* tests/actions.at (Actions after errors): New test case.

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

diff --git a/src/parse-gram.y b/src/parse-gram.y
index 663fd8e75372d148d0599ad89859ed5cc3128c69..8af3e88501c802e8b7f8b0569a9666712eee3ac3 100644 (file)
--- a/src/parse-gram.y
+++ b/src/parse-gram.y
@@ -30,16 +30,32 @@
 
 %{
 #include "system.h"
 
 %{
 #include "system.h"

+#include "complain.h"

 #include "muscle_tab.h"
 #include "files.h"
 #include "getargs.h"
 #include "output.h"
 #include "muscle_tab.h"
 #include "files.h"
 #include "getargs.h"
 #include "output.h"

+#include "symlist.h"

 #include "gram.h"
 #include "reader.h"
 #include "conflicts.h"
 
 #include "gram.h"
 #include "reader.h"
 #include "conflicts.h"
 

-/* Produce verbose parse errors.  */
+/* Produce verbose syntax errors.  */

 #define YYERROR_VERBOSE 1
 #define YYERROR_VERBOSE 1

+#define YYLLOC_DEFAULT(Current, Rhs, N)                        \
+do {                                                   \
+  if (N)                                               \
+  {                                                    \
+    Current.first_column  = Rhs[1].first_column;       \
+    Current.first_line    = Rhs[1].first_line;         \
+    Current.last_column   = Rhs[N].last_column;                \
+    Current.last_line     = Rhs[N].last_line;          \
+  }                                                    \
+  else                                                 \
+  {                                                    \
+    Current = Rhs[0];                                  \
+  }                                                    \
+} while (0)

 
 /* Pass the control structure to YYPARSE and YYLEX. */
 #define YYPARSE_PARAM gram_control
 
 /* Pass the control structure to YYPARSE and YYLEX. */
 #define YYPARSE_PARAM gram_control


@@ -48,24 +64,23 @@
    correctly typed access to it.  */
 #define yycontrol ((gram_control_t *) gram_control)
 
    correctly typed access to it.  */
 #define yycontrol ((gram_control_t *) gram_control)
 

-/* Request detailed parse error messages, and pass them to
-   GRAM_ERROR. */
+/* Request detailed syntax error messages, and pass them to GRAM_ERROR.
+   FIXME: depends on the undocumented availability of YYLLOC.t */

 #undef  yyerror
 #define yyerror(Msg) \
 #undef  yyerror
 #define yyerror(Msg) \

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

 
 

-/* When debugging our pure parser, we want to see values and locations
-   of the tokens.  */

 #define YYPRINT(File, Type, Value) \
 #define YYPRINT(File, Type, Value) \

-        yyprint (File, &yylloc, Type, &Value)
-static void yyprint (FILE *file, const location_t *loc,
-                     int type, const yystype *value);
+        yyprint (File, Type, &Value)
+static void yyprint (FILE *file, int type, const yystype *value);

 
 symbol_class current_class = unknown_sym;
 
 symbol_class current_class = unknown_sym;

-char *current_type = 0;
+struniq_t current_type = 0;

 symbol_t *current_lhs;
 symbol_t *current_lhs;

-associativity current_assoc;
+location_t current_lhs_location;
+assoc_t current_assoc;

 int current_prec = 0;
 int current_prec = 0;

+braced_code_t current_braced_code = action_braced_code;

 %}
 
 
 %}
 
 


@@ -73,91 +88,126 @@ int current_prec = 0;
 %union
 {
   symbol_t *symbol;
 %union
 {
   symbol_t *symbol;

+  symbol_list_t *list;

   int integer;
   char *string;
   int integer;
   char *string;

-  associativity assoc;
+  assoc_t assoc;
+  struniq_t struniq;

 };
 
 };
 

-/* Define the tokens together with there human representation. */
-%token GRAM_EOF 0 "end of string"
-%token STRING CHARACTER
-%token INT
-
-%token PERCENT_TOKEN "%token"
-%token PERCENT_NTERM "%nterm"
-%token PERCENT_TYPE "%type"
-%token PERCENT_UNION "%union"
-%token PERCENT_EXPECT "%expect"
-%token PERCENT_START "%start"
-%token PERCENT_LEFT "%left"
-%token PERCENT_RIGHT "%right"
-%token PERCENT_NONASSOC "%nonassoc"
-%token PERCENT_PREC     "%prec"
-%token PERCENT_VERBOSE  "%verbose"
-%token PERCENT_ERROR_VERBOSE "%error-verbose"
-
-%token PERCENT_OUTPUT "%output"
-%token PERCENT_FILE_PREFIX "%file-prefix"
-%token PERCENT_NAME_PREFIX "%name-prefix"
-
-%token PERCENT_DEFINE "%define"
-%token PERCENT_PURE_PARSER "%pure-parser"
-
-%token PERCENT_DEFINES "%defines"
-
-%token PERCENT_YACC "%yacc"
-
-%token PERCENT_DEBUG "%debug"
-%token PERCENT_LOCATIONS "%locations"
-%token PERCENT_NO_LINES "%no-lines"
-%token PERCENT_SKELETON "%skeleton"
-%token PERCENT_TOKEN_TABLE "%token-table"
-
-%token TYPE
-%token EQUAL "="
-%token SEMICOLON ";"
-%token COLON ":"
-%token PIPE "|"
-%token ID "identifier"
+/* Define the tokens together with their human representation.  */
+%token GRAM_EOF 0 "end of file"
+%token STRING     "string"
+%token INT        "integer"
+
+%token PERCENT_TOKEN       "%token"
+%token PERCENT_NTERM       "%nterm"
+
+%token PERCENT_TYPE        "%type"
+%token PERCENT_DESTRUCTOR  "%destructor"
+%token PERCENT_PRINTER     "%printer"
+
+%token PERCENT_UNION       "%union"
+
+%token PERCENT_LEFT        "%left"
+%token PERCENT_RIGHT       "%right"
+%token PERCENT_NONASSOC    "%nonassoc"
+
+%token PERCENT_PREC          "%prec"
+%token PERCENT_DPREC         "%dprec"
+%token PERCENT_MERGE         "%merge"
+
+
+/*----------------------.
+| Global Declarations.  |
+`----------------------*/
+
+%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_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"
+;
+
+%token TYPE            "type"
+%token EQUAL           "="
+%token SEMICOLON       ";"
+%token COLON           ":"
+%token COMMA           ","
+%token PIPE            "|"
+%token ID              "identifier"

 %token PERCENT_PERCENT "%%"
 %token PERCENT_PERCENT "%%"

-%token PROLOGUE EPILOGUE
-%token BRACED_CODE
+%token PROLOGUE        "%{...%}"
+%token EPILOGUE        "epilogue"
+%token BRACED_CODE     "{...}"

 
 

-%type <string> CHARACTER TYPE  STRING string_content
-               BRACED_CODE PROLOGUE EPILOGUE epilogue.opt action
+
+%type <string> STRING string_content
+               BRACED_CODE code_content action
+               PROLOGUE EPILOGUE
+%type <struniq> TYPE

 %type <integer> INT
 %type <symbol> ID symbol string_as_id
 %type <integer> INT
 %type <symbol> ID symbol string_as_id

-%type <assoc> precedence_directive
+%type <assoc> precedence_declarator
+%type <list>  symbols.1

 %%
 %%

-input: { LOCATION_RESET (yylloc); }
-  directives "%%" gram epilogue.opt
+
+input:
+  declarations "%%" grammar epilogue.opt

     {
       yycontrol->errcode = 0;
     {
       yycontrol->errcode = 0;

-      epilogue_set ($5, @5);

     }
 ;
 
     }
 ;
 

-directives:
+
+       /*------------------------------------.
+       | Declarations: before the first %%.  |
+       `------------------------------------*/
+
+declarations:

   /* Nothing */
   /* Nothing */

-| directives directive
+| declarations declaration semi_colon.opt

 ;
 
 ;
 

-directive:
-  grammar_directives
-| PROLOGUE
-   {
-     prologue_augment ($1, @1);
-   }
+declaration:
+  grammar_declaration
+| PROLOGUE                                 { prologue_augment ($1, @1); }

 | "%debug"                                 { debug_flag = 1; }
 | "%define" string_content string_content  { muscle_insert ($2, $3); }
 | "%defines"                               { defines_flag = 1; }
 | "%error-verbose"                         { error_verbose = 1; }
 | "%expect" INT                            { expected_conflicts = $2; }
 | "%file-prefix" "=" string_content        { spec_file_prefix = $3; }
 | "%debug"                                 { debug_flag = 1; }
 | "%define" string_content string_content  { muscle_insert ($2, $3); }
 | "%defines"                               { defines_flag = 1; }
 | "%error-verbose"                         { error_verbose = 1; }
 | "%expect" INT                            { expected_conflicts = $2; }
 | "%file-prefix" "=" string_content        { spec_file_prefix = $3; }

+| "%glr-parser"                           { glr_parser = 1; }
+| "%lex-param" code_content "," code_content
+                           {
+                            muscle_pair_list_grow ("lex_param", $2, $4);
+                            scanner_last_string_free ();
+                          }

 | "%locations"                             { locations_flag = 1; }
 | "%name-prefix" "=" string_content        { spec_name_prefix = $3; }
 | "%no-lines"                              { no_lines_flag = 1; }
 | "%output" "=" string_content             { spec_outfile = $3; }
 | "%locations"                             { locations_flag = 1; }
 | "%name-prefix" "=" string_content        { spec_name_prefix = $3; }
 | "%no-lines"                              { no_lines_flag = 1; }
 | "%output" "=" string_content             { spec_outfile = $3; }

+| "%parse-param" code_content "," code_content
+                           {
+                            muscle_pair_list_grow ("parse_param", $2, $4);
+                            scanner_last_string_free ();
+                          }

 | "%pure-parser"                           { pure_parser = 1; }
 | "%skeleton" string_content               { skeleton = $2; }
 | "%token-table"                           { token_table_flag = 1; }
 | "%pure-parser"                           { pure_parser = 1; }
 | "%skeleton" string_content               { skeleton = $2; }
 | "%token-table"                           { token_table_flag = 1; }


@@ -165,73 +215,94 @@ directive:
 | "%yacc"                                  { yacc_flag = 1; }
 ;
 
 | "%yacc"                                  { yacc_flag = 1; }
 ;
 

-grammar_directives:
-  precedence_directives
-| "%nterm" { current_class = nterm_sym; } symbol_defs.1
+grammar_declaration:
+  precedence_declaration
+| symbol_declaration
+| "%start" symbol

     {
     {

-      current_class = unknown_sym;
-      current_type = NULL;
+      grammar_start_symbol_set ($2, @2);

     }
     }

-| "%start" symbol
+| "%union" BRACED_CODE

     {
     {

-      grammar_start_symbol_set ($2);
+      typed = 1;
+      MUSCLE_INSERT_INT ("stype_line", @2.first_line);
+      muscle_insert ("stype", $2);

     }
     }

-| "%token" { current_class = token_sym; } symbol_defs.1
+| "%destructor"
+    { current_braced_code = destructor_braced_code; }
+  BRACED_CODE symbols.1
+    {
+      symbol_list_t *list;
+      for (list = $4; list; list = list->next)
+       symbol_destructor_set (list->sym, $3, @3);
+      symbol_list_free ($4);
+      current_braced_code = action_braced_code;
+    }
+| "%printer"
+    { current_braced_code = printer_braced_code; }
+  BRACED_CODE symbols.1
+    {
+      symbol_list_t *list;
+      for (list = $4; list; list = list->next)
+       symbol_printer_set (list->sym, $3, list->location);
+      symbol_list_free ($4);
+      current_braced_code = action_braced_code;
+    }
+;
+
+symbol_declaration:
+  "%nterm" { current_class = nterm_sym; } symbol_defs.1

     {
       current_class = unknown_sym;
       current_type = NULL;
     }
     {
       current_class = unknown_sym;
       current_type = NULL;
     }

-| "%type" TYPE {current_type = $2; } nterms_to_type.1
+| "%token" { current_class = token_sym; } symbol_defs.1

     {
     {

+      current_class = unknown_sym;

       current_type = NULL;
     }
       current_type = NULL;
     }

-| "%union" BRACED_CODE semi_colon_opt
+| "%type" TYPE symbols.1

     {
     {

-      typed = 1;
-      MUSCLE_INSERT_INT ("stype_line", @2.first_line);
-      muscle_insert ("stype", $2);
+      symbol_list_t *list;
+      for (list = $3; list; list = list->next)
+       symbol_type_set (list->sym, $2, @2);
+      symbol_list_free ($3);

     }
 ;
 
     }
 ;
 

-precedence_directives:
-  precedence_directive type.opt
-    { current_assoc = $1; ++current_prec; }
-  terms_to_prec.1
-    { current_assoc = non_assoc; current_type = NULL; }
+precedence_declaration:
+  precedence_declarator type.opt symbols.1
+    {
+      symbol_list_t *list;
+      ++current_prec;
+      for (list = $3; list; list = list->next)
+       {
+         symbol_type_set (list->sym, current_type, @2);
+         symbol_precedence_set (list->sym, current_prec, $1, @1);
+       }
+      symbol_list_free ($3);
+      current_type = NULL;
+    }

 ;
 
 ;
 

-precedence_directive:
+precedence_declarator:

   "%left"     { $$ = left_assoc; }
 | "%right"    { $$ = right_assoc; }
 | "%nonassoc" { $$ = non_assoc; }
 ;
 
 type.opt:
   "%left"     { $$ = left_assoc; }
 | "%right"    { $$ = right_assoc; }
 | "%nonassoc" { $$ = non_assoc; }
 ;
 
 type.opt:

-  /* Nothing. */ { current_type = NULL;}
+  /* Nothing. */ { current_type = NULL; }

 | TYPE           { current_type = $1; }
 ;
 
 /* One or more nonterminals to be %typed. */
 | TYPE           { current_type = $1; }
 ;
 
 /* One or more nonterminals to be %typed. */

-nterms_to_type.1:
-  ID                   { symbol_type_set ($1, current_type); }
-| nterms_to_type.1 ID  { symbol_type_set ($2, current_type); }
-;

 
 

-/* One or more symbols to be given a precedence/associativity.  */
-terms_to_prec.1:
-  symbol
-    {
-      symbol_type_set ($1, current_type);
-      symbol_precedence_set ($1, current_prec, current_assoc);
-    }
-| terms_to_prec.1 symbol
-    {
-      symbol_type_set ($2, current_type);
-      symbol_precedence_set ($2, current_prec, current_assoc);
-    }
+symbols.1:
+  symbol            { $$ = symbol_list_new ($1, @1); }
+| symbols.1 symbol  { $$ = symbol_list_prepend ($1, $2, @2); }

 ;
 
 ;
 

-

 /* One token definition.  */
 symbol_def:
   TYPE
 /* One token definition.  */
 symbol_def:
   TYPE


@@ -240,27 +311,27 @@ symbol_def:
      }
 | ID
      {
      }
 | ID
      {

-       symbol_class_set ($1, current_class);
-       symbol_type_set ($1, current_type);
+       symbol_class_set ($1, current_class, @1);
+       symbol_type_set ($1, current_type, @1);

      }
 | ID INT
     {
      }
 | ID INT
     {

-      symbol_class_set ($1, current_class);
-      symbol_type_set ($1, current_type);
-      symbol_user_token_number_set ($1, $2);
+      symbol_class_set ($1, current_class, @1);
+      symbol_type_set ($1, current_type, @1);
+      symbol_user_token_number_set ($1, $2, @2);

     }
 | ID string_as_id
     {
     }
 | ID string_as_id
     {

-      symbol_class_set ($1, current_class);
-      symbol_type_set ($1, current_type);
-      symbol_make_alias ($1, $2);
+      symbol_class_set ($1, current_class, @1);
+      symbol_type_set ($1, current_type, @1);
+      symbol_make_alias ($1, $2, @$);

     }
 | ID INT string_as_id
     {
     }
 | ID INT string_as_id
     {

-      symbol_class_set ($1, current_class);
-      symbol_type_set ($1, current_type);
-      symbol_user_token_number_set ($1, $2);
-      symbol_make_alias ($1, $3);
+      symbol_class_set ($1, current_class, @1);
+      symbol_type_set ($1, current_type, @1);
+      symbol_user_token_number_set ($1, $2, @2);
+      symbol_make_alias ($1, $3, @$);

     }
 ;
 
     }
 ;
 


@@ -272,36 +343,60 @@ symbol_defs.1:
     {;}
 ;
 
     {;}
 ;
 

-gram:
+
+       /*------------------------------------------.
+       | The grammar section: between the two %%.  |
+       `------------------------------------------*/
+
+grammar:
+  rules_or_grammar_declaration
+| grammar rules_or_grammar_declaration
+;
+
+/* As a Bison extension, one can use the grammar declarations in the
+   body of the grammar.  But to remain LALR(1), they must be ended
+   with a semi-colon.  */
+rules_or_grammar_declaration:

   rules
   rules

-| gram rules
+| grammar_declaration ";"
+    {
+      if (yacc_flag)
+       complain_at (@$, _("POSIX forbids declarations in the grammar"));
+    }
+| error ";"
+    {
+      yyerrok;
+    }

 ;
 
 rules:
 ;
 
 rules:

-  ID ":" { current_lhs = $1; } rhses.1 ";"
+  ID ":" { current_lhs = $1; current_lhs_location = @1; } rhses.1 ";"

     {;}
 ;
 
 rhses.1:
     {;}
 ;
 
 rhses.1:

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

 ;
 
 rhs:
   /* Nothing.  */
 ;
 
 rhs:
   /* Nothing.  */

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

 | rhs symbol
 | rhs symbol

-    { grammar_current_rule_symbol_append ($2); }
+    { grammar_current_rule_symbol_append ($2, @2); }

 | rhs action
 | rhs action

-    { grammar_current_rule_action_append ($2, @2.first_line); }
+    { grammar_current_rule_action_append ($2, @2); }

 | rhs "%prec" symbol
 | rhs "%prec" symbol

-    { grammar_current_rule_prec_set ($3); }
+    { grammar_current_rule_prec_set ($3, @3); }
+| rhs "%dprec" INT
+    { grammar_current_rule_dprec_set ($3, @3); }
+| rhs "%merge" TYPE
+    { grammar_current_rule_merge_set ($3, @3); }

 ;
 
 symbol:
   ID              { $$ = $1; }
 | string_as_id    { $$ = $1; }
 ;
 
 symbol:
   ID              { $$ = $1; }
 | string_as_id    { $$ = $1; }

-| CHARACTER       { $$ = getsym ($1); }

 ;
 
 action:
 ;
 
 action:


@@ -313,8 +408,8 @@ action:
 string_as_id:
   STRING
     {
 string_as_id:
   STRING
     {

-      $$ = getsym ($1);
-      symbol_class_set ($$, token_sym);
+      $$ = symbol_get ($1, @1);
+      symbol_class_set ($$, token_sym, @1);

     }
 ;
 
     }
 ;
 


@@ -327,18 +422,25 @@ string_content:
     };
 
 
     };
 
 

+/* A BRACED_CODE used for its contents.  Strip the braces. */
+code_content:
+  BRACED_CODE
+    {
+      $$ = $1 + 1;
+      $$[strlen ($$) - 1] = '\0';
+    };
+
+

 epilogue.opt:
   /* Nothing.  */
 epilogue.opt:
   /* Nothing.  */

-    {
-      $$ = xstrdup ("");
-    }

 | "%%" EPILOGUE
     {
 | "%%" EPILOGUE
     {

-      $$ = $2;
+      epilogue_augment ($2, @2);
+      scanner_last_string_free ();

     }
 ;
 
     }
 ;
 

-semi_colon_opt:
+semi_colon.opt:

   /* Nothing.  */
 | ";"
 ;
   /* Nothing.  */
 | ";"
 ;


@@ -349,17 +451,11 @@ semi_colon_opt:
 
 static void
 yyprint (FILE *file,
 
 static void
 yyprint (FILE *file,

-         const location_t *loc, int type, const yystype *value)
+         int type, const yystype *value)

 {
 {

-  fputs (" (", file);
-  LOCATION_PRINT (file, *loc);
-  fputs (")", file);
+  fputc (' ', file);

   switch (type)
     {
   switch (type)
     {

-    case CHARACTER:
-      fprintf (file, " = '%s'", value->string);
-      break;
-

     case ID:
       fprintf (file, " = %s", value->symbol->tag);
       break;
     case ID:
       fprintf (file, " = %s", value->symbol->tag);
       break;


@@ -373,7 +469,7 @@ yyprint (FILE *file,
       break;
 
     case TYPE:
       break;
 
     case TYPE:

-      fprintf (file, " = <%s>", value->string);
+      fprintf (file, " = <%s>", value->struniq);

       break;
 
     case BRACED_CODE:
       break;
 
     case BRACED_CODE:


@@ -381,13 +477,15 @@ yyprint (FILE *file,
     case EPILOGUE:
       fprintf (file, " = {{ %s }}", value->string);
       break;
     case EPILOGUE:
       fprintf (file, " = {{ %s }}", value->string);
       break;

+
+    default:
+      fprintf (file, "unknown token type");
+      break;

     }
 }
 
 void
     }
 }
 
 void

-gram_error (gram_control_t *control ATTRIBUTE_UNUSED,
-           location_t *yylloc, const char *msg)
+gram_error (location_t *yylloc, const char *msg)

 {
 {

-  LOCATION_PRINT (stderr, *yylloc);
-  fprintf (stderr, ": %s\n", msg);
+  complain_at (*yylloc, "%s", msg);

 }
 }