X-Git-Url: https://git.saurik.com/bison.git/blobdiff_plain/b729552289115a6da0b700896cdc4ab7f4443b1a..927b425baae16fc1d50e092b78b944e281b521f6:/src/parse-gram.y diff --git a/src/parse-gram.y b/src/parse-gram.y index e241b716..11cd489e 100644 --- a/src/parse-gram.y +++ b/src/parse-gram.y @@ -1,5 +1,6 @@ -/* Bison Grammar Parser -*- C -*- - Copyright (C) 2002 Free Software Foundation, Inc. +%{/* Bison Grammar Parser -*- C -*- + + Copyright (C) 2002, 2003, 2004, 2005, 2006 Free Software Foundation, Inc. This file is part of Bison, the GNU Compiler Compiler. @@ -15,69 +16,89 @@ 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., 59 Temple Place, Suite 330, Boston, MA - 02111-1307 USA + Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA + 02110-1301 USA */ - -%debug -%defines -%locations -%pure-parser -// %error-verbose -%defines -%name-prefix="gram_" - -%{ +#include #include "system.h" + #include "complain.h" -#include "muscle_tab.h" +#include "conflicts.h" #include "files.h" #include "getargs.h" -#include "output.h" -#include "symlist.h" #include "gram.h" +#include "muscle_tab.h" +#include "output.h" +#include "quotearg.h" #include "reader.h" -#include "conflicts.h" - -/* Produce verbose syntax errors. */ -#define YYERROR_VERBOSE 1 +#include "symlist.h" +#include "strverscmp.h" #define YYLLOC_DEFAULT(Current, Rhs, N) (Current) = lloc_default (Rhs, N) static YYLTYPE lloc_default (YYLTYPE const *, int); +#define YY_LOCATION_PRINT(File, Loc) \ + location_print (File, Loc) + +static void version_check (location const *loc, char const *version); + /* Request detailed syntax error messages, and pass them to GRAM_ERROR. FIXME: depends on the undocumented availability of YYLLOC. */ #undef yyerror #define yyerror(Msg) \ - gram_error (&yylloc, Msg) -static void gram_error (location_t const *, char const *); - -#define YYPRINT(File, Type, Value) \ - print_token_value (File, Type, &Value) -static void print_token_value (FILE *, int, YYSTYPE const *); - -static void add_param (char const *, char const *, location_t); - -symbol_class current_class = unknown_sym; -struniq_t current_type = 0; -symbol_t *current_lhs; -location_t current_lhs_location; -assoc_t current_assoc; -int current_prec = 0; -braced_code_t current_braced_code = action_braced_code; + gram_error (&yylloc, Msg) +static void gram_error (location const *, char const *); + +static void add_param (char const *, char *, location); + +static symbol_class current_class = unknown_sym; +static uniqstr current_type = 0; +static symbol *current_lhs; +static location current_lhs_location; +static int current_prec = 0; + +#ifdef UINT_FAST8_MAX +# define YYTYPE_UINT8 uint_fast8_t +#endif +#ifdef INT_FAST8_MAX +# define YYTYPE_INT8 int_fast8_t +#endif +#ifdef UINT_FAST16_MAX +# define YYTYPE_UINT16 uint_fast16_t +#endif +#ifdef INT_FAST16_MAX +# define YYTYPE_INT16 int_fast16_t +#endif %} +%debug +%verbose +%defines +%locations +%pure-parser +%error-verbose +%defines +%name-prefix="gram_" + +%initial-action +{ + /* Bison's grammar can initial empty locations, hence a default + location is needed. */ + @$.start.file = @$.end.file = current_file; + @$.start.line = @$.end.line = 1; + @$.start.column = @$.end.column = 0; +} /* Only NUMBERS have a value. */ %union { - symbol_t *symbol; - symbol_list_t *list; + symbol *symbol; + symbol_list *list; int integer; - char *string; - assoc_t assoc; - struniq_t struniq; + char *chars; + assoc assoc; + uniqstr uniqstr; }; /* Define the tokens together with their human representation. */ @@ -89,10 +110,10 @@ braced_code_t current_braced_code = action_braced_code; %token PERCENT_NTERM "%nterm" %token PERCENT_TYPE "%type" -%token PERCENT_DESTRUCTOR "%destructor" -%token PERCENT_PRINTER "%printer" +%token PERCENT_DESTRUCTOR "%destructor {...}" +%token PERCENT_PRINTER "%printer {...}" -%token PERCENT_UNION "%union" +%token PERCENT_UNION "%union {...}" %token PERCENT_LEFT "%left" %token PERCENT_RIGHT "%right" @@ -108,25 +129,32 @@ braced_code_t current_braced_code = action_braced_code; `----------------------*/ %token - PERCENT_DEBUG "%debug" - PERCENT_DEFINE "%define" - PERCENT_DEFINES "%defines" - PERCENT_ERROR_VERBOSE "%error-verbose" - PERCENT_EXPECT "%expect" - PERCENT_FILE_PREFIX "%file-prefix" - PERCENT_GLR_PARSER "%glr-parser" - PERCENT_LEX_PARAM "%lex-param" - PERCENT_LOCATIONS "%locations" - PERCENT_NAME_PREFIX "%name-prefix" - PERCENT_NO_LINES "%no-lines" - PERCENT_OUTPUT "%output" - PERCENT_PARSE_PARAM "%parse-param" - PERCENT_PURE_PARSER "%pure-parser" - PERCENT_SKELETON "%skeleton" - PERCENT_START "%start" - PERCENT_TOKEN_TABLE "%token-table" - PERCENT_VERBOSE "%verbose" - PERCENT_YACC "%yacc" + 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_FILE_PREFIX "%file-prefix" + PERCENT_GLR_PARSER "%glr-parser" + PERCENT_INITIAL_ACTION "%initial-action {...}" + 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" + PERCENT_OUTPUT "%output" + PERCENT_PARSE_PARAM "%parse-param {...}" + PERCENT_PURE_PARSER "%pure-parser" + PERCENT_REQUIRE "%require" + PERCENT_SKELETON "%skeleton" + PERCENT_START "%start" + PERCENT_TOKEN_TABLE "%token-table" + PERCENT_VERBOSE "%verbose" + PERCENT_YACC "%yacc" ; %token TYPE "type" @@ -141,12 +169,32 @@ braced_code_t current_braced_code = action_braced_code; %token BRACED_CODE "{...}" -%type STRING string_content - BRACED_CODE code_content action - PROLOGUE EPILOGUE -%type TYPE +%type STRING string_content + "%destructor {...}" + "%initial-action {...}" + "%lex-param {...}" + "%parse-param {...}" + "%printer {...}" + "%union {...}" + PROLOGUE EPILOGUE +%printer { fprintf (stderr, "\"%s\"", $$); } + STRING string_content +%printer { fprintf (stderr, "{\n%s\n}", $$); } + "%destructor {...}" + "%initial-action {...}" + "%lex-param {...}" + "%parse-param {...}" + "%printer {...}" + "%union {...}" + PROLOGUE EPILOGUE +%type TYPE +%printer { fprintf (stderr, "<%s>", $$); } TYPE %type INT -%type ID ID_COLON symbol string_as_id +%printer { fprintf (stderr, "%d", $$); } INT +%type ID symbol string_as_id +%printer { fprintf (stderr, "%s", $$->tag); } ID symbol string_as_id +%type ID_COLON +%printer { fprintf (stderr, "%s:", $$->tag); } ID_COLON %type precedence_declarator %type symbols.1 %% @@ -168,25 +216,41 @@ declarations: declaration: grammar_declaration | PROLOGUE { prologue_augment ($1, @1); } -| "%debug" { debug_flag = 1; } +| "%debug" { debug_flag = true; } +| "%define" string_content + { + static char one[] = "1"; + muscle_insert ($2, one); + } | "%define" string_content string_content { muscle_insert ($2, $3); } -| "%defines" { defines_flag = 1; } -| "%error-verbose" { error_verbose = 1; } -| "%expect" INT { expected_conflicts = $2; } +| "%defines" { defines_flag = true; } +| "%error-verbose" { error_verbose = true; } +| "%expect" INT { expected_sr_conflicts = $2; } +| "%expect-rr" INT { expected_rr_conflicts = $2; } | "%file-prefix" "=" string_content { spec_file_prefix = $3; } -| "%glr-parser" { glr_parser = 1; } -| "%lex-param" code_content { add_param ("lex_param", $2, @2); } -| "%locations" { locations_flag = 1; } +| "%glr-parser" + { + nondeterministic_parser = true; + glr_parser = true; + } +| "%initial-action {...}" + { + muscle_code_grow ("initial_action", $1, @1); + } +| "%lex-param {...}" { add_param ("lex_param", $1, @1); } +| "%locations" { locations_flag = true; } | "%name-prefix" "=" string_content { spec_name_prefix = $3; } -| "%no-lines" { no_lines_flag = 1; } +| "%no-lines" { no_lines_flag = true; } +| "%nondeterministic-parser" { nondeterministic_parser = true; } | "%output" "=" string_content { spec_outfile = $3; } -| "%parse-param" code_content { add_param ("parse_param", $2, @2); } -| "%pure-parser" { pure_parser = 1; } +| "%parse-param {...}" { add_param ("parse_param", $1, @1); } +| "%pure-parser" { pure_parser = true; } +| "%require" string_content { version_check (&@2, $2); } | "%skeleton" string_content { skeleton = $2; } -| "%token-table" { token_table_flag = 1; } -| "%verbose" { report_flag = 1; } -| "%yacc" { yacc_flag = 1; } -| ";" +| "%token-table" { token_table_flag = true; } +| "%verbose" { report_flag = report_states; } +| "%yacc" { yacc_flag = true; } +| /*FIXME: Err? What is this horror doing here? */ ";" ; grammar_declaration: @@ -196,31 +260,43 @@ grammar_declaration: { grammar_start_symbol_set ($2, @2); } -| "%union" BRACED_CODE +| "%union {...}" { - typed = 1; - MUSCLE_INSERT_INT ("stype_line", @2.start.line); - muscle_insert ("stype", $2); + char const *body = $1; + + if (typed) + { + /* Concatenate the union bodies, turning the first one's + trailing '}' into '\n', and omitting the second one's '{'. */ + char *code = muscle_find ("stype"); + code[strlen (code) - 1] = '\n'; + body++; + } + + typed = true; + muscle_code_grow ("stype", body, @1); + } +| "%destructor {...}" symbols.1 + { + symbol_list *list; + for (list = $2; list; list = list->next) + symbol_destructor_set (list->sym, $1, @1); + symbol_list_free ($2); } -| "%destructor" - { current_braced_code = destructor_braced_code; } - BRACED_CODE symbols.1 +| "%printer {...}" 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; + symbol_list *list; + for (list = $2; list; list = list->next) + symbol_printer_set (list->sym, $1, @1); + symbol_list_free ($2); } -| "%printer" - { current_braced_code = printer_braced_code; } - BRACED_CODE symbols.1 +| "%default-prec" { - 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; + default_prec = true; + } +| "%no-default-prec" + { + default_prec = false; } ; @@ -237,7 +313,7 @@ symbol_declaration: } | "%type" TYPE symbols.1 { - symbol_list_t *list; + symbol_list *list; for (list = $3; list; list = list->next) symbol_type_set (list->sym, $2, @2); symbol_list_free ($3); @@ -247,7 +323,7 @@ symbol_declaration: precedence_declaration: precedence_declarator type.opt symbols.1 { - symbol_list_t *list; + symbol_list *list; ++current_prec; for (list = $3; list; list = list->next) { @@ -285,24 +361,24 @@ symbol_def: } | ID { - symbol_class_set ($1, current_class, @1); + symbol_class_set ($1, current_class, @1, true); symbol_type_set ($1, current_type, @1); } | ID INT { - symbol_class_set ($1, current_class, @1); + symbol_class_set ($1, current_class, @1, true); 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, @1); + symbol_class_set ($1, current_class, @1, true); symbol_type_set ($1, current_type, @1); symbol_make_alias ($1, $2, @$); } | ID INT string_as_id { - symbol_class_set ($1, current_class, @1); + symbol_class_set ($1, current_class, @1, true); symbol_type_set ($1, current_type, @1); symbol_user_token_number_set ($1, $2, @2); symbol_make_alias ($1, $3, @$); @@ -329,16 +405,11 @@ grammar: body of the grammar. */ rules_or_grammar_declaration: rules -| grammar_declaration - { - if (yacc_flag) - complain_at (@$, _("POSIX forbids declarations in the grammar")); - } +| grammar_declaration ";" | error ";" { yyerrok; } -| ";" ; rules: @@ -346,17 +417,17 @@ rules: ; rhses.1: - rhs { grammar_rule_end (@1); } -| rhses.1 "|" rhs { grammar_rule_end (@3); } + rhs { grammar_current_rule_end (@1); } +| rhses.1 "|" rhs { grammar_current_rule_end (@3); } +| rhses.1 ";" ; rhs: /* Nothing. */ - { grammar_rule_begin (current_lhs, current_lhs_location); } + { grammar_current_rule_begin (current_lhs, current_lhs_location); } | rhs symbol { grammar_current_rule_symbol_append ($2, @2); } | rhs action - { grammar_current_rule_action_append ($2, @2); } | rhs "%prec" symbol { grammar_current_rule_prec_set ($3, @3); } | rhs "%dprec" INT @@ -370,43 +441,44 @@ symbol: | string_as_id { $$ = $1; } ; +/* Handle the semantics of an action specially, with a mid-rule + action, so that grammar_current_rule_action_append is invoked + immediately after the braced code is read by the scanner. + + This implementation relies on the LALR(1) parsing algorithm. + If grammar_current_rule_action_append were executed in a normal + action for this rule, then when the input grammar contains two + successive actions, the scanner would have to read both actions + before reducing this rule. That wouldn't work, since the scanner + relies on all preceding input actions being processed by + grammar_current_rule_action_append before it scans the next + action. */ action: + { grammar_current_rule_action_append (last_string, last_braced_code_loc); } BRACED_CODE - { $$ = $1; } ; -/* A string used as an ID: we have to keep the quotes. */ +/* A string used as an ID: quote it. */ string_as_id: STRING { - $$ = symbol_get ($1, @1); - symbol_class_set ($$, token_sym, @1); + $$ = symbol_get (quotearg_style (c_quoting_style, $1), @1); + symbol_class_set ($$, token_sym, @1, false); } ; -/* A string used for its contents. Strip the quotes. */ +/* A string used for its contents. Don't quote it. */ string_content: STRING - { - $$ = $1 + 1; - $$[strlen ($$) - 1] = '\0'; - }; - - -/* A BRACED_CODE used for its contents. Strip the braces. */ -code_content: - BRACED_CODE - { - $$ = $1 + 1; - $$[strlen ($$) - 1] = '\0'; - }; + { $$ = $1; } +; epilogue.opt: /* Nothing. */ | "%%" EPILOGUE { - epilogue_augment ($2, @2); + muscle_code_grow ("epilogue", $2, @2); scanner_last_string_free (); } ; @@ -423,24 +495,24 @@ static YYLTYPE lloc_default (YYLTYPE const *rhs, int n) { int i; - int j; - YYLTYPE r; - r.start = r.end = rhs[n].end; + YYLTYPE loc; + + /* SGI MIPSpro 7.4.1m miscompiles "loc.start = loc.end = rhs[n].end;". + The bug is fixed in 7.4.2m, but play it safe for now. */ + loc.start = rhs[n].end; + loc.end = rhs[n].end; + /* Ignore empty nonterminals the start of the 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)) { - r.start = rhs[i].start; - - for (j = n; i < j; j--) - if (! equal_boundaries (rhs[j].start, rhs[j].end)) - break; - r.end = rhs[j].end; - + loc.start = rhs[i].start; break; } - return r; + return loc; } @@ -448,21 +520,31 @@ lloc_default (YYLTYPE const *rhs, int n) declaration DECL and location LOC. */ static void -add_param (char const *type, char const *decl, location_t loc) +add_param (char const *type, char *decl, location loc) { - static char const alphanum[] = - "0123456789" + static char const alphanum[26 + 26 + 1 + 10] = "abcdefghijklmnopqrstuvwxyz" "ABCDEFGHIJKLMNOPQRSTUVWXYZ" - "_"; - char const *alpha = alphanum + 10; + "_" + "0123456789"; char const *name_start = NULL; - char const *p; + char *p; - for (p = decl; *p; p++) - if ((p == decl || ! strchr (alphanum, p[-1])) && strchr (alpha, p[0])) + /* 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")); else @@ -471,7 +553,7 @@ add_param (char const *type, char const *decl, location_t loc) size_t name_len; for (name_len = 1; - name_start[name_len] && strchr (alphanum, name_start[name_len]); + memchr (alphanum, name_start[name_len], sizeof alphanum); name_len++) continue; @@ -485,46 +567,25 @@ add_param (char const *type, char const *decl, location_t loc) scanner_last_string_free (); } -/*------------------------------------------------------------------. -| When debugging the parser, display tokens' locations and values. | -`------------------------------------------------------------------*/ - static void -print_token_value (FILE *file, int type, YYSTYPE const *value) +version_check (location const *loc, char const *version) { - fputc (' ', file); - switch (type) + if (strverscmp (version, PACKAGE_VERSION) > 0) { - case ID: - fprintf (file, " = %s", value->symbol->tag); - break; - - case INT: - fprintf (file, " = %d", value->integer); - break; - - case STRING: - fprintf (file, " = \"%s\"", value->string); - break; - - case TYPE: - fprintf (file, " = <%s>", value->struniq); - break; - - case BRACED_CODE: - case PROLOGUE: - case EPILOGUE: - fprintf (file, " = {{ %s }}", value->string); - break; - - default: - fprintf (file, "unknown token type"); - break; + complain_at (*loc, "require bison %s, but have %s", + version, PACKAGE_VERSION); + exit (63); } } static void -gram_error (location_t const *loc, char const *msg) +gram_error (location const *loc, char const *msg) { complain_at (*loc, "%s", msg); } + +char const * +token_name (int type) +{ + return yytname[YYTRANSLATE (type)]; +}