X-Git-Url: https://git.saurik.com/bison.git/blobdiff_plain/2c5690250c9489f1ec3cd083ee754938bb47707f..f377f69fec28013c79db4efe12bbb9d48987fb2c:/src/parse-gram.y?ds=sidebyside diff --git a/src/parse-gram.y b/src/parse-gram.y index d1ebbd67..a6d8ec70 100644 --- a/src/parse-gram.y +++ b/src/parse-gram.y @@ -30,16 +30,32 @@ %{ #include "system.h" +#include "complain.h" #include "muscle_tab.h" #include "files.h" #include "getargs.h" #include "output.h" +#include "symlist.h" #include "gram.h" #include "reader.h" #include "conflicts.h" /* Produce verbose parse errors. */ #define YYERROR_VERBOSE 1 +#define YYLLOC_DEFAULT(Current, Rhs, N) \ +do { \ + if (N) \ + { \ + Current.first_column = Rhs[1].first_column; \ + Current.first_line = Rhs[1].first_line; \ + Current.last_column = Rhs[N].last_column; \ + Current.last_line = Rhs[N].last_line; \ + } \ + else \ + { \ + Current = Rhs[0]; \ + } \ +} while (0) /* Pass the control structure to YYPARSE and YYLEX. */ #define YYPARSE_PARAM gram_control @@ -48,24 +64,23 @@ correctly typed access to it. */ #define yycontrol ((gram_control_t *) gram_control) -/* Request detailed parse error messages, and pass them to - GRAM_ERROR. */ +/* Request detailed parse error messages, and pass them to GRAM_ERROR. + FIXME: depends on the undocumented availability of YYLLOC.t */ #undef yyerror #define yyerror(Msg) \ - gram_error (yycontrol, &yylloc, Msg) + gram_error (&yylloc, Msg) -/* When debugging our pure parser, we want to see values and locations - of the tokens. */ #define YYPRINT(File, Type, Value) \ - yyprint (File, &yylloc, Type, &Value) -static void yyprint (FILE *file, const location_t *loc, - int type, const yystype *value); + yyprint (File, Type, &Value) +static void yyprint (FILE *file, int type, const yystype *value); symbol_class current_class = unknown_sym; char *current_type = 0; symbol_t *current_lhs; -associativity current_assoc; +location_t current_lhs_location; +assoc_t current_assoc; int current_prec = 0; +braced_code_t current_braced_code = action_braced_code; %} @@ -73,69 +88,82 @@ int current_prec = 0; %union { symbol_t *symbol; + symbol_list_t *list; int integer; char *string; - associativity assoc; + assoc_t assoc; }; -/* Define the tokens together with there human representation. */ -%token GRAM_EOF 0 "end of string" -%token STRING CHARACTER -%token INT - -%token PERCENT_TOKEN "%token" -%token PERCENT_NTERM "%nterm" -%token PERCENT_TYPE "%type" -%token PERCENT_UNION "%union" -%token PERCENT_EXPECT "%expect" -%token PERCENT_START "%start" -%token PERCENT_LEFT "%left" -%token PERCENT_RIGHT "%right" -%token PERCENT_NONASSOC "%nonassoc" -%token PERCENT_PREC "%prec" -%token PERCENT_VERBOSE "%verbose" +/* Define the tokens together with their human representation. */ +%token GRAM_EOF 0 "end of file" +%token STRING "string" +%token CHARACTER "character" +%token INT "integer" + +%token PERCENT_TOKEN "%token" +%token PERCENT_NTERM "%nterm" + +%token PERCENT_TYPE "%type" +%token PERCENT_DESTRUCTOR "%destructor" +%token PERCENT_PRINTER "%printer" + +%token PERCENT_UNION "%union" + +%token PERCENT_LEFT "%left" +%token PERCENT_RIGHT "%right" +%token PERCENT_NONASSOC "%nonassoc" + +%token PERCENT_EXPECT "%expect" +%token PERCENT_START "%start" +%token PERCENT_PREC "%prec" +%token PERCENT_DPREC "%dprec" +%token PERCENT_MERGE "%merge" +%token PERCENT_VERBOSE "%verbose" %token PERCENT_ERROR_VERBOSE "%error-verbose" -%token PERCENT_OUTPUT "%output" +%token PERCENT_OUTPUT "%output" %token PERCENT_FILE_PREFIX "%file-prefix" %token PERCENT_NAME_PREFIX "%name-prefix" -%token PERCENT_DEFINE "%define" +%token PERCENT_DEFINE "%define" %token PERCENT_PURE_PARSER "%pure-parser" +%token PERCENT_GLR_PARSER "%glr-parser" %token PERCENT_DEFINES "%defines" %token PERCENT_YACC "%yacc" -%token PERCENT_DEBUG "%debug" -%token PERCENT_LOCATIONS "%locations" -%token PERCENT_NO_LINES "%no-lines" -%token PERCENT_SKELETON "%skeleton" +%token PERCENT_DEBUG "%debug" +%token PERCENT_LOCATIONS "%locations" +%token PERCENT_NO_LINES "%no-lines" +%token PERCENT_SKELETON "%skeleton" %token PERCENT_TOKEN_TABLE "%token-table" -%token TYPE -%token EQUAL "=" -%token SEMICOLON ";" -%token COLON ":" -%token PIPE "|" -%token ID "identifier" +%token TYPE "type" +%token EQUAL "=" +%token SEMICOLON ";" +%token COLON ":" +%token PIPE "|" +%token ID "identifier" %token PERCENT_PERCENT "%%" -%token PROLOGUE EPILOGUE -%token BRACED_CODE +%token PROLOGUE "%{...%}" +%token EPILOGUE "epilogue" +%token BRACED_CODE "{...}" + %type CHARACTER TYPE STRING string_content BRACED_CODE PROLOGUE EPILOGUE epilogue.opt action %type INT %type ID symbol string_as_id %type precedence_declarator - +%type symbols.1 %% -input: { LOCATION_RESET (yylloc); } +input: declarations "%%" grammar epilogue.opt { yycontrol->errcode = 0; - epilogue_set ($5, @5); + epilogue_set ($4, @4); } ; @@ -146,7 +174,7 @@ input: { LOCATION_RESET (yylloc); } declarations: /* Nothing */ -| declarations declaration +| declarations declaration semi_colon.opt ; declaration: @@ -163,6 +191,7 @@ declaration: | "%no-lines" { no_lines_flag = 1; } | "%output" "=" string_content { spec_outfile = $3; } | "%pure-parser" { pure_parser = 1; } +| "%glr-parser" { glr_parser = 1; } | "%skeleton" string_content { skeleton = $2; } | "%token-table" { token_table_flag = 1; } | "%verbose" { report_flag = 1; } @@ -174,14 +203,34 @@ grammar_declaration: | symbol_declaration | "%start" symbol { - grammar_start_symbol_set ($2); + grammar_start_symbol_set ($2, @2); } -| "%union" BRACED_CODE semi_colon_opt +| "%union" BRACED_CODE { typed = 1; MUSCLE_INSERT_INT ("stype_line", @2.first_line); muscle_insert ("stype", $2); } +| "%destructor" + { current_braced_code = destructor_braced_code; } + BRACED_CODE symbols.1 + { + symbol_list_t *list; + for (list = $4; list; list = list->next) + symbol_destructor_set (list->sym, $3, @3); + symbol_list_free ($4); + current_braced_code = action_braced_code; + } +| "%printer" + { current_braced_code = printer_braced_code; } + BRACED_CODE symbols.1 + { + symbol_list_t *list; + for (list = $4; list; list = list->next) + symbol_printer_set (list->sym, $3, list->location); + symbol_list_free ($4); + current_braced_code = action_braced_code; + } ; symbol_declaration: @@ -195,17 +244,28 @@ symbol_declaration: current_class = unknown_sym; current_type = NULL; } -| "%type" TYPE {current_type = $2; } nterms_to_type.1 +| "%type" TYPE symbols.1 { - current_type = NULL; + symbol_list_t *list; + for (list = $3; list; list = list->next) + symbol_type_set (list->sym, $2, @2); + symbol_list_free ($3); } ; precedence_declaration: - precedence_declarator type.opt - { current_assoc = $1; ++current_prec; } - terms_to_prec.1 - { current_assoc = non_assoc; current_type = NULL; } + precedence_declarator type.opt symbols.1 + { + symbol_list_t *list; + ++current_prec; + for (list = $3; list; list = list->next) + { + symbol_type_set (list->sym, current_type, @2); + symbol_precedence_set (list->sym, current_prec, $1, @1); + } + symbol_list_free ($3); + current_type = NULL; + } ; precedence_declarator: @@ -220,23 +280,10 @@ type.opt: ; /* One or more nonterminals to be %typed. */ -nterms_to_type.1: - ID { symbol_type_set ($1, current_type); } -| nterms_to_type.1 ID { symbol_type_set ($2, current_type); } -; -/* One or more symbols to be given a precedence/associativity. */ -terms_to_prec.1: - symbol - { - symbol_type_set ($1, current_type); - symbol_precedence_set ($1, current_prec, current_assoc); - } -| terms_to_prec.1 symbol - { - symbol_type_set ($2, current_type); - symbol_precedence_set ($2, current_prec, current_assoc); - } +symbols.1: + symbol { $$ = symbol_list_new ($1, @1); } +| symbols.1 symbol { $$ = symbol_list_prepend ($1, $2, @2); } ; /* One token definition. */ @@ -247,27 +294,27 @@ symbol_def: } | ID { - symbol_class_set ($1, current_class); - symbol_type_set ($1, current_type); + symbol_class_set ($1, current_class, @1); + symbol_type_set ($1, current_type, @1); } | ID INT { - symbol_class_set ($1, current_class); - symbol_type_set ($1, current_type); - symbol_user_token_number_set ($1, $2); + symbol_class_set ($1, current_class, @1); + symbol_type_set ($1, current_type, @1); + symbol_user_token_number_set ($1, $2, @2); } | ID string_as_id { - symbol_class_set ($1, current_class); - symbol_type_set ($1, current_type); - symbol_make_alias ($1, $2); + symbol_class_set ($1, current_class, @1); + symbol_type_set ($1, current_type, @1); + symbol_make_alias ($1, $2, @$); } | ID INT string_as_id { - symbol_class_set ($1, current_class); - symbol_type_set ($1, current_type); - symbol_user_token_number_set ($1, $2); - symbol_make_alias ($1, $3); + symbol_class_set ($1, current_class, @1); + symbol_type_set ($1, current_type, @1); + symbol_user_token_number_set ($1, $2, @2); + symbol_make_alias ($1, $3, @$); } ; @@ -285,35 +332,55 @@ symbol_defs.1: `------------------------------------------*/ grammar: + rules_or_grammar_declaration +| grammar rules_or_grammar_declaration +; + +/* As a Bison extension, one can use the grammar declarations in the + body of the grammar. But to remain LALR(1), they must be ended + with a semi-colon. */ +rules_or_grammar_declaration: rules -| grammar rules +| grammar_declaration ";" + { + if (yacc_flag) + complain_at (@$, _("POSIX forbids declarations in the grammar")); + } +| error ";" + { + yyerrok; + } ; rules: - ID ":" { current_lhs = $1; } rhses.1 ";" + ID ":" { current_lhs = $1; current_lhs_location = @1; } rhses.1 ";" {;} ; rhses.1: - rhs { grammar_rule_end (); } -| rhses.1 "|" rhs { grammar_rule_end (); } + rhs { grammar_rule_end (@1); } +| rhses.1 "|" rhs { grammar_rule_end (@3); } ; rhs: /* Nothing. */ - { grammar_rule_begin (current_lhs); } + { grammar_rule_begin (current_lhs, current_lhs_location); } | rhs symbol - { grammar_current_rule_symbol_append ($2); } + { grammar_current_rule_symbol_append ($2, @2); } | rhs action - { grammar_current_rule_action_append ($2, @2.first_line); } + { grammar_current_rule_action_append ($2, @2); } | rhs "%prec" symbol - { grammar_current_rule_prec_set ($3); } + { grammar_current_rule_prec_set ($3, @3); } +| rhs "%dprec" INT + { grammar_current_rule_dprec_set ($3, @3); } +| rhs "%merge" TYPE + { grammar_current_rule_merge_set ($3, @3); } ; symbol: ID { $$ = $1; } | string_as_id { $$ = $1; } -| CHARACTER { $$ = getsym ($1); } +| CHARACTER { $$ = symbol_get ($1, @1); } ; action: @@ -325,8 +392,8 @@ action: string_as_id: STRING { - $$ = getsym ($1); - symbol_class_set ($$, token_sym); + $$ = symbol_get ($1, @1); + symbol_class_set ($$, token_sym, @1); } ; @@ -350,7 +417,7 @@ epilogue.opt: } ; -semi_colon_opt: +semi_colon.opt: /* Nothing. */ | ";" ; @@ -361,11 +428,9 @@ semi_colon_opt: static void yyprint (FILE *file, - const location_t *loc, int type, const yystype *value) + int type, const yystype *value) { - fputs (" (", file); - LOCATION_PRINT (file, *loc); - fputs (")", file); + fputc (' ', file); switch (type) { case CHARACTER: @@ -397,9 +462,7 @@ yyprint (FILE *file, } void -gram_error (gram_control_t *control ATTRIBUTE_UNUSED, - location_t *yylloc, const char *msg) +gram_error (location_t *yylloc, const char *msg) { - LOCATION_PRINT (stderr, *yylloc); - fprintf (stderr, ": %s\n", msg); + complain_at (*yylloc, "%s", msg); }