Allow specification of semantic predicates.

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

diff --git a/src/parse-gram.y b/src/parse-gram.y
index 7277105658e95f8152e22701e5c166164515091d..d588dfa8ab0f2ba015414a37cf7f8059e058d6cd 100644 (file)
--- a/src/parse-gram.y
+++ b/src/parse-gram.y
@@ -1,7 +1,7 @@
 %{/* Bison Grammar Parser                             -*- C -*-
 
 %{/* Bison Grammar Parser                             -*- C -*-
 

-   Copyright (C) 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009 Free Software
-   Foundation, Inc.
+   Copyright (C) 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010
+   Free Software Foundation, Inc.

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


@@ -26,7 +26,8 @@
 #include "files.h"
 #include "getargs.h"
 #include "gram.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 "quotearg.h"
 #include "reader.h"
 #include "symlist.h"


@@ -49,34 +50,29 @@ static void version_check (location const *loc, char const *version);
 static void gram_error (location const *, char const *);
 
 static char const *char_name (char);
 static void gram_error (location const *, char const *);
 
 static char const *char_name (char);

-
-/** Add a lex-param or a parse-param.
- *
- * \param type  \a lex_param or \a parse_param
- * \param decl  the formal argument
- * \param loc   the location in the source.
- */
-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 *current_lhs;
-static location current_lhs_location;
-static int current_prec = 0;
-
-#define YYTYPE_INT16 int_fast16_t
-#define YYTYPE_INT8 int_fast8_t
-#define YYTYPE_UINT16 uint_fast16_t
-#define YYTYPE_UINT8 uint_fast8_t

 %}
 
 %}
 

+%code
+{
+  static int current_prec = 0;
+  static location current_lhs_location;
+  static named_ref *current_lhs_named_ref;
+  static symbol *current_lhs;
+  static symbol_class current_class = unknown_sym;
+  static uniqstr current_type = NULL;
+
+  #define YYTYPE_INT16 int_fast16_t
+  #define YYTYPE_INT8 int_fast8_t
+  #define YYTYPE_UINT16 uint_fast16_t
+  #define YYTYPE_UINT8 uint_fast8_t
+}
+

 %debug
 %verbose
 %defines
 %locations
 %pure-parser
 %debug
 %verbose
 %defines
 %locations
 %pure-parser

-%error-verbose
+%define parse.error "verbose"

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


@@ -90,12 +86,13 @@ static int current_prec = 0;
 
 %union
 {
 
 %union
 {

+  assoc assoc;
+  char *code;
+  char const *chars;
+  int integer;
+  named_ref *named_ref;

   symbol *symbol;
   symbol_list *list;
   symbol *symbol;
   symbol_list *list;

-  int integer;
-  char const *chars;
-  char *code;
-  assoc assoc;

   uniqstr uniqstr;
   unsigned char character;
 };
   uniqstr uniqstr;
   unsigned char character;
 };


@@ -121,7 +118,6 @@ static int current_prec = 0;
 %token PERCENT_DPREC         "%dprec"
 %token PERCENT_MERGE         "%merge"
 
 %token PERCENT_DPREC         "%dprec"
 %token PERCENT_MERGE         "%merge"
 

-

 /*----------------------.
 | Global Declarations.  |
 `----------------------*/
 /*----------------------.
 | Global Declarations.  |
 `----------------------*/


@@ -131,23 +127,21 @@ static int current_prec = 0;
   PERCENT_DEFAULT_PREC    "%default-prec"
   PERCENT_DEFINE          "%define"
   PERCENT_DEFINES         "%defines"
   PERCENT_DEFAULT_PREC    "%default-prec"
   PERCENT_DEFINE          "%define"
   PERCENT_DEFINES         "%defines"

+  PERCENT_ERROR_VERBOSE   "%error-verbose"

   PERCENT_EXPECT          "%expect"
   PERCENT_EXPECT          "%expect"

-  PERCENT_EXPECT_RR      "%expect-rr"
+  PERCENT_EXPECT_RR       "%expect-rr"

   PERCENT_FLAG            "%<flag>"
   PERCENT_FILE_PREFIX     "%file-prefix"
   PERCENT_GLR_PARSER      "%glr-parser"
   PERCENT_INITIAL_ACTION  "%initial-action"
   PERCENT_LANGUAGE        "%language"
   PERCENT_FLAG            "%<flag>"
   PERCENT_FILE_PREFIX     "%file-prefix"
   PERCENT_GLR_PARSER      "%glr-parser"
   PERCENT_INITIAL_ACTION  "%initial-action"
   PERCENT_LANGUAGE        "%language"

-  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
   PERCENT_NAME_PREFIX     "%name-prefix"
   PERCENT_NO_DEFAULT_PREC "%no-default-prec"
   PERCENT_NO_LINES        "%no-lines"
   PERCENT_NONDETERMINISTIC_PARSER

-                         "%nondeterministic-parser"
+                          "%nondeterministic-parser"

   PERCENT_OUTPUT          "%output"
   PERCENT_OUTPUT          "%output"

-  PERCENT_PARSE_PARAM     "%parse-param"
-  PERCENT_REQUIRE        "%require"
+  PERCENT_REQUIRE         "%require"

   PERCENT_SKELETON        "%skeleton"
   PERCENT_START           "%start"
   PERCENT_TOKEN_TABLE     "%token-table"
   PERCENT_SKELETON        "%skeleton"
   PERCENT_START           "%start"
   PERCENT_TOKEN_TABLE     "%token-table"


@@ -156,6 +150,8 @@ static int current_prec = 0;
 ;
 
 %token BRACED_CODE     "{...}"
 ;
 
 %token BRACED_CODE     "{...}"

+%token BRACED_PREDICATE "%?{...}"
+%token BRACKETED_ID    "[identifier]"

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


@@ -175,27 +171,80 @@ static int current_prec = 0;
 /* braceless is not to be used for rule or symbol actions, as it
    calls code_props_plain_init.  */
 %type <chars> STRING "%{...%}" EPILOGUE braceless content.opt
 /* braceless is not to be used for rule or symbol actions, as it
    calls code_props_plain_init.  */
 %type <chars> STRING "%{...%}" EPILOGUE braceless content.opt

-%type <code> "{...}"
+%type <code> "{...}" "%?{...}"

 %printer { fputs (quotearg_style (c_quoting_style, $$), stderr); }
         STRING
 %printer { fprintf (stderr, "{\n%s\n}", $$); }
         braceless content.opt "{...}" "%{...%}" EPILOGUE
 
 %printer { fputs (quotearg_style (c_quoting_style, $$), stderr); }
         STRING
 %printer { fprintf (stderr, "{\n%s\n}", $$); }
         braceless content.opt "{...}" "%{...%}" EPILOGUE
 

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

 %printer { fprintf (stderr, "%s:", $$); } ID_COLON
 %printer { fprintf (stderr, "%%%s", $$); } PERCENT_FLAG
 %printer { fprintf (stderr, "<%s>", $$); } TAG
 
 %type <integer> INT
 %printer { fprintf (stderr, "%s:", $$); } ID_COLON
 %printer { fprintf (stderr, "%%%s", $$); } PERCENT_FLAG
 %printer { fprintf (stderr, "<%s>", $$); } TAG
 
 %type <integer> INT

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

 
 

-%type <symbol> id id_colon symbol symbol.prec 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
 %type <list>  symbols.1 symbols.prec generic_symlist generic_symlist_item
 %printer { fprintf (stderr, "%s:", $$->tag); } id_colon
 
 %type <assoc> precedence_declarator
 %type <list>  symbols.1 symbols.prec generic_symlist generic_symlist_item

+%type <named_ref> named_ref.opt
+
+/*---------.
+| %param.  |
+`---------*/
+%code requires
+{
+# ifndef PARAM_TYPE
+#  define PARAM_TYPE
+  typedef enum
+  {
+    param_none   = 0,
+    param_lex    = 1 << 0,
+    param_parse  = 1 << 1,
+    param_both   = param_lex | param_parse
+  } param_type;
+# endif
+};
+%code
+{
+  /** Add a lex-param and/or a parse-param.
+   *
+   * \param type  where to push this formal argument.
+   * \param decl  the formal argument.  Destroyed.
+   * \param loc   the location in the source.
+   */
+  static void add_param (param_type type, char *decl, location loc);
+  static param_type current_param = param_none;
+};
+%union
+{
+  param_type param;
+}
+%token <param> PERCENT_PARAM "%param";
+%printer
+{
+  switch ($$)
+    {
+#define CASE(In, Out)                                           \
+      case param_ ## In: fputs ("%" #Out, stderr); break
+      CASE(lex,   lex-param);
+      CASE(parse, parse-param);
+      CASE(both,  param);
+#undef CASE
+      case param_none: aver (false); break;
+    }
+} <param>;
+
+
+                     /*==========\
+                     | Grammar.  |
+                     \==========*/

 %%
 
 input:
 %%
 
 input:


@@ -230,7 +279,8 @@ prologue_declaration:
     }
 | "%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


@@ -238,6 +288,11 @@ prologue_declaration:
       defines_flag = true;
       spec_defines_file = xstrdup ($2);
     }
       defines_flag = true;
       spec_defines_file = xstrdup ($2);
     }

+| "%error-verbose"
+    {
+      muscle_percent_define_insert ("parse.error", @1, "verbose",
+                                    MUSCLE_PERCENT_DEFINE_GRAMMAR_FILE);
+    }

 | "%expect" INT                    { expected_sr_conflicts = $2; }
 | "%expect-rr" INT                { expected_rr_conflicts = $2; }
 | "%file-prefix" STRING            { spec_file_prefix = $2; }
 | "%expect" INT                    { expected_sr_conflicts = $2; }
 | "%expect-rr" INT                { expected_rr_conflicts = $2; }
 | "%file-prefix" STRING            { spec_file_prefix = $2; }


@@ -257,15 +312,13 @@ prologue_declaration:
       code_scanner_last_string_free ();
     }
 | "%language" STRING           { language_argmatch ($2, grammar_prio, @1); }
       code_scanner_last_string_free ();
     }
 | "%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; }
 | "%name-prefix" "=" STRING     { spec_name_prefix = $3; } /* deprecated */
 | "%no-lines"                   { no_lines_flag = true; }
 | "%nondeterministic-parser"   { nondeterministic_parser = true; }
 | "%output" STRING              { spec_outfile = $2; }
 | "%output" "=" STRING          { spec_outfile = $3; }  /* deprecated */
 | "%name-prefix" STRING         { spec_name_prefix = $2; }
 | "%name-prefix" "=" STRING     { spec_name_prefix = $3; } /* deprecated */
 | "%no-lines"                   { no_lines_flag = true; }
 | "%nondeterministic-parser"   { nondeterministic_parser = true; }
 | "%output" STRING              { spec_outfile = $2; }
 | "%output" "=" STRING          { spec_outfile = $3; }  /* deprecated */

-| "%parse-param" "{...}"       { add_param ("parse_param", $2, @2); }
+| "%param" { current_param = $1; } params { current_param = param_none; }

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


@@ -297,6 +350,16 @@ prologue_declaration:
 | /*FIXME: Err?  What is this horror doing here? */ ";"
 ;
 
 | /*FIXME: Err?  What is this horror doing here? */ ";"
 ;
 

+params:
+   params "{...}"  { add_param (current_param, $2, @2); }
+| "{...}"          { add_param (current_param, $1, @1); }
+;
+
+
+/*----------------------.
+| grammar_declaration.  |
+`----------------------*/
+

 grammar_declaration:
   precedence_declaration
 | symbol_declaration
 grammar_declaration:
   precedence_declaration
 | symbol_declaration


@@ -506,7 +569,8 @@ 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; current_lhs_location = @1;
+    current_lhs_named_ref = $2; } rhses.1

 ;
 
 rhses.1:
 ;
 
 rhses.1:


@@ -517,11 +581,14 @@ 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, 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, false); }
+| rhs "%?{...}" 
+    { grammar_current_rule_action_append ($2, @2, NULL, true); }

 | 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


@@ -530,22 +597,27 @@ 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
 ;
 


@@ -650,34 +722,33 @@ lloc_default (YYLTYPE const *rhs, int n)
 }
 
 
 }
 
 

-/* Add a lex-param or a parse-param (depending on TYPE) with
-   declaration DECL and location LOC.  */
-

 static void
 static void

-add_param (char const *type, char *decl, location loc)
+add_param (param_type type, char *decl, location loc)

 {
   static char const alphanum[26 + 26 + 1 + 10] =
     "abcdefghijklmnopqrstuvwxyz"
     "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
     "_"
     "0123456789";
 {
   static char const alphanum[26 + 26 + 1 + 10] =
     "abcdefghijklmnopqrstuvwxyz"
     "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
     "_"
     "0123456789";

+

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

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


@@ -694,7 +765,10 @@ add_param (char const *type, char *decl, location loc)
       name = xmalloc (name_len + 1);
       memcpy (name, name_start, name_len);
       name[name_len] = '\0';
       name = xmalloc (name_len + 1);
       memcpy (name, name_start, name_len);
       name[name_len] = '\0';

-      muscle_pair_list_grow (type, decl, name);
+      if (type & param_lex)
+        muscle_pair_list_grow ("lex_param", decl, name);
+      if (type & param_parse)
+        muscle_pair_list_grow ("parse_param", decl, name);

       free (name);
     }
 
       free (name);
     }