X-Git-Url: https://git.saurik.com/bison.git/blobdiff_plain/1c7b7e1d87f698efb2e3daad0f6484829563d96f..775673717fc9b7e5174418f7420ba21040a8e675:/src/parse-gram.y?ds=sidebyside diff --git a/src/parse-gram.y b/src/parse-gram.y index a8a1335b..532349bb 100644 --- a/src/parse-gram.y +++ b/src/parse-gram.y @@ -1,13 +1,12 @@ %{/* Bison Grammar Parser -*- C -*- - Copyright (C) 2002, 2003, 2004, 2005, 2006, 2007 Free Software Foundation, - Inc. + Copyright (C) 2002-2012 Free Software Foundation, Inc. This file is part of Bison, the GNU Compiler Compiler. - This program is free software; you can redistribute it and/or modify + This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2 of the License, or + the Free Software Foundation, either version 3 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, @@ -16,22 +15,18 @@ GNU General Public License for more details. You should have received a copy of the GNU General Public License - along with this program; if not, write to the Free Software - Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA - 02110-1301 USA -*/ + along with this program. If not, see . */ #include #include "system.h" -#include - #include "complain.h" #include "conflicts.h" #include "files.h" #include "getargs.h" #include "gram.h" -#include "muscle_tab.h" +#include "muscle-tab.h" +#include "named-ref.h" #include "quotearg.h" #include "reader.h" #include "symlist.h" @@ -42,7 +37,7 @@ static YYLTYPE lloc_default (YYLTYPE const *, int); #define YY_LOCATION_PRINT(File, Loc) \ - location_print (File, Loc) + location_print (File, Loc) static void version_check (location const *loc, char const *version); @@ -50,39 +45,52 @@ static void version_check (location const *loc, char const *version); FIXME: depends on the undocumented availability of YYLLOC. */ #undef yyerror #define yyerror(Msg) \ - gram_error (&yylloc, Msg) + gram_error (&yylloc, Msg) static void gram_error (location const *, char const *); static char const *char_name (char); - -/** Add a lex-param or a parse-param. - * - * \param type \a lex_param or \a parse_param - * \param decl the formal argument - * \param loc the location in the source. - */ -static void add_param (char const *type, char *decl, location loc); - - -static symbol_class current_class = unknown_sym; -static uniqstr current_type = NULL; -static symbol *current_lhs; -static location current_lhs_location; -static int current_prec = 0; - -#define YYTYPE_INT16 int_fast16_t -#define YYTYPE_INT8 int_fast8_t -#define YYTYPE_UINT16 uint_fast16_t -#define YYTYPE_UINT8 uint_fast8_t %} +%code +{ + static int current_prec = 0; + static location current_lhs_location; + static named_ref *current_lhs_named_ref; + static symbol *current_lhs_symbol; + static symbol_class current_class = unknown_sym; + static uniqstr current_type = NULL; + + /** Set the new current left-hand side symbol, possibly common + * to several right-hand side parts of rule. + */ + static + void + current_lhs(symbol *sym, location loc, named_ref *ref) + { + current_lhs_symbol = sym; + current_lhs_location = loc; + /* In order to simplify memory management, named references for lhs + are always assigned by deep copy into the current symbol_list + node. This is because a single named-ref in the grammar may + result in several uses when the user factors lhs between several + rules using "|". Therefore free the parser's original copy. */ + free (current_lhs_named_ref); + current_lhs_named_ref = ref; + } + + #define YYTYPE_INT16 int_fast16_t + #define YYTYPE_INT8 int_fast8_t + #define YYTYPE_UINT16 uint_fast16_t + #define YYTYPE_UINT8 uint_fast8_t +} + %debug %verbose %defines %locations %pure-parser -%error-verbose -%defines +%define parse.error "verbose" +%define parse.lac full %name-prefix="gram_" %expect 0 @@ -96,12 +104,13 @@ static int current_prec = 0; %union { + assoc assoc; + char *code; + char const *chars; + int integer; + named_ref *named_ref; symbol *symbol; symbol_list *list; - int integer; - char const *chars; - char *code; - assoc assoc; uniqstr uniqstr; unsigned char character; }; @@ -121,43 +130,36 @@ static int current_prec = 0; %token PERCENT_LEFT "%left" %token PERCENT_RIGHT "%right" %token PERCENT_NONASSOC "%nonassoc" +%token PERCENT_PRECEDENCE "%precedence" %token PERCENT_PREC "%prec" %token PERCENT_DPREC "%dprec" %token PERCENT_MERGE "%merge" - /*----------------------. | Global Declarations. | `----------------------*/ %token PERCENT_CODE "%code" - PERCENT_DEBUG "%debug" PERCENT_DEFAULT_PREC "%default-prec" PERCENT_DEFINE "%define" PERCENT_DEFINES "%defines" PERCENT_ERROR_VERBOSE "%error-verbose" PERCENT_EXPECT "%expect" - PERCENT_EXPECT_RR "%expect-rr" + PERCENT_EXPECT_RR "%expect-rr" + PERCENT_FLAG "%" PERCENT_FILE_PREFIX "%file-prefix" PERCENT_GLR_PARSER "%glr-parser" PERCENT_INITIAL_ACTION "%initial-action" PERCENT_LANGUAGE "%language" - PERCENT_LEX_PARAM "%lex-param" - PERCENT_LOCATIONS "%locations" PERCENT_NAME_PREFIX "%name-prefix" PERCENT_NO_DEFAULT_PREC "%no-default-prec" PERCENT_NO_LINES "%no-lines" PERCENT_NONDETERMINISTIC_PARSER - "%nondeterministic-parser" + "%nondeterministic-parser" PERCENT_OUTPUT "%output" - PERCENT_PARSE_PARAM "%parse-param" - PERCENT_PURE_PARSER "%pure-parser" - PERCENT_PUSH_PARSER "%push-parser" - PERCENT_PUSH_PULL_PARSER - "%push-pull-parser" - PERCENT_REQUIRE "%require" + PERCENT_REQUIRE "%require" PERCENT_SKELETON "%skeleton" PERCENT_START "%start" PERCENT_TOKEN_TABLE "%token-table" @@ -166,6 +168,8 @@ static int current_prec = 0; ; %token BRACED_CODE "{...}" +%token BRACED_PREDICATE "%?{...}" +%token BRACKETED_ID "[identifier]" %token CHAR "char" %token EPILOGUE "epilogue" %token EQUAL "=" @@ -175,36 +179,90 @@ static int current_prec = 0; %token PIPE "|" %token PROLOGUE "%{...%}" %token SEMICOLON ";" -%token TYPE "type" -%token TYPE_TAG_ANY "<*>" -%token TYPE_TAG_NONE "<>" +%token TAG "" +%token TAG_ANY "<*>" +%token TAG_NONE "<>" %type CHAR %printer { fputs (char_name ($$), stderr); } CHAR /* braceless is not to be used for rule or symbol actions, as it calls code_props_plain_init. */ -%type STRING "%{...%}" EPILOGUE braceless content content.opt -%type "{...}" +%type STRING "%{...%}" EPILOGUE braceless content.opt +%type "{...}" "%?{...}" %printer { fputs (quotearg_style (c_quoting_style, $$), stderr); } - STRING + STRING %printer { fprintf (stderr, "{\n%s\n}", $$); } - braceless content content.opt "{...}" "%{...%}" EPILOGUE + braceless content.opt "{...}" "%{...%}" EPILOGUE -%type TYPE ID ID_COLON -%printer { fprintf (stderr, "<%s>", $$); } TYPE -%printer { fputs ($$, stderr); } ID +%type BRACKETED_ID ID ID_COLON PERCENT_FLAG TAG variable +%printer { fputs ($$, stderr); } +%printer { fprintf (stderr, "[%s]", $$); } BRACKETED_ID %printer { fprintf (stderr, "%s:", $$); } ID_COLON +%printer { fprintf (stderr, "%%%s", $$); } PERCENT_FLAG +%printer { fprintf (stderr, "<%s>", $$); } TAG %type INT -%printer { fprintf (stderr, "%d", $$); } INT +%printer { fprintf (stderr, "%d", $$); } -%type id id_colon symbol string_as_id -%printer { fprintf (stderr, "%s", $$->tag); } id symbol string_as_id +%type id id_colon string_as_id symbol symbol.prec +%printer { fprintf (stderr, "%s", $$->tag); } %printer { fprintf (stderr, "%s:", $$->tag); } id_colon %type precedence_declarator -%type symbols.1 generic_symlist generic_symlist_item +%type symbols.1 symbols.prec generic_symlist generic_symlist_item +%type named_ref.opt + +/*---------. +| %param. | +`---------*/ +%code requires +{ +# ifndef PARAM_TYPE +# define PARAM_TYPE + typedef enum + { + param_none = 0, + param_lex = 1 << 0, + param_parse = 1 << 1, + param_both = param_lex | param_parse + } param_type; +# endif +}; +%code +{ + /** Add a lex-param and/or a parse-param. + * + * \param type where to push this formal argument. + * \param decl the formal argument. Destroyed. + * \param loc the location in the source. + */ + static void add_param (param_type type, char *decl, location loc); + static param_type current_param = param_none; +}; +%union +{ + param_type param; +} +%token PERCENT_PARAM "%param"; +%printer +{ + switch ($$) + { +#define CASE(In, Out) \ + case param_ ## In: fputs ("%" #Out, stderr); break + CASE(lex, lex-param); + CASE(parse, parse-param); + CASE(both, param); +#undef CASE + case param_none: aver (false); break; + } +} ; + + + /*==========\ + | Grammar. | + \==========*/ %% input: @@ -212,9 +270,9 @@ input: ; - /*------------------------------------. - | Declarations: before the first %%. | - `------------------------------------*/ + /*------------------------------------. + | Declarations: before the first %%. | + `------------------------------------*/ prologue_declarations: /* Nothing */ @@ -229,20 +287,32 @@ prologue_declaration: code_props_plain_init (&plain_code, $1, @1); code_props_translate_code (&plain_code); gram_scanner_last_string_free (); - prologue_augment (plain_code.code, @1, union_seen); + muscle_code_grow (union_seen ? "post_prologue" : "pre_prologue", + plain_code.code, @1); code_scanner_last_string_free (); } -| "%debug" { debug_flag = true; } -| "%define" STRING content.opt { muscle_insert ($2, $3); } +| "%" + { + muscle_percent_define_ensure ($1, @1, true); + } +| "%define" variable content.opt + { + muscle_percent_define_insert ($2, @2, $3, + MUSCLE_PERCENT_DEFINE_GRAMMAR_FILE); + } | "%defines" { defines_flag = true; } | "%defines" STRING { defines_flag = true; spec_defines_file = xstrdup ($2); } -| "%error-verbose" { error_verbose = true; } +| "%error-verbose" + { + muscle_percent_define_insert ("parse.error", @1, "verbose", + MUSCLE_PERCENT_DEFINE_GRAMMAR_FILE); + } | "%expect" INT { expected_sr_conflicts = $2; } -| "%expect-rr" INT { expected_rr_conflicts = $2; } +| "%expect-rr" INT { expected_rr_conflicts = $2; } | "%file-prefix" STRING { spec_file_prefix = $2; } | "%file-prefix" "=" STRING { spec_file_prefix = $3; } /* deprecated */ | "%glr-parser" @@ -259,27 +329,55 @@ prologue_declaration: muscle_code_grow ("initial_action", action.code, @2); code_scanner_last_string_free (); } -| "%language" STRING { language_argmatch ($2, 1, &@1); } -| "%lex-param" "{...}" { add_param ("lex_param", $2, @2); } -| "%locations" { locations_flag = true; } +| "%language" STRING { language_argmatch ($2, grammar_prio, @1); } | "%name-prefix" STRING { spec_name_prefix = $2; } | "%name-prefix" "=" STRING { spec_name_prefix = $3; } /* deprecated */ | "%no-lines" { no_lines_flag = true; } -| "%nondeterministic-parser" { nondeterministic_parser = true; } +| "%nondeterministic-parser" { nondeterministic_parser = true; } | "%output" STRING { spec_outfile = $2; } | "%output" "=" STRING { spec_outfile = $3; } /* deprecated */ -| "%parse-param" "{...}" { add_param ("parse_param", $2, @2); } -| "%pure-parser" { pure_parser = true; } -| "%push-parser" { push_parser = true; pull_parser = false; } -| "%push-pull-parser" { push_parser = true; pull_parser = true; } +| "%param" { current_param = $1; } params { current_param = param_none; } | "%require" STRING { version_check (&@2, $2); } -| "%skeleton" STRING { skeleton_arg ($2, 1, &@1); } +| "%skeleton" STRING + { + char const *skeleton_user = $2; + if (mbschr (skeleton_user, '/')) + { + size_t dir_length = strlen (current_file); + char *skeleton_build; + while (dir_length && current_file[dir_length - 1] != '/') + --dir_length; + while (dir_length && current_file[dir_length - 1] == '/') + --dir_length; + skeleton_build = + xmalloc (dir_length + 1 + strlen (skeleton_user) + 1); + if (dir_length > 0) + { + strncpy (skeleton_build, current_file, dir_length); + skeleton_build[dir_length++] = '/'; + } + strcpy (skeleton_build + dir_length, skeleton_user); + skeleton_user = uniqstr_new (skeleton_build); + free (skeleton_build); + } + skeleton_arg (skeleton_user, grammar_prio, @1); + } | "%token-table" { token_table_flag = true; } -| "%verbose" { report_flag = report_states; } +| "%verbose" { report_flag |= report_states; } | "%yacc" { yacc_flag = true; } | /*FIXME: Err? What is this horror doing here? */ ";" ; +params: + params "{...}" { add_param (current_param, $2, @2); } +| "{...}" { add_param (current_param, $1, @1); } +; + + +/*----------------------. +| grammar_declaration. | +`----------------------*/ + grammar_declaration: precedence_declaration | symbol_declaration @@ -291,14 +389,14 @@ grammar_declaration: { symbol_list *list; for (list = $3; list; list = list->next) - symbol_list_destructor_set (list, $2, @2); + symbol_list_destructor_set (list, $2, @2); symbol_list_free ($3); } | "%printer" "{...}" generic_symlist { symbol_list *list; for (list = $3; list; list = list->next) - symbol_list_printer_set (list, $2, @2); + symbol_list_printer_set (list, $2, @2); symbol_list_free ($3); } | "%default-prec" @@ -311,28 +409,22 @@ grammar_declaration: } | "%code" braceless { - muscle_code_grow ("percent_code", $2, @2); + /* Do not invoke muscle_percent_code_grow here since it invokes + muscle_user_name_list_grow. */ + muscle_code_grow ("percent_code()", $2, @2); code_scanner_last_string_free (); } -| "%code" STRING braceless +| "%code" ID braceless { - char const name_prefix[] = "percent_code_"; - char *name = xmalloc (sizeof name_prefix + strlen ($2)); - strcpy (name, name_prefix); - strcpy (name + sizeof name_prefix - 1, $2); - muscle_code_grow (uniqstr_new (name), $3, @3); - free (name); + muscle_percent_code_grow ($2, @2, $3, @3); code_scanner_last_string_free (); - muscle_grow ("used_percent_code_qualifiers", "[[", ","); - muscle_grow ("used_percent_code_qualifiers", $2, ""); - muscle_grow ("used_percent_code_qualifiers", "]]", ""); } ; -/*----------* - | %union. | - *----------*/ +/*---------. +| %union. | +`---------*/ %token PERCENT_UNION "%union"; @@ -342,28 +434,11 @@ union_name: ; grammar_declaration: - "%union" union_name "{...}" - { - char const *body = $3; - - /* Concatenate the %union bodies. If this is the first %union, make sure - the synchronization line appears after the opening '{' so as not to - confuse Doxygen. Otherwise, turn the previous %union's trailing '}' - into '\n', and omit the new %union's leading '{'. */ - if (!union_seen) - { - muscle_grow ("stype", "{", ""); - } - else - { - char *code = muscle_find ("stype"); - code[strlen (code) - 1] = '\n'; - } - body++; - + "%union" union_name braceless + { union_seen = true; - muscle_code_grow ("stype", body, @3); - gram_scanner_last_string_free (); + muscle_code_grow ("stype", $3, @3); + code_scanner_last_string_free (); } ; @@ -381,42 +456,56 @@ symbol_declaration: current_class = unknown_sym; current_type = NULL; } -| "%type" TYPE symbols.1 +| "%type" TAG symbols.1 { symbol_list *list; tag_seen = true; for (list = $3; list; list = list->next) - symbol_type_set (list->content.sym, $2, @2); + symbol_type_set (list->content.sym, $2, @2); symbol_list_free ($3); } ; precedence_declaration: - precedence_declarator type.opt symbols.1 + precedence_declarator tag.opt symbols.prec { symbol_list *list; ++current_prec; for (list = $3; list; list = list->next) - { - symbol_type_set (list->content.sym, current_type, @2); - symbol_precedence_set (list->content.sym, current_prec, $1, @1); - } + { + symbol_type_set (list->content.sym, current_type, @2); + symbol_precedence_set (list->content.sym, current_prec, $1, @1); + } symbol_list_free ($3); current_type = NULL; } ; precedence_declarator: - "%left" { $$ = left_assoc; } -| "%right" { $$ = right_assoc; } -| "%nonassoc" { $$ = non_assoc; } + "%left" { $$ = left_assoc; } +| "%right" { $$ = right_assoc; } +| "%nonassoc" { $$ = non_assoc; } +| "%precedence" { $$ = precedence_assoc; } ; -type.opt: +tag.opt: /* Nothing. */ { current_type = NULL; } -| TYPE { current_type = $1; tag_seen = true; } +| TAG { current_type = $1; tag_seen = true; } +; + +/* Just like symbols.1 but accept INT for the sake of POSIX. */ +symbols.prec: + symbol.prec + { $$ = symbol_list_sym_new ($1, @1); } +| symbols.prec symbol.prec + { $$ = symbol_list_prepend ($1, symbol_list_sym_new ($2, @2)); } ; +symbol.prec: + symbol { $$ = $1; } + | symbol INT { $$ = $1; symbol_user_token_number_set ($1, $2, @2); } + ; + /* One or more symbols to be %typed. */ symbols.1: symbol @@ -431,15 +520,15 @@ generic_symlist: ; generic_symlist_item: - symbol { $$ = symbol_list_sym_new ($1, @1); } -| TYPE { $$ = symbol_list_type_new ($1, @1); } -| "<*>" { $$ = symbol_list_default_tagged_new (@1); } -| "<>" { $$ = symbol_list_default_tagless_new (@1); } + symbol { $$ = symbol_list_sym_new ($1, @1); } +| TAG { $$ = symbol_list_type_new ($1, @1); } +| "<*>" { $$ = symbol_list_default_tagged_new (@1); } +| "<>" { $$ = symbol_list_default_tagless_new (@1); } ; /* One token definition. */ symbol_def: - TYPE + TAG { current_type = $1; tag_seen = true; @@ -477,9 +566,9 @@ symbol_defs.1: ; - /*------------------------------------------. - | The grammar section: between the two %%. | - `------------------------------------------*/ + /*------------------------------------------. + | The grammar section: between the two %%. | + `------------------------------------------*/ grammar: rules_or_grammar_declaration @@ -498,7 +587,11 @@ rules_or_grammar_declaration: ; rules: - id_colon { current_lhs = $1; current_lhs_location = @1; } rhses.1 + id_colon named_ref.opt { current_lhs ($1, @1, $2); } rhses.1 + { + /* Free the current lhs. */ + current_lhs (0, @1, 0); + } ; rhses.1: @@ -509,40 +602,51 @@ rhses.1: rhs: /* Nothing. */ - { grammar_current_rule_begin (current_lhs, current_lhs_location); } -| rhs symbol - { grammar_current_rule_symbol_append ($2, @2); } -| rhs "{...}" - { grammar_current_rule_action_append ($2, @2); } + { grammar_current_rule_begin (current_lhs_symbol, current_lhs_location, + current_lhs_named_ref); } +| rhs symbol named_ref.opt + { grammar_current_rule_symbol_append ($2, @2, $3); } +| rhs "{...}" named_ref.opt + { grammar_current_rule_action_append ($2, @2, $3, false); } +| rhs "%?{...}" + { grammar_current_rule_action_append ($2, @2, NULL, true); } | rhs "%prec" symbol { grammar_current_rule_prec_set ($3, @3); } | rhs "%dprec" INT { grammar_current_rule_dprec_set ($3, @3); } -| rhs "%merge" TYPE +| rhs "%merge" TAG { grammar_current_rule_merge_set ($3, @3); } ; +named_ref.opt: + /* Nothing. */ { $$ = 0; } +| + BRACKETED_ID { $$ = named_ref_new($1, @1); } +; -/*-----------* - | content. | - *-----------*/ +/*---------------------------. +| variable and content.opt. | +`---------------------------*/ -content: - STRING -| braceless +/* The STRING form of variable is deprecated and is not M4-friendly. + For example, M4 fails for `%define "[" "value"'. */ +variable: + ID +| STRING { $$ = uniqstr_new ($1); } ; -/* Some content or "1" by default. */ +/* Some content or empty by default. */ content.opt: - /* Nothing. */ - { - static char one[] = "1"; - $$ = one; - } -| content + /* Nothing. */ { $$ = ""; } +| ID { $$ = $1; } +| STRING ; +/*------------. +| braceless. | +`------------*/ + braceless: "{...}" { @@ -556,9 +660,9 @@ braceless: ; -/*---------------* - | Identifiers. | - *---------------*/ +/*--------------. +| Identifiers. | +`--------------*/ /* Identifiers are returned as uniqstr values by the scanner. Depending on their use, we may need to make them genuine symbols. */ @@ -625,48 +729,47 @@ lloc_default (YYLTYPE const *rhs, int n) loc.start = rhs[n].end; loc.end = rhs[n].end; - /* Ignore empty nonterminals the start of the the right-hand side. + /* Ignore empty nonterminals the start of the right-hand side. Do not bother to ignore them at the end of the right-hand side, since empty nonterminals have the same end as their predecessors. */ for (i = 1; i <= n; i++) if (! equal_boundaries (rhs[i].start, rhs[i].end)) { - loc.start = rhs[i].start; - break; + loc.start = rhs[i].start; + break; } return loc; } -/* Add a lex-param or a parse-param (depending on TYPE) with - declaration DECL and location LOC. */ - static void -add_param (char const *type, char *decl, location loc) +add_param (param_type type, char *decl, location loc) { static char const alphanum[26 + 26 + 1 + 10] = "abcdefghijklmnopqrstuvwxyz" "ABCDEFGHIJKLMNOPQRSTUVWXYZ" "_" "0123456789"; + char const *name_start = NULL; - char *p; - - /* Stop on last actual character. */ - for (p = decl; p[1]; p++) - if ((p == decl - || ! memchr (alphanum, p[-1], sizeof alphanum)) - && memchr (alphanum, p[0], sizeof alphanum - 10)) - name_start = p; - - /* Strip the surrounding '{' and '}', and any blanks just inside - the braces. */ - while (*--p == ' ' || *p == '\t') - continue; - p[1] = '\0'; - while (*++decl == ' ' || *decl == '\t') - continue; + { + char *p; + /* Stop on last actual character. */ + for (p = decl; p[1]; p++) + if ((p == decl + || ! memchr (alphanum, p[-1], sizeof alphanum)) + && memchr (alphanum, p[0], sizeof alphanum - 10)) + name_start = p; + + /* Strip the surrounding '{' and '}', and any blanks just inside + the braces. */ + while (*--p == ' ' || *p == '\t') + continue; + p[1] = '\0'; + while (*++decl == ' ' || *decl == '\t') + continue; + } if (! name_start) complain_at (loc, _("missing identifier in parameter declaration")); @@ -676,14 +779,17 @@ add_param (char const *type, char *decl, location loc) size_t name_len; for (name_len = 1; - memchr (alphanum, name_start[name_len], sizeof alphanum); - name_len++) - continue; + memchr (alphanum, name_start[name_len], sizeof alphanum); + name_len++) + continue; name = xmalloc (name_len + 1); memcpy (name, name_start, name_len); name[name_len] = '\0'; - muscle_pair_list_grow (type, decl, name); + if (type & param_lex) + muscle_pair_list_grow ("lex_param", decl, name); + if (type & param_parse) + muscle_pair_list_grow ("parse_param", decl, name); free (name); } @@ -697,8 +803,8 @@ version_check (location const *loc, char const *version) if (strverscmp (version, PACKAGE_VERSION) > 0) { complain_at (*loc, "require bison %s, but have %s", - version, PACKAGE_VERSION); - exit (63); + version, PACKAGE_VERSION); + exit (EX_MISMATCH); } }