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