src/parse-gram.y

   1 /* Bison Grammar Parser                             -*- C -*-
   2    Copyright (C) 2002 Free Software Foundation, Inc.
   3
   4    This file is part of Bison, the GNU Compiler Compiler.
   5
   6    This program is free software; you can redistribute it and/or modify
   7    it under the terms of the GNU General Public License as published by
   8    the Free Software Foundation; either version 2 of the License, or
   9    (at your option) any later version.
  10
  11    This program is distributed in the hope that it will be useful,
  12    but WITHOUT ANY WARRANTY; without even the implied warranty of
  13    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  14    GNU General Public License for more details.
  15
  16    You should have received a copy of the GNU General Public License
  17    along with this program; if not, write to the Free Software
  18    Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
  19    02111-1307  USA
  20 */
  21
  22
  23 %debug
  24 %defines
  25 %locations
  26 %pure-parser
  27 // %error-verbose
  28 %defines
  29 %name-prefix="gram_"
  30
  31 %{
  32 #include "system.h"
  33 #include "complain.h"
  34 #include "muscle_tab.h"
  35 #include "files.h"
  36 #include "getargs.h"
  37 #include "output.h"
  38 #include "symlist.h"
  39 #include "gram.h"
  40 #include "reader.h"
  41 #include "conflicts.h"
  42
  43 /* Produce verbose parse errors.  */
  44 #define YYERROR_VERBOSE 1
  45 #define YYLLOC_DEFAULT(Current, Rhs, N)                 \
  46 do {                                                    \
  47   if (N)                                                \
  48   {                                                     \
  49     Current.first_column  = Rhs[1].first_column;        \
  50     Current.first_line    = Rhs[1].first_line;          \
  51     Current.last_column   = Rhs[N].last_column;         \
  52     Current.last_line     = Rhs[N].last_line;           \
  53   }                                                     \
  54   else                                                  \
  55   {                                                     \
  56     Current = Rhs[0];                                   \
  57   }                                                     \
  58 } while (0)
  59
  60 /* Pass the control structure to YYPARSE and YYLEX. */
  61 #define YYPARSE_PARAM gram_control
  62 #define YYLEX_PARAM gram_control
  63 /* YYPARSE receives GRAM_CONTROL as a void *.  Provide a
  64    correctly typed access to it.  */
  65 #define yycontrol ((gram_control_t *) gram_control)
  66
  67 /* Request detailed parse error messages, and pass them to GRAM_ERROR.
  68    FIXME: depends on the undocumented availability of YYLLOC.t */
  69 #undef  yyerror
  70 #define yyerror(Msg) \
  71         gram_error (&yylloc, Msg)
  72
  73 #define YYPRINT(File, Type, Value) \
  74         yyprint (File, Type, &Value)
  75 static void yyprint (FILE *file, int type, const yystype *value);
  76
  77 symbol_class current_class = unknown_sym;
  78 char *current_type = 0;
  79 symbol_t *current_lhs;
  80 location_t current_lhs_location;
  81 assoc_t current_assoc;
  82 int current_prec = 0;
  83 braced_code_t current_braced_code = action_braced_code;
  84 %}
  85
  86
  87 /* Only NUMBERS have a value.  */
  88 %union
  89 {
  90   symbol_t *symbol;
  91   symbol_list_t *list;
  92   int integer;
  93   char *string;
  94   assoc_t assoc;
  95 };
  96
  97 /* Define the tokens together with their human representation.  */
  98 %token GRAM_EOF 0 "end of file"
  99 %token STRING     "string"
 100 %token CHARACTER  "character"
 101 %token INT        "integer"
 102
 103 %token PERCENT_TOKEN       "%token"
 104 %token PERCENT_NTERM       "%nterm"
 105
 106 %token PERCENT_TYPE        "%type"
 107 %token PERCENT_DESTRUCTOR  "%destructor"
 108 %token PERCENT_PRINTER     "%printer"
 109
 110 %token PERCENT_UNION       "%union"
 111
 112 %token PERCENT_LEFT        "%left"
 113 %token PERCENT_RIGHT       "%right"
 114 %token PERCENT_NONASSOC    "%nonassoc"
 115
 116 %token PERCENT_PREC          "%prec"
 117 %token PERCENT_DPREC         "%dprec"
 118 %token PERCENT_MERGE         "%merge"
 119
 120
 121 /*----------------------.
 122 | Global Declarations.  |
 123 `----------------------*/
 124
 125 %token
 126   PERCENT_DEBUG         "%debug"
 127   PERCENT_DEFINE        "%define"
 128   PERCENT_DEFINES       "%defines"
 129   PERCENT_ERROR_VERBOSE "%error-verbose"
 130   PERCENT_EXPECT        "%expect"
 131   PERCENT_FILE_PREFIX   "%file-prefix"
 132   PERCENT_GLR_PARSER    "%glr-parser"
 133   PERCENT_LEX_PARAM     "%lex-param"
 134   PERCENT_LOCATIONS     "%locations"
 135   PERCENT_NAME_PREFIX   "%name-prefix"
 136   PERCENT_NO_LINES      "%no-lines"
 137   PERCENT_OUTPUT        "%output"
 138   PERCENT_PARSE_PARAM   "%parse-param"
 139   PERCENT_PURE_PARSER   "%pure-parser"
 140   PERCENT_SKELETON      "%skeleton"
 141   PERCENT_START         "%start"
 142   PERCENT_TOKEN_TABLE   "%token-table"
 143   PERCENT_VERBOSE       "%verbose"
 144   PERCENT_YACC          "%yacc"
 145 ;
 146
 147 %token TYPE            "type"
 148 %token EQUAL           "="
 149 %token SEMICOLON       ";"
 150 %token COLON           ":"
 151 %token COMMA           ","
 152 %token PIPE            "|"
 153 %token ID              "identifier"
 154 %token PERCENT_PERCENT "%%"
 155 %token PROLOGUE        "%{...%}"
 156 %token EPILOGUE        "epilogue"
 157 %token BRACED_CODE     "{...}"
 158
 159
 160 %type <string> CHARACTER TYPE STRING string_content
 161                BRACED_CODE PROLOGUE EPILOGUE epilogue.opt action
 162 %type <integer> INT
 163 %type <symbol> ID symbol string_as_id
 164 %type <assoc> precedence_declarator
 165 %type <list>  symbols.1
 166 %%
 167
 168 input:
 169   declarations "%%" grammar epilogue.opt
 170     {
 171       yycontrol->errcode = 0;
 172       epilogue_set ($4, @4);
 173     }
 174 ;
 175
 176
 177         /*------------------------------------.
 178         | Declarations: before the first %%.  |
 179         `------------------------------------*/
 180
 181 declarations:
 182   /* Nothing */
 183 | declarations declaration semi_colon.opt
 184 ;
 185
 186 declaration:
 187   grammar_declaration
 188 | PROLOGUE                                 { prologue_augment ($1, @1); }
 189 | "%debug"                                 { debug_flag = 1; }
 190 | "%define" string_content string_content  { muscle_insert ($2, $3); }
 191 | "%defines"                               { defines_flag = 1; }
 192 | "%error-verbose"                         { error_verbose = 1; }
 193 | "%expect" INT                            { expected_conflicts = $2; }
 194 | "%file-prefix" "=" string_content        { spec_file_prefix = $3; }
 195 | "%glr-parser"                            { glr_parser = 1; }
 196 | "%lex-param" string_content "," string_content
 197                            { muscle_pair_list_grow ("lex_param", $2, $4); }
 198 | "%locations"                             { locations_flag = 1; }
 199 | "%name-prefix" "=" string_content        { spec_name_prefix = $3; }
 200 | "%no-lines"                              { no_lines_flag = 1; }
 201 | "%output" "=" string_content             { spec_outfile = $3; }
 202 | "%parse-param" string_content "," string_content
 203                            { muscle_pair_list_grow ("parse_param", $2, $4); }
 204 | "%pure-parser"                           { pure_parser = 1; }
 205 | "%skeleton" string_content               { skeleton = $2; }
 206 | "%token-table"                           { token_table_flag = 1; }
 207 | "%verbose"                               { report_flag = 1; }
 208 | "%yacc"                                  { yacc_flag = 1; }
 209 ;
 210
 211 grammar_declaration:
 212   precedence_declaration
 213 | symbol_declaration
 214 | "%start" symbol
 215     {
 216       grammar_start_symbol_set ($2, @2);
 217     }
 218 | "%union" BRACED_CODE
 219     {
 220       typed = 1;
 221       MUSCLE_INSERT_INT ("stype_line", @2.first_line);
 222       muscle_insert ("stype", $2);
 223     }
 224 | "%destructor"
 225     { current_braced_code = destructor_braced_code; }
 226   BRACED_CODE symbols.1
 227     {
 228       symbol_list_t *list;
 229       for (list = $4; list; list = list->next)
 230         symbol_destructor_set (list->sym, $3, @3);
 231       symbol_list_free ($4);
 232       current_braced_code = action_braced_code;
 233     }
 234 | "%printer"
 235     { current_braced_code = printer_braced_code; }
 236   BRACED_CODE symbols.1
 237     {
 238       symbol_list_t *list;
 239       for (list = $4; list; list = list->next)
 240         symbol_printer_set (list->sym, $3, list->location);
 241       symbol_list_free ($4);
 242       current_braced_code = action_braced_code;
 243     }
 244 ;
 245
 246 symbol_declaration:
 247   "%nterm" { current_class = nterm_sym; } symbol_defs.1
 248     {
 249       current_class = unknown_sym;
 250       current_type = NULL;
 251     }
 252 | "%token" { current_class = token_sym; } symbol_defs.1
 253     {
 254       current_class = unknown_sym;
 255       current_type = NULL;
 256     }
 257 | "%type" TYPE symbols.1
 258     {
 259       symbol_list_t *list;
 260       for (list = $3; list; list = list->next)
 261         symbol_type_set (list->sym, $2, @2);
 262       symbol_list_free ($3);
 263     }
 264 ;
 265
 266 precedence_declaration:
 267   precedence_declarator type.opt symbols.1
 268     {
 269       symbol_list_t *list;
 270       ++current_prec;
 271       for (list = $3; list; list = list->next)
 272         {
 273           symbol_type_set (list->sym, current_type, @2);
 274           symbol_precedence_set (list->sym, current_prec, $1, @1);
 275         }
 276       symbol_list_free ($3);
 277       current_type = NULL;
 278     }
 279 ;
 280
 281 precedence_declarator:
 282   "%left"     { $$ = left_assoc; }
 283 | "%right"    { $$ = right_assoc; }
 284 | "%nonassoc" { $$ = non_assoc; }
 285 ;
 286
 287 type.opt:
 288   /* Nothing. */ { current_type = NULL;}
 289 | TYPE           { current_type = $1; }
 290 ;
 291
 292 /* One or more nonterminals to be %typed. */
 293
 294 symbols.1:
 295   symbol            { $$ = symbol_list_new ($1, @1); }
 296 | symbols.1 symbol  { $$ = symbol_list_prepend ($1, $2, @2); }
 297 ;
 298
 299 /* One token definition.  */
 300 symbol_def:
 301   TYPE
 302      {
 303        current_type = $1;
 304      }
 305 | ID
 306      {
 307        symbol_class_set ($1, current_class, @1);
 308        symbol_type_set ($1, current_type, @1);
 309      }
 310 | ID INT
 311     {
 312       symbol_class_set ($1, current_class, @1);
 313       symbol_type_set ($1, current_type, @1);
 314       symbol_user_token_number_set ($1, $2, @2);
 315     }
 316 | ID string_as_id
 317     {
 318       symbol_class_set ($1, current_class, @1);
 319       symbol_type_set ($1, current_type, @1);
 320       symbol_make_alias ($1, $2, @$);
 321     }
 322 | ID INT string_as_id
 323     {
 324       symbol_class_set ($1, current_class, @1);
 325       symbol_type_set ($1, current_type, @1);
 326       symbol_user_token_number_set ($1, $2, @2);
 327       symbol_make_alias ($1, $3, @$);
 328     }
 329 ;
 330
 331 /* One or more symbol definitions. */
 332 symbol_defs.1:
 333   symbol_def
 334     {;}
 335 | symbol_defs.1 symbol_def
 336     {;}
 337 ;
 338
 339
 340         /*------------------------------------------.
 341         | The grammar section: between the two %%.  |
 342         `------------------------------------------*/
 343
 344 grammar:
 345   rules_or_grammar_declaration
 346 | grammar rules_or_grammar_declaration
 347 ;
 348
 349 /* As a Bison extension, one can use the grammar declarations in the
 350    body of the grammar.  But to remain LALR(1), they must be ended
 351    with a semi-colon.  */
 352 rules_or_grammar_declaration:
 353   rules
 354 | grammar_declaration ";"
 355     {
 356       if (yacc_flag)
 357         complain_at (@$, _("POSIX forbids declarations in the grammar"));
 358     }
 359 | error ";"
 360     {
 361       yyerrok;
 362     }
 363 ;
 364
 365 rules:
 366   ID ":" { current_lhs = $1; current_lhs_location = @1; } rhses.1 ";"
 367     {;}
 368 ;
 369
 370 rhses.1:
 371   rhs                { grammar_rule_end (@1); }
 372 | rhses.1 "|" rhs    { grammar_rule_end (@3); }
 373 ;
 374
 375 rhs:
 376   /* Nothing.  */
 377     { grammar_rule_begin (current_lhs, current_lhs_location); }
 378 | rhs symbol
 379     { grammar_current_rule_symbol_append ($2, @2); }
 380 | rhs action
 381     { grammar_current_rule_action_append ($2, @2); }
 382 | rhs "%prec" symbol
 383     { grammar_current_rule_prec_set ($3, @3); }
 384 | rhs "%dprec" INT
 385     { grammar_current_rule_dprec_set ($3, @3); }
 386 | rhs "%merge" TYPE
 387     { grammar_current_rule_merge_set ($3, @3); }
 388 ;
 389
 390 symbol:
 391   ID              { $$ = $1; }
 392 | string_as_id    { $$ = $1; }
 393 | CHARACTER       { $$ = symbol_get ($1, @1); }
 394 ;
 395
 396 action:
 397   BRACED_CODE
 398    { $$ = $1; }
 399 ;
 400
 401 /* A string used as an ID: we have to keep the quotes. */
 402 string_as_id:
 403   STRING
 404     {
 405       $$ = symbol_get ($1, @1);
 406       symbol_class_set ($$, token_sym, @1);
 407     }
 408 ;
 409
 410 /* A string used for its contents.  Strip the quotes. */
 411 string_content:
 412   STRING
 413     {
 414       $$ = $1 + 1;
 415       $$[strlen ($$) - 1] = '\0';
 416     };
 417
 418
 419 epilogue.opt:
 420   /* Nothing.  */
 421     {
 422       $$ = xstrdup ("");
 423     }
 424 | "%%" EPILOGUE
 425     {
 426       $$ = $2;
 427     }
 428 ;
 429
 430 semi_colon.opt:
 431   /* Nothing.  */
 432 | ";"
 433 ;
 434 %%
 435 /*------------------------------------------------------------------.
 436 | When debugging the parser, display tokens' locations and values.  |
 437 `------------------------------------------------------------------*/
 438
 439 static void
 440 yyprint (FILE *file,
 441          int type, const yystype *value)
 442 {
 443   fputc (' ', file);
 444   switch (type)
 445     {
 446     case CHARACTER:
 447       fprintf (file, " = '%s'", value->string);
 448       break;
 449
 450     case ID:
 451       fprintf (file, " = %s", value->symbol->tag);
 452       break;
 453
 454     case INT:
 455       fprintf (file, " = %d", value->integer);
 456       break;
 457
 458     case STRING:
 459       fprintf (file, " = \"%s\"", value->string);
 460       break;
 461
 462     case TYPE:
 463       fprintf (file, " = <%s>", value->string);
 464       break;
 465
 466     case BRACED_CODE:
 467     case PROLOGUE:
 468     case EPILOGUE:
 469       fprintf (file, " = {{ %s }}", value->string);
 470       break;
 471     }
 472 }
 473
 474 void
 475 gram_error (location_t *yylloc, const char *msg)
 476 {
 477   complain_at (*yylloc, "%s", msg);
 478 }