X-Git-Url: https://git.saurik.com/bison.git/blobdiff_plain/4f82b42a9d5ce6f0baa70714f25346ba9303be73..457bf91968a87a01dfb5317b008497ac4dcde8a4:/src/parse-gram.y diff --git a/src/parse-gram.y b/src/parse-gram.y index e9ec6366..3120bfc2 100644 --- a/src/parse-gram.y +++ b/src/parse-gram.y @@ -1,12 +1,12 @@ %{/* Bison Grammar Parser -*- C -*- - Copyright (C) 2002, 2003, 2004, 2005, 2006 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, @@ -15,27 +15,25 @@ 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 "c-ctype.h" #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" #include "scan-gram.h" #include "scan-code.h" +#include "xmemdup0.h" #define YYLLOC_DEFAULT(Current, Rhs, N) (Current) = lloc_default (Rhs, N) static YYLTYPE lloc_default (YYLTYPE const *, int); @@ -65,10 +63,30 @@ 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 symbol *current_lhs_symbol; static location current_lhs_location; +static named_ref *current_lhs_named_ref; static int current_prec = 0; +/** 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 @@ -81,7 +99,7 @@ static int current_prec = 0; %locations %pure-parser %error-verbose -%defines +%define parse.lac full %name-prefix="gram_" %expect 0 @@ -93,16 +111,17 @@ static int current_prec = 0; boundary_set (&@$.end, current_file, 1, 1); } -/* Only NUMBERS have a value. */ %union { symbol *symbol; symbol_list *list; int integer; - char *chars; + char const *chars; + char *code; assoc assoc; uniqstr uniqstr; unsigned char character; + named_ref *named_ref; }; /* Define the tokens together with their human representation. */ @@ -116,8 +135,6 @@ static int current_prec = 0; %token PERCENT_TYPE "%type" %token PERCENT_DESTRUCTOR "%destructor" %token PERCENT_PRINTER "%printer" -%token PERCENT_SYMBOL_DEFAULT - "%symbol-default" %token PERCENT_LEFT "%left" %token PERCENT_RIGHT "%right" @@ -133,19 +150,18 @@ static int current_prec = 0; `----------------------*/ %token - PERCENT_AFTER_HEADER "%after-header" - PERCENT_BEFORE_HEADER "%before-header" + PERCENT_CODE "%code" PERCENT_DEBUG "%debug" PERCENT_DEFAULT_PREC "%default-prec" PERCENT_DEFINE "%define" PERCENT_DEFINES "%defines" - PERCENT_END_HEADER "%end-header" PERCENT_ERROR_VERBOSE "%error-verbose" PERCENT_EXPECT "%expect" PERCENT_EXPECT_RR "%expect-rr" 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" @@ -156,17 +172,16 @@ static int current_prec = 0; PERCENT_OUTPUT "%output" PERCENT_PARSE_PARAM "%parse-param" PERCENT_PURE_PARSER "%pure-parser" - PERCENT_PUSH_PARSER "%push-parser" PERCENT_REQUIRE "%require" PERCENT_SKELETON "%skeleton" PERCENT_START "%start" - PERCENT_START_HEADER "%start-header" PERCENT_TOKEN_TABLE "%token-table" PERCENT_VERBOSE "%verbose" PERCENT_YACC "%yacc" ; %token BRACED_CODE "{...}" +%token BRACKETED_ID "[identifier]" %token CHAR "char" %token EPILOGUE "epilogue" %token EQUAL "=" @@ -177,32 +192,38 @@ static int current_prec = 0; %token PROLOGUE "%{...%}" %token SEMICOLON ";" %token TYPE "type" +%token TYPE_TAG_ANY "<*>" +%token TYPE_TAG_NONE "<>" %type CHAR %printer { fputs (char_name ($$), stderr); } CHAR /* braceless is not to be used for rule or symbol actions, as it - calls translate_code. */ -%type STRING "{...}" "%{...%}" EPILOGUE braceless content content.opt + calls code_props_plain_init. */ +%type STRING "%{...%}" EPILOGUE braceless content.opt +%type "{...}" %printer { fputs (quotearg_style (c_quoting_style, $$), 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 ($$, stderr); } ID +%type BRACKETED_ID ID ID_COLON TYPE variable +%printer { fputs ($$, stderr); } +%printer { fprintf (stderr, "[%s]", $$); } BRACKETED_ID %printer { fprintf (stderr, "%s:", $$); } ID_COLON +%printer { fprintf (stderr, "<%s>", $$); } TYPE %type INT -%printer { fprintf (stderr, "%d", $$); } 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 + %% input: @@ -221,17 +242,33 @@ prologue_declarations: prologue_declaration: grammar_declaration -| "%{...%}" { prologue_augment (translate_code ($1, @1), @1, union_seen); } -| "%after-header" braceless { prologue_augment ($2, @2, true); } -| "%before-header" braceless { prologue_augment ($2, @2, false); } +| "%{...%}" + { + code_props plain_code; + code_props_plain_init (&plain_code, $1, @1); + code_props_translate_code (&plain_code); + gram_scanner_last_string_free (); + 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); } +| "%define" variable content.opt + { + muscle_percent_define_insert ($2, @2, $3, + MUSCLE_PERCENT_DEFINE_GRAMMAR_FILE); + } | "%defines" { defines_flag = true; } -| "%end-header" braceless { muscle_code_grow ("end_header", $2, @2); } +| "%defines" STRING + { + defines_flag = true; + spec_defines_file = xstrdup ($2); + } | "%error-verbose" { error_verbose = true; } | "%expect" INT { expected_sr_conflicts = $2; } | "%expect-rr" INT { expected_rr_conflicts = $2; } -| "%file-prefix" "=" STRING { spec_file_prefix = $3; } +| "%file-prefix" STRING { spec_file_prefix = $2; } +| "%file-prefix" "=" STRING { spec_file_prefix = $3; } /* deprecated */ | "%glr-parser" { nondeterministic_parser = true; @@ -239,22 +276,65 @@ prologue_declaration: } | "%initial-action" "{...}" { - muscle_code_grow ("initial_action", translate_symbol_action ($2, @2), @2); + code_props action; + code_props_symbol_action_init (&action, $2, @2); + code_props_translate_code (&action); + gram_scanner_last_string_free (); + muscle_code_grow ("initial_action", action.code, @2); + code_scanner_last_string_free (); } +| "%language" STRING { language_argmatch ($2, grammar_prio, @1); } | "%lex-param" "{...}" { add_param ("lex_param", $2, @2); } | "%locations" { locations_flag = true; } -| "%name-prefix" "=" STRING { spec_name_prefix = $3; } +| "%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; } -| "%output" "=" STRING { spec_outfile = $3; } +| "%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; } +| "%pure-parser" + { + /* %pure-parser is deprecated in favor of `%define api.pure', so use + `%define api.pure' in a backward-compatible manner here. First, don't + complain if %pure-parser is specified multiple times. */ + if (!muscle_find_const ("percent_define(api.pure)")) + muscle_percent_define_insert ("api.pure", @1, "", + MUSCLE_PERCENT_DEFINE_GRAMMAR_FILE); + /* In all cases, use api.pure now so that the backend doesn't complain if + the skeleton ignores api.pure, but do warn now if there's a previous + conflicting definition from an actual %define. */ + if (!muscle_percent_define_flag_if ("api.pure")) + muscle_percent_define_insert ("api.pure", @1, "", + MUSCLE_PERCENT_DEFINE_GRAMMAR_FILE); + } | "%require" STRING { version_check (&@2, $2); } -| "%skeleton" STRING { skeleton = $2; } -| "%start-header" braceless { muscle_code_grow ("start_header", $2, @2); } +| "%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) + { + memcpy (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? */ ";" ; @@ -268,19 +348,27 @@ grammar_declaration: } | "%destructor" "{...}" generic_symlist { - symbol_list *list; - const char *action = translate_symbol_action ($2, @2); - for (list = $3; list; list = list->next) - symbol_list_destructor_set (list, action, @2); - symbol_list_free ($3); + code_props code; + code_props_symbol_action_init (&code, $2, @2); + code_props_translate_code (&code); + { + symbol_list *list; + for (list = $3; list; list = list->next) + symbol_list_destructor_set (list, &code); + symbol_list_free ($3); + } } | "%printer" "{...}" generic_symlist { - symbol_list *list; - const char *action = translate_symbol_action ($2, @2); - for (list = $3; list; list = list->next) - symbol_list_printer_set (list, action, @2); - symbol_list_free ($3); + code_props code; + code_props_symbol_action_init (&code, $2, @2); + code_props_translate_code (&code); + { + symbol_list *list; + for (list = $3; list; list = list->next) + symbol_list_printer_set (list, &code); + symbol_list_free ($3); + } } | "%default-prec" { @@ -290,6 +378,18 @@ grammar_declaration: { default_prec = false; } +| "%code" braceless + { + /* 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" ID braceless + { + muscle_percent_code_grow ($2, @2, $3, @3); + code_scanner_last_string_free (); + } ; @@ -305,27 +405,11 @@ union_name: ; grammar_declaration: - "%union" union_name "{...}" + "%union" union_name braceless { - 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_seen = true; - muscle_code_grow ("stype", body, @3); + muscle_code_grow ("stype", $3, @3); + code_scanner_last_string_free (); } ; @@ -354,7 +438,7 @@ symbol_declaration: ; precedence_declaration: - precedence_declarator type.opt symbols.1 + precedence_declarator type.opt symbols.prec { symbol_list *list; ++current_prec; @@ -379,6 +463,19 @@ type.opt: | TYPE { 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 @@ -395,7 +492,8 @@ generic_symlist: generic_symlist_item: symbol { $$ = symbol_list_sym_new ($1, @1); } | TYPE { $$ = symbol_list_type_new ($1, @1); } -| "%symbol-default" { $$ = symbol_list_default_new (@1); } +| "<*>" { $$ = symbol_list_default_tagged_new (@1); } +| "<>" { $$ = symbol_list_default_tagless_new (@1); } ; /* One token definition. */ @@ -459,7 +557,11 @@ rules_or_grammar_declaration: ; rules: - id_colon { current_lhs = $1; current_lhs_location = @1; } rhses.1 + id_colon named_ref.opt { current_lhs ($1, @1, $2); } rhses.1 + { + /* Free the current lhs. */ + current_lhs (0, @1, 0); + } ; rhses.1: @@ -470,11 +572,12 @@ rhses.1: rhs: /* Nothing. */ - { grammar_current_rule_begin (current_lhs, current_lhs_location); } -| rhs symbol - { grammar_current_rule_symbol_append ($2, @2); } -| rhs "{...}" - { grammar_current_rule_action_append ($2, @2); } + { grammar_current_rule_begin (current_lhs_symbol, current_lhs_location, + current_lhs_named_ref); } +| rhs symbol named_ref.opt + { grammar_current_rule_symbol_append ($2, @2, $3); } +| rhs "{...}" named_ref.opt + { grammar_current_rule_action_append ($2, @2, $3); } | rhs "%prec" symbol { grammar_current_rule_prec_set ($3, @3); } | rhs "%dprec" INT @@ -483,32 +586,45 @@ rhs: { grammar_current_rule_merge_set ($3, @3); } ; +named_ref.opt: + /* Nothing. */ { $$ = 0; } +| + BRACKETED_ID { $$ = named_ref_new($1, @1); } +; -/*-----------* - | content. | - *-----------*/ -content: - STRING -| braceless +/*----------------------------* + | variable and content.opt. | + *---------------------------*/ + +/* The STRING form of variable is deprecated and is not M4-friendly. + For example, M4 fails for `%define "[" "value"'. */ +variable: + ID +| STRING { $$ = 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 { $$ = $1; } +| STRING ; +/*-------------* + | braceless. | + *-------------*/ + braceless: "{...}" { + code_props plain_code; $1[strlen ($1) - 1] = '\n'; - $$ = translate_code ($1 + 1, @1); + code_props_plain_init (&plain_code, $1+1, @1); + code_props_translate_code (&plain_code); + gram_scanner_last_string_free (); + $$ = plain_code.code; } ; @@ -554,8 +670,12 @@ epilogue.opt: /* Nothing. */ | "%%" EPILOGUE { - muscle_code_grow ("epilogue", translate_code ($2, @2), @2); + code_props plain_code; + code_props_plain_init (&plain_code, $2, @2); + code_props_translate_code (&plain_code); gram_scanner_last_string_free (); + muscle_code_grow ("epilogue", plain_code.code, @2); + code_scanner_last_string_free (); } ; @@ -578,7 +698,7 @@ 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++) @@ -615,27 +735,19 @@ add_param (char const *type, char *decl, location loc) /* Strip the surrounding '{' and '}', and any blanks just inside the braces. */ - while (*--p == ' ' || *p == '\t') - continue; + --p; + while (c_isspace ((unsigned char) *p)) + --p; p[1] = '\0'; - while (*++decl == ' ' || *decl == '\t') - continue; + ++decl; + while (c_isspace ((unsigned char) *decl)) + ++decl; if (! name_start) complain_at (loc, _("missing identifier in parameter declaration")); else { - char *name; - size_t name_len; - - for (name_len = 1; - 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'; + char *name = xmemdup0 (name_start, strspn (name_start, alphanum)); muscle_pair_list_grow (type, decl, name); free (name); } @@ -650,8 +762,8 @@ 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); } }