glr.cc, yacc.c: initialize yylloc properly

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

diff --git a/src/parse-gram.y b/src/parse-gram.y
index 18e76fa8f532e4e8fed9c2ae045845fd2a78058d..5f77a5bd08abd0355531dadbbebe53503442c77a 100644 (file)
--- a/src/parse-gram.y
+++ b/src/parse-gram.y
@@ -1,7 +1,6 @@
 %{/* 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.
 


@@ -21,19 +20,20 @@
 #include <config.h>
 #include "system.h"
 
 #include <config.h>
 #include "system.h"
 

-#include <strverscmp.h>
-
+#include "c-ctype.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);


@@ -63,10 +63,30 @@ 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


@@ -74,13 +94,14 @@ static int current_prec = 0;
 %}
 
 %debug
 %}
 
 %debug

-%verbose
+%define api.prefix "gram_"
+%define api.pure
+%define parse.lac full

 %defines
 %defines

-%locations
-%pure-parser

 %error-verbose
 %error-verbose

-%name-prefix="gram_"

 %expect 0
 %expect 0

+%locations
+%verbose

 
 %initial-action
 {
 
 %initial-action
 {


@@ -100,6 +121,7 @@ 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.  */


@@ -150,9 +172,6 @@ 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"


@@ -162,6 +181,7 @@ 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           "="


@@ -187,20 +207,23 @@ static int current_prec = 0;
 %printer { fprintf (stderr, "{\n%s\n}", $$); }
         braceless content.opt "{...}" "%{...%}" EPILOGUE
 
 %printer { fprintf (stderr, "{\n%s\n}", $$); }
         braceless content.opt "{...}" "%{...%}" EPILOGUE
 

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


@@ -229,10 +252,11 @@ prologue_declaration:
                         plain_code.code, @1);
       code_scanner_last_string_free ();
     }
                         plain_code.code, @1);
       code_scanner_last_string_free ();
     }

-| "%debug"                         { debug_flag = true; }
+| "%debug"                         { debug = true; }

 | "%define" variable content.opt
     {
 | "%define" variable content.opt
     {

-      muscle_percent_define_insert ($2, @2, $3);
+      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 +283,7 @@ 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; }


@@ -269,14 +293,26 @@ 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); }
 | "%skeleton" STRING
     {
       char const *skeleton_user = $2;
 | "%require" STRING             { version_check (&@2, $2); }
 | "%skeleton" STRING
     {
       char const *skeleton_user = $2;

-      if (strchr (skeleton_user, '/'))
+      if (mbschr (skeleton_user, '/'))

         {
           size_t dir_length = strlen (current_file);
           char *skeleton_build;
         {
           size_t dir_length = strlen (current_file);
           char *skeleton_build;


@@ -288,17 +324,17 @@ prologue_declaration:
             xmalloc (dir_length + 1 + strlen (skeleton_user) + 1);
           if (dir_length > 0)
             {
             xmalloc (dir_length + 1 + strlen (skeleton_user) + 1);
           if (dir_length > 0)
             {

-              strncpy (skeleton_build, current_file, dir_length);
+              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_build[dir_length++] = '/';
             }
           strcpy (skeleton_build + dir_length, skeleton_user);
           skeleton_user = uniqstr_new (skeleton_build);
           free (skeleton_build);
         }

-      skeleton_arg (skeleton_user, 1, &@1);
+      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? */ ";"
 ;


@@ -312,17 +348,27 @@ grammar_declaration:
     }
 | "%destructor" "{...}" generic_symlist
     {
     }
 | "%destructor" "{...}" generic_symlist
     {

-      symbol_list *list;
-      for (list = $3; list; list = list->next)
-       symbol_list_destructor_set (list, $2, @2);
-      symbol_list_free ($3);
+      code_props code;
+      code_props_symbol_action_init (&code, $2, @2);
+      code_props_translate_code (&code);
+      {
+        symbol_list *list;
+        for (list = $3; list; list = list->next)
+          symbol_list_destructor_set (list, &code);
+        symbol_list_free ($3);
+      }

     }
 | "%printer" "{...}" generic_symlist
     {
     }
 | "%printer" "{...}" generic_symlist
     {

-      symbol_list *list;
-      for (list = $3; list; list = list->next)
-       symbol_list_printer_set (list, $2, @2);
-      symbol_list_free ($3);
+      code_props code;
+      code_props_symbol_action_init (&code, $2, @2);
+      code_props_translate_code (&code);
+      {
+        symbol_list *list;
+        for (list = $3; list; list = list->next)
+          symbol_list_printer_set (list, &code);
+        symbol_list_free ($3);
+      }

     }
 | "%default-prec"
     {
     }
 | "%default-prec"
     {


@@ -392,7 +438,7 @@ 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 +463,19 @@ 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 +557,11 @@ 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 +572,12 @@ 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


@@ -522,22 +586,28 @@ 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); }
+;
+

 
 /*----------------------------*
  | variable and content.opt.  |
  *---------------------------*/
 
 
 /*----------------------------*
  | 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
 variable:
   ID

-  | STRING { $$ = uniqstr_new ($1); } /* deprecated and not M4-friendly */
-  ;
+| STRING { $$ = uniqstr_new ($1); }
+;

 
 /* Some content or empty by default. */
 content.opt:
 
 /* Some content or empty by default. */
 content.opt:

-  /* Nothing. */
-    {
-      $$ = "";
-    }
+  /* Nothing. */   { $$ = ""; }
+| ID { $$ = $1; }

 | STRING
 ;
 
 | STRING
 ;
 


@@ -628,7 +698,7 @@ 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++)


@@ -665,27 +735,19 @@ add_param (char const *type, char *decl, location loc)
 
   /* Strip the surrounding '{' and '}', and any blanks just inside
      the braces.  */
 
   /* Strip the surrounding '{' and '}', and any blanks just inside
      the braces.  */

-  while (*--p == ' ' || *p == '\t')
-    continue;
+  --p;
+  while (c_isspace ((unsigned char) *p))
+    --p;

   p[1] = '\0';
   p[1] = '\0';

-  while (*++decl == ' ' || *decl == '\t')
-    continue;
+  ++decl;
+  while (c_isspace ((unsigned char) *decl))
+    ++decl;

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


@@ -700,8 +762,8 @@ 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);

     }
 }
 
     }
 }