCENT_PURE_PARSER     "%pure-parser"

-  PERCENT_PUSH_PARSER     "%push-parser"
-  PERCENT_PUSH_PULL_PARSER
-                          "%push-pull-parser"

   PERCENT_REQUIRE        "%require"
   PERCENT_SKELETON        "%skeleton"
   PERCENT_START           "%start"
   PERCENT_REQUIRE        "%require"
   PERCENT_SKELETON        "%skeleton"
   PERCENT_START           "%start"


@@ -166,6 +180,7 @@ static int current_prec = 0;
 ;
 
 %token BRACED_CODE     "{...}"
 ;
 
 %token BRACED_CODE     "{...}"

+%token BRACKETED_ID    "[identifier]"

 %token CHAR            "char"
 %token EPILOGUE        "epilogue"
 %token EQUAL           "="
 %token CHAR            "char"
 %token EPILOGUE        "epilogue"
 %token EQUAL           "="


@@ -184,27 +199,30 @@ static int current_prec = 0;
 
 /* braceless is not to be used for rule or symbol actions, as it
    calls code_props_plain_init.  */
 
 /* braceless is not to be used for rule or symbol actions, as it
    calls code_props_plain_init.  */

-%type <chars> STRING "%{...%}" EPILOGUE braceless content content.opt
+%type <chars> STRING "%{...%}" EPILOGUE braceless content.opt

 %type <code> "{...}"
 %printer { fputs (quotearg_style (c_quoting_style, $$), stderr); }
         STRING
 %printer { fprintf (stderr, "{\n%s\n}", $$); }
 %type <code> "{...}"
 %printer { fputs (quotearg_style (c_quoting_style, $$), stderr); }
         STRING
 %printer { fprintf (stderr, "{\n%s\n}", $$); }

-        braceless content content.opt "{...}" "%{...%}" EPILOGUE
+        braceless content.opt "{...}" "%{...%}" EPILOGUE

 
 

-%type <uniqstr> TYPE ID ID_COLON
-%printer { fprintf (stderr, "<%s>", $$); } TYPE
-%printer { fputs ($$, stderr); } ID
+%type <uniqstr> BRACKETED_ID ID ID_COLON TYPE variable
+%printer { fputs ($$, stderr); } <uniqstr>
+%printer { fprintf (stderr, "[%s]", $$); } BRACKETED_ID

 %printer { fprintf (stderr, "%s:", $$); } ID_COLON
 %printer { fprintf (stderr, "%s:", $$); } ID_COLON

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

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

-%printer { fprintf (stderr, "%d", $$); } INT
+%printer { fprintf (stderr, "%d", $$); } <integer>

 
 

-%type <symbol> id id_colon symbol string_as_id
-%printer { fprintf (stderr, "%s", $$->tag); } id symbol string_as_id
+%type <symbol> id id_colon string_as_id symbol symbol.prec
+%printer { fprintf (stderr, "%s", $$->tag); } <symbol>

 %printer { fprintf (stderr, "%s:", $$->tag); } id_colon
 
 %type <assoc> precedence_declarator
 %printer { fprintf (stderr, "%s:", $$->tag); } id_colon
 
 %type <assoc> precedence_declarator

-%type <list>  symbols.1 generic_symlist generic_symlist_item
+%type <list>  symbols.1 symbols.prec generic_symlist generic_symlist_item
+%type <named_ref> named_ref.opt
+

 %%
 
 input:
 %%
 
 input:


@@ -229,11 +247,16 @@ prologue_declaration:
       code_props_plain_init (&plain_code, $1, @1);
       code_props_translate_code (&plain_code);
       gram_scanner_last_string_free ();
       code_props_plain_init (&plain_code, $1, @1);
       code_props_translate_code (&plain_code);
       gram_scanner_last_string_free ();

-      prologue_augment (plain_code.code, @1, union_seen);
+      muscle_code_grow (union_seen ? "post_prologue" : "pre_prologue",
+                        plain_code.code, @1);

       code_scanner_last_string_free ();
     }
 | "%debug"                         { debug_flag = true; }
       code_scanner_last_string_free ();
     }
 | "%debug"                         { debug_flag = true; }

-| "%define" STRING content.opt     { muscle_insert ($2, $3); }
+| "%define" variable content.opt
+    {
+      muscle_percent_define_insert ($2, @2, $3,
+                                    MUSCLE_PERCENT_DEFINE_GRAMMAR_FILE);
+    }

 | "%defines"                       { defines_flag = true; }
 | "%defines" STRING
     {
 | "%defines"                       { defines_flag = true; }
 | "%defines" STRING
     {


@@ -259,7 +282,7 @@ prologue_declaration:
       muscle_code_grow ("initial_action", action.code, @2);
       code_scanner_last_string_free ();
     }
       muscle_code_grow ("initial_action", action.code, @2);
       code_scanner_last_string_free ();
     }

-| "%language" STRING           { language_argmatch ($2, 1, &@1); }
+| "%language" STRING           { language_argmatch ($2, grammar_prio, @1); }

 | "%lex-param" "{...}"         { add_param ("lex_param", $2, @2); }
 | "%locations"                  { locations_flag = true; }
 | "%name-prefix" STRING         { spec_name_prefix = $2; }
 | "%lex-param" "{...}"         { add_param ("lex_param", $2, @2); }
 | "%locations"                  { locations_flag = true; }
 | "%name-prefix" STRING         { spec_name_prefix = $2; }


@@ -269,13 +292,48 @@ prologue_declaration:
 | "%output" STRING              { spec_outfile = $2; }
 | "%output" "=" STRING          { spec_outfile = $3; }  /* deprecated */
 | "%parse-param" "{...}"       { add_param ("parse_param", $2, @2); }
 | "%output" STRING              { spec_outfile = $2; }
 | "%output" "=" STRING          { spec_outfile = $3; }  /* deprecated */
 | "%parse-param" "{...}"       { add_param ("parse_param", $2, @2); }

-| "%pure-parser"                { pure_parser = true; }
-| "%push-parser"                { push_parser = true; pull_parser = false; }
-| "%push-pull-parser"           { push_parser = true; pull_parser = true; }
+| "%pure-parser"
+    {
+      /* %pure-parser is deprecated in favor of `%define api.pure', so use
+         `%define api.pure' in a backward-compatible manner here.  First, don't
+         complain if %pure-parser is specified multiple times.  */
+      if (!muscle_find_const ("percent_define(api.pure)"))
+        muscle_percent_define_insert ("api.pure", @1, "",
+                                      MUSCLE_PERCENT_DEFINE_GRAMMAR_FILE);
+      /* In all cases, use api.pure now so that the backend doesn't complain if
+         the skeleton ignores api.pure, but do warn now if there's a previous
+         conflicting definition from an actual %define.  */
+      if (!muscle_percent_define_flag_if ("api.pure"))
+        muscle_percent_define_insert ("api.pure", @1, "",
+                                      MUSCLE_PERCENT_DEFINE_GRAMMAR_FILE);
+    }

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

-| "%skeleton" STRING            { skeleton_arg ($2, 1, &@1); }
+| "%skeleton" STRING
+    {
+      char const *skeleton_user = $2;
+      if (mbschr (skeleton_user, '/'))
+        {
+          size_t dir_length = strlen (current_file);
+          char *skeleton_build;
+          while (dir_length && current_file[dir_length - 1] != '/')
+            --dir_length;
+          while (dir_length && current_file[dir_length - 1] == '/')
+            --dir_length;
+          skeleton_build =
+            xmalloc (dir_length + 1 + strlen (skeleton_user) + 1);
+          if (dir_length > 0)
+            {
+              memcpy (skeleton_build, current_file, dir_length);
+              skeleton_build[dir_length++] = '/';
+            }
+          strcpy (skeleton_build + dir_length, skeleton_user);
+          skeleton_user = uniqstr_new (skeleton_build);
+          free (skeleton_build);
+        }
+      skeleton_arg (skeleton_user, grammar_prio, @1);
+    }

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

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

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


@@ -311,21 +369,15 @@ grammar_declaration:
     }
 | "%code" braceless
     {
     }
 | "%code" braceless
     {

-      muscle_code_grow ("percent_code", $2, @2);
+      /* Do not invoke muscle_percent_code_grow here since it invokes
+         muscle_user_name_list_grow.  */
+      muscle_code_grow ("percent_code()", $2, @2);

       code_scanner_last_string_free ();
     }
       code_scanner_last_string_free ();
     }

-| "%code" STRING braceless
+| "%code" ID braceless

     {
     {

-      char const name_prefix[] = "percent_code_";
-      char *name = xmalloc (sizeof name_prefix + strlen ($2));
-      strcpy (name, name_prefix);
-      strcpy (name + sizeof name_prefix - 1, $2);
-      muscle_code_grow (uniqstr_new (name), $3, @3);
-      free (name);
+      muscle_percent_code_grow ($2, @2, $3, @3);

       code_scanner_last_string_free ();
       code_scanner_last_string_free ();

-      muscle_grow ("used_percent_code_qualifiers", "[[", ",");
-      muscle_grow ("used_percent_code_qualifiers", $2, "");
-      muscle_grow ("used_percent_code_qualifiers", "]]", "");

     }
 ;
 
     }
 ;
 


@@ -342,28 +394,11 @@ union_name:
 ;
 
 grammar_declaration:
 ;
 
 grammar_declaration:

-  "%union" union_name "{...}"
+  "%union" union_name braceless

     {
     {

-      char const *body = $3;
-
-      /* Concatenate the %union bodies.  If this is the first %union, make sure
-        the synchronization line appears after the opening '{' so as not to
-        confuse Doxygen.  Otherwise, turn the previous %union's trailing '}'
-        into '\n', and omit the new %union's leading '{'.  */
-      if (!union_seen)
-       {
-         muscle_grow ("stype", "{", "");
-       }
-      else
-       {
-         char *code = muscle_find ("stype");
-         code[strlen (code) - 1] = '\n';
-       }
-      body++;
-

       union_seen = true;
       union_seen = true;

-      muscle_code_grow ("stype", body, @3);
-      gram_scanner_last_string_free ();
+      muscle_code_grow ("stype", $3, @3);
+      code_scanner_last_string_free ();

     }
 ;
 
     }
 ;
 


@@ -392,7 +427,7 @@ symbol_declaration:
 ;
 
 precedence_declaration:
 ;
 
 precedence_declaration:

-  precedence_declarator type.opt symbols.1
+  precedence_declarator type.opt symbols.prec

     {
       symbol_list *list;
       ++current_prec;
     {
       symbol_list *list;
       ++current_prec;


@@ -417,6 +452,19 @@ type.opt:
 | TYPE           { current_type = $1; tag_seen = true; }
 ;
 
 | TYPE           { current_type = $1; tag_seen = true; }
 ;
 

+/* Just like symbols.1 but accept INT for the sake of POSIX.  */
+symbols.prec:
+  symbol.prec
+    { $$ = symbol_list_sym_new ($1, @1); }
+| symbols.prec symbol.prec
+    { $$ = symbol_list_prepend ($1, symbol_list_sym_new ($2, @2)); }
+;
+
+symbol.prec:
+    symbol { $$ = $1; }
+  | symbol INT { $$ = $1; symbol_user_token_number_set ($1, $2, @2); }
+  ;
+

 /* One or more symbols to be %typed. */
 symbols.1:
   symbol
 /* One or more symbols to be %typed. */
 symbols.1:
   symbol


@@ -498,7 +546,11 @@ rules_or_grammar_declaration:
 ;
 
 rules:
 ;
 
 rules:

-  id_colon { current_lhs = $1; current_lhs_location = @1; } rhses.1
+  id_colon named_ref.opt { current_lhs ($1, @1, $2); } rhses.1
+  {
+    /* Free the current lhs. */
+    current_lhs (0, @1, 0);
+  }

 ;
 
 rhses.1:
 ;
 
 rhses.1:


@@ -509,11 +561,12 @@ rhses.1:
 
 rhs:
   /* Nothing.  */
 
 rhs:
   /* Nothing.  */

-    { grammar_current_rule_begin (current_lhs, current_lhs_location); }
-| rhs symbol
-    { grammar_current_rule_symbol_append ($2, @2); }
-| rhs "{...}"
-    { grammar_current_rule_action_append ($2, @2); }
+    { grammar_current_rule_begin (current_lhs_symbol, current_lhs_location,
+                                 current_lhs_named_ref); }
+| rhs symbol named_ref.opt
+    { grammar_current_rule_symbol_append ($2, @2, $3); }
+| rhs "{...}" named_ref.opt
+    { grammar_current_rule_action_append ($2, @2, $3); }

 | 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


@@ -522,27 +575,36 @@ rhs:
     { grammar_current_rule_merge_set ($3, @3); }
 ;
 
     { grammar_current_rule_merge_set ($3, @3); }
 ;
 

+named_ref.opt:
+  /* Nothing. */ { $$ = 0; }
+|
+  BRACKETED_ID   { $$ = named_ref_new($1, @1); }
+;

 
 

-/*-----------*
- | content.  |
- *-----------*/

 
 

-content:
-  STRING
-| braceless
+/*----------------------------*
+ | variable and content.opt.  |
+ *---------------------------*/
+
+/* The STRING form of variable is deprecated and is not M4-friendly.
+   For example, M4 fails for `%define "[" "value"'.  */
+variable:
+  ID
+| STRING { $$ = uniqstr_new ($1); }

 ;
 
 ;
 

-/* Some content or "1" by default. */
+/* Some content or empty by default. */

 content.opt:
 content.opt:

-  /* Nothing. */
-    {
-      static char one[] = "1";
-      $$ = one;
-    }
-| content
+  /* Nothing. */   { $$ = ""; }
+| ID { $$ = $1; }
+| STRING

 ;
 
 
 ;
 
 

+/*-------------*
+ | braceless.  |
+ *-------------*/
+

 braceless:
   "{...}"
     {
 braceless:
   "{...}"
     {


@@ -625,7 +687,7 @@ lloc_default (YYLTYPE const *rhs, int n)
   loc.start = rhs[n].end;
   loc.end = rhs[n].end;
 
   loc.start = rhs[n].end;
   loc.end = rhs[n].end;
 

-  /* Ignore empty nonterminals the start of the the right-hand side.
+  /* Ignore empty nonterminals the start of the right-hand side.

      Do not bother to ignore them at the end of the right-hand side,
      since empty nonterminals have the same end as their predecessors.  */
   for (i = 1; i <= n; i++)
      Do not bother to ignore them at the end of the right-hand side,
      since empty nonterminals have the same end as their predecessors.  */
   for (i = 1; i <= n; i++)


@@ -672,17 +734,7 @@ add_param (char const *type, char *decl, location loc)
     complain_at (loc, _("missing identifier in parameter declaration"));
   else
     {
     complain_at (loc, _("missing identifier in parameter declaration"));
   else
     {

-      char *name;
-      size_t name_len;
-
-      for (name_len = 1;
-          memchr (alphanum, name_start[name_len], sizeof alphanum);
-          name_len++)
-       continue;
-
-      name = xmalloc (name_len + 1);
-      memcpy (name, name_start, name_len);
-      name[name_len] = '\0';
+      char *name = xmemdup0 (name_start, strspn (name_start, alphanum));

       muscle_pair_list_grow (type, decl, name);
       free (name);
     }
       muscle_pair_list_grow (type, decl, name);
       free (name);
     }


@@ -697,8 +749,8 @@ version_check (location const *loc, char const *version)
   if (strverscmp (version, PACKAGE_VERSION) > 0)
     {
       complain_at (*loc, "require bison %s, but have %s",
   if (strverscmp (version, PACKAGE_VERSION) > 0)
     {
       complain_at (*loc, "require bison %s, but have %s",

-                  version, PACKAGE_VERSION);
-      exit (63);
+                   version, PACKAGE_VERSION);
+      exit (EX_MISMATCH);

     }
 }
 
     }
 }