X-Git-Url: https://git.saurik.com/bison.git/blobdiff_plain/84866159725a9df9febc1050643c0355eba5916c..95021767a1e93fa7b83f8a6c5ec5fd5f8064dd2c:/src/reader.c diff --git a/src/reader.c b/src/reader.c index 573cb14d..587a4f09 100644 --- a/src/reader.c +++ b/src/reader.c @@ -1,7 +1,7 @@ /* Input parser for Bison Copyright (C) 1984, 1986, 1989, 1992, 1998, 2000, 2001, 2002, 2003, - 2005, 2006 Free Software Foundation, Inc. + 2005, 2006, 2007 Free Software Foundation, Inc. This file is part of Bison, the GNU Compiler Compiler. @@ -31,17 +31,23 @@ #include "getargs.h" #include "gram.h" #include "muscle_tab.h" -#include "output.h" #include "reader.h" #include "symlist.h" #include "symtab.h" +#include "scan-gram.h" +#include "scan-code.h" + +static void check_and_convert_grammar (void); static symbol_list *grammar = NULL; static bool start_flag = false; merger_list *merge_functions; /* Was %union seen? */ -bool typed = false; +bool union_seen = false; + +/* Was a tag seen? */ +bool tag_seen = false; /* Should rules have a default precedence? */ bool default_prec = true; @@ -64,18 +70,20 @@ grammar_start_symbol_set (symbol *sym, location loc) } -/*----------------------------------------------------------------. -| There are two prologues: one before %union, one after. Augment | -| the current one. | -`----------------------------------------------------------------*/ +/*---------------------------------------------------------------------. +| There are two prologues: one before the first %union and one after. | +| Augment the one specified by POST. | +`---------------------------------------------------------------------*/ void -prologue_augment (const char *prologue, location loc) +prologue_augment (const char *prologue, location loc, bool post) { struct obstack *oout = - !typed ? &pre_prologue_obstack : &post_prologue_obstack; + !post ? &pre_prologue_obstack : &post_prologue_obstack; obstack_fgrow1 (oout, "]b4_syncline(%d, [[", loc.start.line); + /* FIXME: Protection of M4 characters missing here. See + output.c:escaped_output. */ MUSCLE_OBSTACK_SGROW (oout, quotearg_style (c_quoting_style, loc.start.file)); obstack_sgrow (oout, "]])[\n"); @@ -84,14 +92,13 @@ prologue_augment (const char *prologue, location loc) -/*-------------------------------------------------------------------. -| Return the merger index for a merging function named NAME, whose | -| arguments have type TYPE. Records the function, if new, in | -| MERGER_LIST. | -`-------------------------------------------------------------------*/ +/*------------------------------------------------------------------------. +| Return the merger index for a merging function named NAME. Records the | +| function, if new, in MERGER_LIST. | +`------------------------------------------------------------------------*/ static int -get_merge_function (uniqstr name, uniqstr type, location loc) +get_merge_function (uniqstr name) { merger_list *syms; merger_list head; @@ -100,9 +107,6 @@ get_merge_function (uniqstr name, uniqstr type, location loc) if (! glr_parser) return 0; - if (type == NULL) - type = uniqstr_new (""); - head.next = merge_functions; for (syms = &head, n = 1; syms->next; syms = syms->next, n += 1) if (UNIQSTR_EQ (name, syms->next->name)) @@ -111,16 +115,51 @@ get_merge_function (uniqstr name, uniqstr type, location loc) { syms->next = xmalloc (sizeof syms->next[0]); syms->next->name = uniqstr_new (name); - syms->next->type = uniqstr_new (type); + /* After all symbol type declarations have been parsed, packgram invokes + record_merge_function_type to set the type. */ + syms->next->type = NULL; syms->next->next = NULL; merge_functions = head.next; } - else if (!UNIQSTR_EQ (type, syms->next->type)) - warn_at (loc, _("result type clash on merge function %s: <%s> != <%s>"), - name, type, syms->next->type); return n; } +/*-------------------------------------------------------------------------. +| For the existing merging function with index MERGER, record the result | +| type as TYPE as required by the lhs of the rule whose %merge declaration | +| is at DECLARATION_LOC. | +`-------------------------------------------------------------------------*/ + +static void +record_merge_function_type (int merger, uniqstr type, location declaration_loc) +{ + int merger_find; + merger_list *merge_function; + + if (merger <= 0) + return; + + if (type == NULL) + type = uniqstr_new (""); + + merger_find = 1; + for (merge_function = merge_functions; + merge_function != NULL && merger_find != merger; + merge_function = merge_function->next) + merger_find += 1; + aver (merge_function != NULL && merger_find == merger); + if (merge_function->type != NULL && !UNIQSTR_EQ (merge_function->type, type)) + { + complain_at (declaration_loc, + _("result type clash on merge function `%s': <%s> != <%s>"), + merge_function->name, type, merge_function->type); + complain_at (merge_function->type_declaration_location, + _("previous declaration")); + } + merge_function->type = uniqstr_new (type); + merge_function->type_declaration_location = declaration_loc; +} + /*--------------------------------------. | Free all merge-function definitions. | `--------------------------------------*/ @@ -156,7 +195,7 @@ static symbol_list *grammar_end = NULL; static void grammar_symbol_append (symbol *sym, location loc) { - symbol_list *p = symbol_list_new (sym, loc); + symbol_list *p = symbol_list_sym_new (sym, loc); if (grammar_end) grammar_end->next = p; @@ -174,7 +213,7 @@ grammar_symbol_append (symbol *sym, location loc) /* The rule currently being defined, and the previous rule. CURRENT_RULE points to the first LHS of the current rule, while PREVIOUS_RULE_END points to the *end* of the previous rule (NULL). */ -symbol_list *current_rule = NULL; +static symbol_list *current_rule = NULL; static symbol_list *previous_rule_end = NULL; @@ -185,13 +224,6 @@ static symbol_list *previous_rule_end = NULL; void grammar_current_rule_begin (symbol *lhs, location loc) { - if (!start_flag) - { - startsymbol = lhs; - startsymbol_location = loc; - start_flag = true; - } - /* Start a new rule and record its lhs. */ ++nrules; previous_rule_end = grammar_end; @@ -210,18 +242,26 @@ grammar_current_rule_begin (symbol *lhs, location loc) } -/*-----------------------------------------------------------------. -| A symbol is typed if it has a declared %type, or if it is a | -| mid-rule symbol (i.e., the generated LHS replacing a mid-rule | -| action) that was assigned to, as in `exp: { $$ = 1; } { $$ = $1; | -| }'. | -`-----------------------------------------------------------------*/ +/*----------------------------------------------------------------------. +| A symbol should be used if either: | +| 1. It has a destructor. | +| 2. --warnings=midrule-values and the symbol is a mid-rule symbol | +| (i.e., the generated LHS replacing a mid-rule action) that was | +| assigned to or used, as in "exp: { $$ = 1; } { $$ = $1; }". | +`----------------------------------------------------------------------*/ static bool -symbol_typed_p (const symbol_list *s) +symbol_should_be_used (symbol_list const *s) { - return (s->sym->type_name - || s->mid_rule && s->mid_rule->used); + if (symbol_destructor_get (s->content.sym)->code) + return true; + if (warnings_flag & warnings_midrule_values) + return ((s->midrule && s->midrule->action_props.is_value_used) + || (s->midrule_parent_rule + && symbol_list_n_get (s->midrule_parent_rule, + s->midrule_parent_rhs_index) + ->action_props.is_value_used)); + return false; } /*----------------------------------------------------------------. @@ -239,13 +279,13 @@ grammar_rule_check (const symbol_list *r) Don't worry about the default action if $$ is untyped, since $$'s value can't be used. */ - if (!r->action && r->sym->type_name) + if (!r->action_props.code && r->content.sym->type_name) { - symbol *first_rhs = r->next->sym; + symbol *first_rhs = r->next->content.sym; /* If $$ is being set in default way, report if any type mismatch. */ if (first_rhs) { - char const *lhs_type = r->sym->type_name; + char const *lhs_type = r->content.sym->type_name; const char *rhs_type = first_rhs->type_name ? first_rhs->type_name : ""; if (!UNIQSTR_EQ (lhs_type, rhs_type)) @@ -259,15 +299,15 @@ grammar_rule_check (const symbol_list *r) _("empty rule for typed nonterminal, and no action")); } - /* Check that typed symbol values are used. */ + /* Check that symbol values that should be used are in fact used. */ { symbol_list const *l = r; int n = 0; - for (; l && l->sym; l = l->next, ++n) - if (! (l->used - || !symbol_typed_p (l) + for (; l && l->content.sym; l = l->next, ++n) + if (! (l->action_props.is_value_used + || !symbol_should_be_used (l) /* The default action, $$ = $1, `uses' both. */ - || (!r->action && (n == 0 || n == 1)))) + || (!r->action_props.code && (n == 0 || n == 1)))) { if (n) warn_at (r->location, _("unused value: $%d"), n); @@ -288,7 +328,6 @@ grammar_current_rule_end (location loc) /* Put an empty link in the list to mark the end of this rule */ grammar_symbol_append (NULL, grammar_end->location); current_rule->location = loc; - grammar_rule_check (current_rule); } @@ -299,7 +338,7 @@ grammar_current_rule_end (location loc) | rule. | `-------------------------------------------------------------------*/ -static void +void grammar_midrule_action (void) { /* Since the action was written out with this rule's number, we must @@ -308,9 +347,9 @@ grammar_midrule_action (void) /* Make a DUMMY nonterminal, whose location is that of the midrule action. Create the MIDRULE. */ - location dummy_location = current_rule->action_location; + location dummy_location = current_rule->action_props.location; symbol *dummy = dummy_symbol_get (dummy_location); - symbol_list *midrule = symbol_list_new (dummy, dummy_location); + symbol_list *midrule = symbol_list_sym_new (dummy, dummy_location); /* Make a new rule, whose body is empty, before the current one, so that the action just read can belong to it. */ @@ -318,13 +357,11 @@ grammar_midrule_action (void) ++nritems; /* Attach its location and actions to that of the DUMMY. */ midrule->location = dummy_location; - midrule->action = current_rule->action; - midrule->action_location = dummy_location; - current_rule->action = NULL; - /* If $$ was used in the action, the LHS of the enclosing rule was - incorrectly flagged as used. */ - midrule->used = current_rule->used; - current_rule->used = false; + code_props_rule_action_init (&midrule->action_props, + current_rule->action_props.code, + current_rule->action_props.location, + midrule); + code_props_none_init (¤t_rule->action_props); if (previous_rule_end) previous_rule_end->next = midrule; @@ -332,8 +369,7 @@ grammar_midrule_action (void) grammar = midrule; /* End the dummy's rule. */ - midrule->next = symbol_list_new (NULL, dummy_location); - grammar_rule_check (midrule); + midrule->next = symbol_list_sym_new (NULL, dummy_location); midrule->next->next = current_rule; previous_rule_end = midrule->next; @@ -341,7 +377,9 @@ grammar_midrule_action (void) /* Insert the dummy nonterminal replacing the midrule action into the current rule. Bind it to its dedicated rule. */ grammar_current_rule_symbol_append (dummy, dummy_location); - grammar_end->mid_rule = midrule; + grammar_end->midrule = midrule; + midrule->midrule_parent_rule = current_rule; + midrule->midrule_parent_rhs_index = symbol_list_length (current_rule->next); } /* Set the precedence symbol of the current rule to PRECSYM. */ @@ -349,6 +387,7 @@ grammar_midrule_action (void) void grammar_current_rule_prec_set (symbol *precsym, location loc) { + symbol_class_set (precsym, token_sym, loc, false); if (current_rule->ruleprec) complain_at (loc, _("only one %s allowed per rule"), "%prec"); current_rule->ruleprec = precsym; @@ -378,8 +417,8 @@ grammar_current_rule_merge_set (uniqstr name, location loc) warn_at (loc, _("%s affects only GLR parsers"), "%merge"); if (current_rule->merger != 0) complain_at (loc, _("only one %s allowed per rule"), "%merge"); - current_rule->merger = - get_merge_function (name, current_rule->sym->type_name, loc); + current_rule->merger = get_merge_function (name); + current_rule->merger_declaration_location = loc; } /* Attach SYM to the current rule. If needed, move the previous @@ -388,21 +427,22 @@ grammar_current_rule_merge_set (uniqstr name, location loc) void grammar_current_rule_symbol_append (symbol *sym, location loc) { - if (current_rule->action) + if (current_rule->action_props.code) grammar_midrule_action (); grammar_symbol_append (sym, loc); } -/* Attach an ACTION to the current rule. If needed, move the previous - action as a mid-rule action. */ +/* Attach an ACTION to the current rule. */ void grammar_current_rule_action_append (const char *action, location loc) { - if (current_rule->action) + if (current_rule->action_props.code) grammar_midrule_action (); - current_rule->action = action; - current_rule->action_location = loc; + /* After all symbol declarations have been parsed, packgram invokes + code_props_translate_code. */ + code_props_rule_action_init (¤t_rule->action_props, action, loc, + current_rule); } @@ -418,15 +458,36 @@ packgram (void) rule_number ruleno = 0; symbol_list *p = grammar; - ritem = xnmalloc (nritems, sizeof *ritem); + ritem = xnmalloc (nritems + 1, sizeof *ritem); + + /* This sentinel is used by build_relations in gram.c. */ + *ritem++ = 0; + rules = xnmalloc (nrules, sizeof *rules); + /* Before invoking grammar_rule_check on any rule, make sure all actions have + already been scanned in order to set `used' flags. Otherwise, checking + that a midrule's $$ should be set will not always work properly because + the check must forward-reference the midrule's parent rule. For the same + reason, all the `used' flags must be set before checking whether to remove + `$' from any midrule symbol name. */ + while (p) + { + code_props_translate_code (&p->action_props); + if (p) + p = p->next; + } + + p = grammar; while (p) { + int rule_length = 0; symbol *ruleprec = p->ruleprec; + record_merge_function_type (p->merger, p->content.sym->type_name, + p->merger_declaration_location); rules[ruleno].user_number = ruleno; rules[ruleno].number = ruleno; - rules[ruleno].lhs = p->sym; + rules[ruleno].lhs = p->content.sym; rules[ruleno].rhs = ritem + itemno; rules[ruleno].prec = NULL; rules[ruleno].dprec = p->dprec; @@ -434,21 +495,42 @@ packgram (void) rules[ruleno].precsym = NULL; rules[ruleno].location = p->location; rules[ruleno].useful = true; - rules[ruleno].action = p->action; - rules[ruleno].action_location = p->action_location; - - p = p->next; - while (p && p->sym) + rules[ruleno].action = p->action_props.code; + rules[ruleno].action_location = p->action_props.location; + + /* If the midrule's $$ is set or its $n is used, remove the `$' from the + symbol name so that it's a user-defined symbol so that the default + %destructor and %printer apply. */ + if (p->midrule_parent_rule + && (p->action_props.is_value_used + || symbol_list_n_get (p->midrule_parent_rule, + p->midrule_parent_rhs_index) + ->action_props.is_value_used)) + p->content.sym->tag += 1; + + /* Don't check the generated rule 0. It has no action, so some rhs + symbols may appear unused, but the parsing algorithm ensures that + %destructor's are invoked appropriately. */ + if (p != grammar) + grammar_rule_check (p); + + for (p = p->next; p && p->content.sym; p = p->next) { + ++rule_length; + + /* Don't allow rule_length == INT_MAX, since that might + cause confusion with strtol if INT_MAX == LONG_MAX. */ + if (rule_length == INT_MAX) + fatal_at (rules[ruleno].location, _("rule is too long")); + /* item_number = symbol_number. But the former needs to contain more: negative rule numbers. */ - ritem[itemno++] = symbol_number_as_item_number (p->sym->number); + ritem[itemno++] = + symbol_number_as_item_number (p->content.sym->number); /* A rule gets by default the precedence and associativity - of the last token in it. */ - if (p->sym->class == token_sym && default_prec) - rules[ruleno].prec = p->sym; - if (p) - p = p->next; + of its last token. */ + if (p->content.sym->class == token_sym && default_prec) + rules[ruleno].prec = p->content.sym; } /* If this rule has a %prec, @@ -458,15 +540,17 @@ packgram (void) rules[ruleno].precsym = ruleprec; rules[ruleno].prec = ruleprec; } + /* An item ends by the rule number (negated). */ ritem[itemno++] = rule_number_as_item_number (ruleno); + aver (itemno < ITEM_NUMBER_MAX); ++ruleno; + aver (ruleno < RULE_NUMBER_MAX); if (p) p = p->next; } - if (itemno != nritems) - abort (); + aver (itemno == nritems); if (trace_flag & trace_sets) ritem_print (stderr); @@ -509,15 +593,25 @@ reader (void) gram__flex_debug = trace_flag & trace_scan; gram_debug = trace_flag & trace_parse; - scanner_initialize (); + gram_scanner_initialize (); gram_parse (); - /* If something went wrong during the parsing, don't try to - continue. */ - if (complaint_issued) - return; + if (! complaint_issued) + check_and_convert_grammar (); + + xfclose (gram_in); +} - /* Grammar has been read. Do some checking */ + +/*-------------------------------------------------------------. +| Check the grammar that has just been read, and convert it to | +| internal form. | +`-------------------------------------------------------------*/ + +static void +check_and_convert_grammar (void) +{ + /* Grammar has been read. Do some checking. */ if (nrules == 0) fatal (_("no rules in the input grammar")); @@ -534,26 +628,41 @@ reader (void) endtoken->user_token_number = 0; } - /* Insert the initial rule, which line is that of the first rule + /* Find the start symbol if no %start. */ + if (!start_flag) + { + symbol_list *node; + for (node = grammar; + node != NULL && symbol_is_dummy (node->content.sym); + node = node->next) + { + for (node = node->next; + node != NULL && node->content.sym != NULL; + node = node->next) + ; + } + aver (node != NULL); + grammar_start_symbol_set (node->content.sym, + node->content.sym->location); + } + + /* Insert the initial rule, whose line is that of the first rule (not that of the start symbol): accept: %start EOF. */ { - symbol_list *p = symbol_list_new (accept, empty_location); + symbol_list *p = symbol_list_sym_new (accept, empty_location); p->location = grammar->location; - p->next = symbol_list_new (startsymbol, empty_location); - p->next->next = symbol_list_new (endtoken, empty_location); - p->next->next->next = symbol_list_new (NULL, empty_location); + p->next = symbol_list_sym_new (startsymbol, empty_location); + p->next->next = symbol_list_sym_new (endtoken, empty_location); + p->next->next->next = symbol_list_sym_new (NULL, empty_location); p->next->next->next->next = grammar; nrules += 1; nritems += 3; grammar = p; } - if (! (nsyms <= SYMBOL_NUMBER_MAXIMUM && nsyms == ntokens + nvars)) - abort (); - - xfclose (gram_in); + aver (nsyms <= SYMBOL_NUMBER_MAXIMUM && nsyms == ntokens + nvars); /* Assign the symbols their symbol numbers. Write #defines for the token symbols into FDEFINES if requested. */ @@ -563,5 +672,5 @@ reader (void) packgram (); /* The grammar as a symbol_list is no longer needed. */ - LIST_FREE (symbol_list, grammar); + symbol_list_free (grammar); }