src/gram.h

   1 /* Data definitions for internal representation of bison's input,
   2    Copyright 1984, 1986, 1989, 1992, 2001, 2002
   3    Free Software Foundation, Inc.
   4
   5    This file is part of Bison, the GNU Compiler Compiler.
   6
   7    Bison is free software; you can redistribute it and/or modify
   8    it under the terms of the GNU General Public License as published by
   9    the Free Software Foundation; either version 2, or (at your option)
  10    any later version.
  11
  12    Bison is distributed in the hope that it will be useful,
  13    but WITHOUT ANY WARRANTY; without even the implied warranty of
  14    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  15    GNU General Public License for more details.
  16
  17    You should have received a copy of the GNU General Public License
  18    along with Bison; see the file COPYING.  If not, write to
  19    the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
  20    Boston, MA 02111-1307, USA.  */
  21
  22 #ifndef GRAM_H_
  23 # define GRAM_H_
  24
  25 /* Representation of the grammar rules:
  26
  27    NTOKENS is the number of tokens, and NVARS is the number of
  28    variables (nonterminals).  NSYMS is the total number, ntokens +
  29    nvars.
  30
  31    Each symbol (either token or variable) receives a symbol number.
  32    Numbers 0 to NTOKENS - 1 are for tokens, and NTOKENS to NSYMS - 1
  33    are for variables.  Symbol number zero is the end-of-input token.
  34    This token is counted in ntokens.  The true number of token values
  35    assigned is NTOKENS reduced by one for each alias declaration.
  36
  37    The rules receive rule numbers 1 to NRULES in the order they are
  38    written.  More precisely Bison augments the grammar with the
  39    initial rule, `$axiom: START-SYMBOL EOF', which is numbered 1, all
  40    the user rules are 2, 3 etc.  Each time a rule number is presented
  41    to the user, we subtract 1, so *displayed* rule numbers are 0, 1,
  42    2...
  43
  44    Internally, we cannot use the number 0 for a rule because for
  45    instance RITEM stores both symbol (the RHS) and rule numbers: the
  46    symbols are shorts >= 0, and rule number are stored negative.
  47    Therefore 0 cannot be used, since it would be both the rule number
  48    0, and the token EOF).
  49
  50    Actions and guards are accessed via the rule number.
  51
  52    The rules themselves are described by several arrays: amongst which
  53    RITEM, and RULES.
  54
  55    RULES is an array of struct rule_s, which members are:
  56
  57    RULES[R].lhs -- the symbol number of the left hand side of rule R.
  58    If -1, the rule has been thrown out by reduce.c and should be
  59    ignored.
  60
  61    RULES[R].rhs -- the index in RITEM of the beginning of the portion
  62    for rule R.
  63
  64    RULES[R].prec -- the precedence level of R.
  65
  66    RULES[R].precsym -- the symbol-number of the symbol in %prec for R
  67    (if any).
  68
  69    RULES[R].assoc -- the associativity of R.
  70
  71    RULES[R].line -- the line where R was defined.
  72
  73    RULES[R].useful -- TRUE iff the rule is used.
  74
  75    The right hand side is stored as symbol numbers in a portion of
  76    RITEM.
  77
  78    The length of the portion is one greater than the number of symbols
  79    in the rule's right hand side.  The last element in the portion
  80    contains minus R, which identifies it as the end of a portion and
  81    says which rule it is for.
  82
  83    The portions of RITEM come in order of increasing rule number and
  84    are followed by an element which is zero to mark the end.  nitems
  85    is the total length of ritem, not counting the final zero.  Each
  86    element of RITEM is called an "item" and its index in RITEM is an
  87    item number.
  88
  89    Item numbers are used in the finite state machine to represent
  90    places that parsing can get to.
  91
  92    SYMBOLS[I]->prec records the precedence level of each symbol.
  93
  94    Precedence levels are assigned in increasing order starting with 1
  95    so that numerically higher precedence values mean tighter binding
  96    as they ought to.  Zero as a symbol or rule's precedence means none
  97    is assigned.
  98
  99    Associativities are recorded similarly in SYMBOLS[I]->assoc.  */
 100
 101
 102 #define ISTOKEN(s)      ((s) < ntokens)
 103 #define ISVAR(s)        ((s) >= ntokens)
 104
 105 extern int nitems;
 106 extern int nrules;
 107 extern int nsyms;
 108 extern int ntokens;
 109 extern int nvars;
 110
 111 extern short *ritem;
 112 extern int nritems;
 113
 114 extern int start_symbol;
 115
 116 /* Associativity values for tokens and rules.  */
 117 typedef enum
 118 {
 119   right_assoc,
 120   left_assoc,
 121   non_assoc
 122 } associativity;
 123
 124
 125 typedef struct rule_s
 126 {
 127   short lhs;
 128   short *rhs;
 129   short prec;
 130   short precsym;
 131   associativity assoc;
 132   short line;
 133   bool useful;
 134
 135   const char *action;
 136   short action_line;
 137
 138   const char *guard;
 139   short guard_line;
 140 } rule_t;
 141
 142 extern struct rule_s *rules;
 143
 144 /* Table of the symbols, indexed by the symbol number. */
 145 extern struct bucket **symbols;
 146
 147 /* token translation table: indexed by a token number as returned by
 148    the user's yylex routine, it yields the internal token number used
 149    by the parser and throughout bison.  */
 150
 151 extern short *token_translations;
 152 extern int max_user_token_number;
 153
 154 /* SEMANTIC_PARSER is nonzero if the input file says to use the hairy
 155    parser that provides for semantic error recovery.  If it is zero,
 156    the yacc-compatible simplified parser is used.  */
 157
 158 extern int semantic_parser;
 159
 160 /* PURE_PARSER is nonzero if should generate a parser that is all pure
 161    and reentrant.  */
 162
 163 extern int pure_parser;
 164
 165 /* ERROR_TOKEN_NUMBER is the token number of the error token.  */
 166
 167 extern int error_token_number;
 168
 169
 170 /* Dump RITEM for traces. */
 171 void ritem_print PARAMS ((FILE *out));
 172
 173 /* Return the size of the longest rule RHS.  */
 174 size_t ritem_longest_rhs PARAMS ((void));
 175
 176 #endif /* !GRAM_H_ */