X-Git-Url: http://drtracing.org/?a=blobdiff_plain;f=gdb%2Fc-exp.y;h=cda91774b16291b5ea35a72c18e031303ab8cca4;hb=6aba47ca06d9150c6196a374b745c2711b46e045;hp=0e7d39ac6c2f69b6ee2b3c1585005ef3a4dbf7f5;hpb=f843c95fc3d55bb449ed9e681848808c9f637ee4;p=deliverable%2Fbinutils-gdb.git diff --git a/gdb/c-exp.y b/gdb/c-exp.y index 0e7d39ac6c..cda91774b1 100644 --- a/gdb/c-exp.y +++ b/gdb/c-exp.y @@ -1,5 +1,6 @@ /* YACC parser for C expressions, for GDB. - Copyright (C) 1986, 1989, 1990, 1991 Free Software Foundation, Inc. + Copyright (C) 1986, 1989, 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1997, + 1998, 1999, 2000, 2003, 2004, 2006, 2007 Free Software Foundation, Inc. This file is part of GDB. @@ -15,7 +16,8 @@ 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., 675 Mass Ave, Cambridge, MA 02139, USA. */ +Foundation, Inc., 51 Franklin Street, Fifth Floor, +Boston, MA 02110-1301, USA. */ /* Parse a C expression from text in a string, and return the result as a struct expression pointer. @@ -37,11 +39,19 @@ Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ %{ #include "defs.h" +#include "gdb_string.h" +#include #include "expression.h" -#include "parser-defs.h" #include "value.h" +#include "parser-defs.h" #include "language.h" #include "c-lang.h" +#include "bfd.h" /* Required by objfiles.h. */ +#include "symfile.h" /* Required by objfiles.h. */ +#include "objfiles.h" /* For have_full_symbols and have_partial_symbols */ +#include "charset.h" +#include "block.h" +#include "cp-support.h" /* Remap normal yacc parser interface names (yyparse, yylex, yyerror, etc), as well as gratuitiously global symbol names, so we can have multiple @@ -79,19 +89,29 @@ Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ #define yylloc c_lloc #define yyreds c_reds /* With YYDEBUG defined */ #define yytoks c_toks /* With YYDEBUG defined */ +#define yyname c_name /* With YYDEBUG defined */ +#define yyrule c_rule /* With YYDEBUG defined */ +#define yylhs c_yylhs +#define yylen c_yylen +#define yydefred c_yydefred +#define yydgoto c_yydgoto +#define yysindex c_yysindex +#define yyrindex c_yyrindex +#define yygindex c_yygindex +#define yytable c_yytable +#define yycheck c_yycheck #ifndef YYDEBUG -#define YYDEBUG 0 /* Default to no yydebug support */ +#define YYDEBUG 1 /* Default to yydebug support */ #endif -int -yyparse PARAMS ((void)); +#define YYFPRINTF parser_fprintf -static int -yylex PARAMS ((void)); +int yyparse (void); -void -yyerror PARAMS ((char *)); +static int yylex (void); + +void yyerror (char *); %} @@ -105,8 +125,11 @@ yyerror PARAMS ((char *)); struct { LONGEST val; struct type *type; - } typed_val; - double dval; + } typed_val_int; + struct { + DOUBLEST dval; + struct type *type; + } typed_val_float; struct symbol *sym; struct type *tval; struct stoken sval; @@ -123,13 +146,12 @@ yyerror PARAMS ((char *)); %{ /* YYSTYPE gets defined by %union */ -static int -parse_number PARAMS ((char *, int, int, YYSTYPE *)); +static int parse_number (char *, int, int, YYSTYPE *); %} %type exp exp1 type_exp start variable qualified_name lcurly %type rcurly -%type type typebase +%type type typebase qualified_type %type nonempty_typelist /* %type block */ @@ -138,8 +160,8 @@ parse_number PARAMS ((char *, int, int, YYSTYPE *)); %type ptype %type array_mod -%token INT -%token FLOAT +%token INT +%token FLOAT /* Both NAME and TYPENAME tokens represent symbols in the input, and both convey their data as strings. @@ -169,15 +191,16 @@ parse_number PARAMS ((char *, int, int, YYSTYPE *)); /* Special type cases, put in to allow the parser to distinguish different legal basetypes. */ -%token SIGNED_KEYWORD LONG SHORT INT_KEYWORD CONST_KEYWORD VOLATILE_KEYWORD -%token LAST REGNAME +%token SIGNED_KEYWORD LONG SHORT INT_KEYWORD CONST_KEYWORD VOLATILE_KEYWORD DOUBLE_KEYWORD -%token VARIABLE +%token VARIABLE %token ASSIGN_MODIFY /* C++ */ -%token THIS +%token TRUEKEYWORD +%token FALSEKEYWORD + %left ',' %left ABOVE_COMMA @@ -197,6 +220,7 @@ parse_number PARAMS ((char *, int, int, YYSTYPE *)); %right UNARY INCREMENT DECREMENT %right ARROW '.' '[' '(' %token BLOCKNAME +%token FILENAME %type block %left COLONCOLON @@ -222,14 +246,20 @@ exp1 : exp /* Expressions, not including the comma operator. */ exp : '*' exp %prec UNARY { write_exp_elt_opcode (UNOP_IND); } + ; exp : '&' exp %prec UNARY { write_exp_elt_opcode (UNOP_ADDR); } + ; exp : '-' exp %prec UNARY { write_exp_elt_opcode (UNOP_NEG); } ; +exp : '+' exp %prec UNARY + { write_exp_elt_opcode (UNOP_PLUS); } + ; + exp : '!' exp %prec UNARY { write_exp_elt_opcode (UNOP_LOGICAL_NOT); } ; @@ -271,6 +301,7 @@ exp : exp ARROW qualified_name write_exp_elt_opcode (UNOP_ADDR); write_exp_elt_opcode (STRUCTOP_MPTR); } ; + exp : exp ARROW '*' exp { write_exp_elt_opcode (STRUCTOP_MPTR); } ; @@ -451,8 +482,8 @@ exp : NAME_OR_INT { YYSTYPE val; parse_number ($1.stoken.ptr, $1.stoken.length, 0, &val); write_exp_elt_opcode (OP_LONG); - write_exp_elt_type (val.typed_val.type); - write_exp_elt_longcst ((LONGEST)val.typed_val.val); + write_exp_elt_type (val.typed_val_int.type); + write_exp_elt_longcst ((LONGEST)val.typed_val_int.val); write_exp_elt_opcode (OP_LONG); } ; @@ -460,35 +491,22 @@ exp : NAME_OR_INT exp : FLOAT { write_exp_elt_opcode (OP_DOUBLE); - write_exp_elt_type (builtin_type_double); - write_exp_elt_dblcst ($1); + write_exp_elt_type ($1.type); + write_exp_elt_dblcst ($1.dval); write_exp_elt_opcode (OP_DOUBLE); } ; exp : variable ; -exp : LAST - { write_exp_elt_opcode (OP_LAST); - write_exp_elt_longcst ((LONGEST) $1); - write_exp_elt_opcode (OP_LAST); } - ; - -exp : REGNAME - { write_exp_elt_opcode (OP_REGISTER); - write_exp_elt_longcst ((LONGEST) $1); - write_exp_elt_opcode (OP_REGISTER); } - ; - exp : VARIABLE - { write_exp_elt_opcode (OP_INTERNALVAR); - write_exp_elt_intern ($1); - write_exp_elt_opcode (OP_INTERNALVAR); } + /* Already written by write_dollar_variable. */ ; exp : SIZEOF '(' type ')' %prec UNARY { write_exp_elt_opcode (OP_LONG); - write_exp_elt_type (builtin_type_int); + write_exp_elt_type (builtin_type (current_gdbarch)->builtin_int); + CHECK_TYPEDEF ($3); write_exp_elt_longcst ((LONGEST) TYPE_LENGTH ($3)); write_exp_elt_opcode (OP_LONG); } ; @@ -503,12 +521,12 @@ exp : STRING while (count-- > 0) { write_exp_elt_opcode (OP_LONG); - write_exp_elt_type (builtin_type_char); + write_exp_elt_type (builtin_type (current_gdbarch)->builtin_char); write_exp_elt_longcst ((LONGEST)(*sp++)); write_exp_elt_opcode (OP_LONG); } write_exp_elt_opcode (OP_LONG); - write_exp_elt_type (builtin_type_char); + write_exp_elt_type (builtin_type (current_gdbarch)->builtin_char); write_exp_elt_longcst ((LONGEST)'\0'); write_exp_elt_opcode (OP_LONG); write_exp_elt_opcode (OP_ARRAY); @@ -518,35 +536,40 @@ exp : STRING ; /* C++. */ -exp : THIS - { write_exp_elt_opcode (OP_THIS); - write_exp_elt_opcode (OP_THIS); } +exp : TRUEKEYWORD + { write_exp_elt_opcode (OP_LONG); + write_exp_elt_type (builtin_type (current_gdbarch)->builtin_bool); + write_exp_elt_longcst ((LONGEST) 1); + write_exp_elt_opcode (OP_LONG); } + ; + +exp : FALSEKEYWORD + { write_exp_elt_opcode (OP_LONG); + write_exp_elt_type (builtin_type (current_gdbarch)->builtin_bool); + write_exp_elt_longcst ((LONGEST) 0); + write_exp_elt_opcode (OP_LONG); } ; /* end of C++. */ block : BLOCKNAME { - if ($1.sym != 0) - $$ = SYMBOL_BLOCK_VALUE ($1.sym); + if ($1.sym) + $$ = SYMBOL_BLOCK_VALUE ($1.sym); else - { - struct symtab *tem = - lookup_symtab (copy_name ($1.stoken)); - if (tem) - $$ = BLOCKVECTOR_BLOCK - (BLOCKVECTOR (tem), STATIC_BLOCK); - else - error ("No file or function \"%s\".", - copy_name ($1.stoken)); - } + error ("No file or function \"%s\".", + copy_name ($1.stoken)); + } + | FILENAME + { + $$ = $1; } ; block : block COLONCOLON name { struct symbol *tem = lookup_symbol (copy_name ($3), $1, - VAR_NAMESPACE, (int *) NULL, + VAR_DOMAIN, (int *) NULL, (struct symtab **) NULL); if (!tem || SYMBOL_CLASS (tem) != LOC_BLOCK) error ("No function \"%s\" in specified context.", @@ -557,7 +580,7 @@ block : block COLONCOLON name variable: block COLONCOLON name { struct symbol *sym; sym = lookup_symbol (copy_name ($3), $1, - VAR_NAMESPACE, (int *) NULL, + VAR_DOMAIN, (int *) NULL, (struct symtab **) NULL); if (sym == 0) error ("No symbol \"%s\" in specified context.", @@ -574,7 +597,8 @@ qualified_name: typebase COLONCOLON name { struct type *type = $1; if (TYPE_CODE (type) != TYPE_CODE_STRUCT - && TYPE_CODE (type) != TYPE_CODE_UNION) + && TYPE_CODE (type) != TYPE_CODE_UNION + && TYPE_CODE (type) != TYPE_CODE_NAMESPACE) error ("`%s' is not defined as an aggregate type.", TYPE_NAME (type)); @@ -588,19 +612,19 @@ qualified_name: typebase COLONCOLON name struct type *type = $1; struct stoken tmp_token; if (TYPE_CODE (type) != TYPE_CODE_STRUCT - && TYPE_CODE (type) != TYPE_CODE_UNION) + && TYPE_CODE (type) != TYPE_CODE_UNION + && TYPE_CODE (type) != TYPE_CODE_NAMESPACE) error ("`%s' is not defined as an aggregate type.", TYPE_NAME (type)); - if (!STREQ (type_name_no_tag (type), $4.ptr)) - error ("invalid destructor `%s::~%s'", - type_name_no_tag (type), $4.ptr); - tmp_token.ptr = (char*) alloca ($4.length + 2); tmp_token.length = $4.length + 1; tmp_token.ptr[0] = '~'; memcpy (tmp_token.ptr+1, $4.ptr, $4.length); tmp_token.ptr[tmp_token.length] = 0; + + /* Check for valid destructor name. */ + destructor_name_p (tmp_token.ptr, type); write_exp_elt_opcode (OP_SCOPE); write_exp_elt_type (type); write_exp_string (tmp_token); @@ -617,7 +641,7 @@ variable: qualified_name sym = lookup_symbol (name, (const struct block *) NULL, - VAR_NAMESPACE, (int *) NULL, + VAR_DOMAIN, (int *) NULL, (struct symtab **) NULL); if (sym) { @@ -628,23 +652,12 @@ variable: qualified_name break; } - msymbol = lookup_minimal_symbol (name, - (struct objfile *) NULL); + msymbol = lookup_minimal_symbol (name, NULL, NULL); if (msymbol != NULL) { - write_exp_elt_opcode (OP_LONG); - write_exp_elt_type (builtin_type_long); - write_exp_elt_longcst ((LONGEST) SYMBOL_VALUE_ADDRESS (msymbol)); - write_exp_elt_opcode (OP_LONG); - write_exp_elt_opcode (UNOP_MEMVAL); - if (msymbol -> type == mst_data || - msymbol -> type == mst_bss) - write_exp_elt_type (builtin_type_int); - else if (msymbol -> type == mst_text) - write_exp_elt_type (lookup_function_type (builtin_type_int)); - else - write_exp_elt_type (builtin_type_char); - write_exp_elt_opcode (UNOP_MEMVAL); + write_exp_msymbol (msymbol, + lookup_function_type (builtin_type (current_gdbarch)->builtin_int), + builtin_type (current_gdbarch)->builtin_int); } else if (!have_full_symbols () && !have_partial_symbols ()) @@ -692,25 +705,15 @@ variable: name_not_typename else { struct minimal_symbol *msymbol; - register char *arg = copy_name ($1.stoken); + char *arg = copy_name ($1.stoken); - msymbol = lookup_minimal_symbol (arg, - (struct objfile *) NULL); + msymbol = + lookup_minimal_symbol (arg, NULL, NULL); if (msymbol != NULL) { - write_exp_elt_opcode (OP_LONG); - write_exp_elt_type (builtin_type_long); - write_exp_elt_longcst ((LONGEST) SYMBOL_VALUE_ADDRESS (msymbol)); - write_exp_elt_opcode (OP_LONG); - write_exp_elt_opcode (UNOP_MEMVAL); - if (msymbol -> type == mst_data || - msymbol -> type == mst_bss) - write_exp_elt_type (builtin_type_int); - else if (msymbol -> type == mst_text) - write_exp_elt_type (lookup_function_type (builtin_type_int)); - else - write_exp_elt_type (builtin_type_char); - write_exp_elt_opcode (UNOP_MEMVAL); + write_exp_msymbol (msymbol, + lookup_function_type (builtin_type (current_gdbarch)->builtin_int), + builtin_type (current_gdbarch)->builtin_int); } else if (!have_full_symbols () && !have_partial_symbols ()) error ("No symbol table is loaded. Use the \"file\" command."); @@ -721,24 +724,26 @@ variable: name_not_typename } ; +space_identifier : '@' NAME + { push_type_address_space (copy_name ($2.stoken)); + push_type (tp_space_identifier); + } + ; -/* shift/reduce conflict: "typebase ." and the token is '('. (Shows up - twice, once where qualified_name is a possibility and once where - it is not). */ -/* shift/reduce conflict: "typebase CONST_KEYWORD ." and the token is '('. */ -/* shift/reduce conflict: "typebase VOLATILE_KEYWORD ." and the token is - '('. */ -ptype : typebase - /* "const" and "volatile" are curently ignored. A type qualifier - before the type is currently handled in the typebase rule. */ - | typebase CONST_KEYWORD - | typebase VOLATILE_KEYWORD - | typebase abs_decl - { $$ = follow_types ($1); } - | typebase CONST_KEYWORD abs_decl - { $$ = follow_types ($1); } - | typebase VOLATILE_KEYWORD abs_decl - { $$ = follow_types ($1); } +const_or_volatile: const_or_volatile_noopt + | + ; + +cv_with_space_id : const_or_volatile space_identifier const_or_volatile + ; + +const_or_volatile_or_space_identifier_noopt: cv_with_space_id + | const_or_volatile_noopt + ; + +const_or_volatile_or_space_identifier: + const_or_volatile_or_space_identifier_noopt + | ; abs_decl: '*' @@ -766,9 +771,6 @@ direct_abs_decl: '(' abs_decl ')' $$ = 0; } - /* shift/reduce conflict. "direct_abs_decl . func_mod", and the token - is '('. */ - | direct_abs_decl func_mod { push_type (tp_function); } | func_mod @@ -784,50 +786,79 @@ array_mod: '[' ']' func_mod: '(' ')' { $$ = 0; } | '(' nonempty_typelist ')' - { free ((PTR)$2); $$ = 0; } + { free ($2); $$ = 0; } ; -/* shift/reduce conflict: "type '(' typebase COLONCOLON '*' ')' ." and the - token is '('. */ +/* We used to try to recognize pointer to member types here, but + that didn't work (shift/reduce conflicts meant that these rules never + got executed). The problem is that + int (foo::bar::baz::bizzle) + is a function type but + int (foo::bar::baz::bizzle::*) + is a pointer to member type. Stroustrup loses again! */ + type : ptype - | typebase COLONCOLON '*' - { $$ = lookup_member_type (builtin_type_int, $1); } - | type '(' typebase COLONCOLON '*' ')' - { $$ = lookup_member_type ($1, $3); } - | type '(' typebase COLONCOLON '*' ')' '(' ')' - { $$ = lookup_member_type - (lookup_function_type ($1), $3); } - | type '(' typebase COLONCOLON '*' ')' '(' nonempty_typelist ')' - { $$ = lookup_member_type - (lookup_function_type ($1), $3); - free ((PTR)$8); } ; typebase /* Implements (approximately): (type-qualifier)* type-specifier */ : TYPENAME { $$ = $1.type; } | INT_KEYWORD - { $$ = builtin_type_int; } + { $$ = builtin_type (current_gdbarch)->builtin_int; } | LONG - { $$ = builtin_type_long; } + { $$ = builtin_type (current_gdbarch)->builtin_long; } | SHORT - { $$ = builtin_type_short; } + { $$ = builtin_type (current_gdbarch)->builtin_short; } | LONG INT_KEYWORD - { $$ = builtin_type_long; } + { $$ = builtin_type (current_gdbarch)->builtin_long; } + | LONG SIGNED_KEYWORD INT_KEYWORD + { $$ = builtin_type (current_gdbarch)->builtin_long; } + | LONG SIGNED_KEYWORD + { $$ = builtin_type (current_gdbarch)->builtin_long; } + | SIGNED_KEYWORD LONG INT_KEYWORD + { $$ = builtin_type (current_gdbarch)->builtin_long; } | UNSIGNED LONG INT_KEYWORD - { $$ = builtin_type_unsigned_long; } + { $$ = builtin_type (current_gdbarch)->builtin_unsigned_long; } + | LONG UNSIGNED INT_KEYWORD + { $$ = builtin_type (current_gdbarch)->builtin_unsigned_long; } + | LONG UNSIGNED + { $$ = builtin_type (current_gdbarch)->builtin_unsigned_long; } | LONG LONG - { $$ = builtin_type_long_long; } + { $$ = builtin_type (current_gdbarch)->builtin_long_long; } | LONG LONG INT_KEYWORD - { $$ = builtin_type_long_long; } + { $$ = builtin_type (current_gdbarch)->builtin_long_long; } + | LONG LONG SIGNED_KEYWORD INT_KEYWORD + { $$ = builtin_type (current_gdbarch)->builtin_long_long; } + | LONG LONG SIGNED_KEYWORD + { $$ = builtin_type (current_gdbarch)->builtin_long_long; } + | SIGNED_KEYWORD LONG LONG + { $$ = builtin_type (current_gdbarch)->builtin_long_long; } + | SIGNED_KEYWORD LONG LONG INT_KEYWORD + { $$ = builtin_type (current_gdbarch)->builtin_long_long; } | UNSIGNED LONG LONG - { $$ = builtin_type_unsigned_long_long; } + { $$ = builtin_type (current_gdbarch)->builtin_unsigned_long_long; } | UNSIGNED LONG LONG INT_KEYWORD - { $$ = builtin_type_unsigned_long_long; } + { $$ = builtin_type (current_gdbarch)->builtin_unsigned_long_long; } + | LONG LONG UNSIGNED + { $$ = builtin_type (current_gdbarch)->builtin_unsigned_long_long; } + | LONG LONG UNSIGNED INT_KEYWORD + { $$ = builtin_type (current_gdbarch)->builtin_unsigned_long_long; } | SHORT INT_KEYWORD - { $$ = builtin_type_short; } + { $$ = builtin_type (current_gdbarch)->builtin_short; } + | SHORT SIGNED_KEYWORD INT_KEYWORD + { $$ = builtin_type (current_gdbarch)->builtin_short; } + | SHORT SIGNED_KEYWORD + { $$ = builtin_type (current_gdbarch)->builtin_short; } | UNSIGNED SHORT INT_KEYWORD - { $$ = builtin_type_unsigned_short; } + { $$ = builtin_type (current_gdbarch)->builtin_unsigned_short; } + | SHORT UNSIGNED + { $$ = builtin_type (current_gdbarch)->builtin_unsigned_short; } + | SHORT UNSIGNED INT_KEYWORD + { $$ = builtin_type (current_gdbarch)->builtin_unsigned_short; } + | DOUBLE_KEYWORD + { $$ = builtin_type (current_gdbarch)->builtin_double; } + | LONG DOUBLE_KEYWORD + { $$ = builtin_type (current_gdbarch)->builtin_long_double; } | STRUCT name { $$ = lookup_struct (copy_name ($2), expression_context_block); } @@ -843,20 +874,93 @@ typebase /* Implements (approximately): (type-qualifier)* type-specifier */ | UNSIGNED typename { $$ = lookup_unsigned_typename (TYPE_NAME($2.type)); } | UNSIGNED - { $$ = builtin_type_unsigned_int; } + { $$ = builtin_type (current_gdbarch)->builtin_unsigned_int; } | SIGNED_KEYWORD typename { $$ = lookup_signed_typename (TYPE_NAME($2.type)); } | SIGNED_KEYWORD - { $$ = builtin_type_int; } + { $$ = builtin_type (current_gdbarch)->builtin_int; } + /* It appears that this rule for templates is never + reduced; template recognition happens by lookahead + in the token processing code in yylex. */ | TEMPLATE name '<' type '>' { $$ = lookup_template_type(copy_name($2), $4, expression_context_block); } - /* "const" and "volatile" are curently ignored. A type qualifier - after the type is handled in the ptype rule. I think these could - be too. */ - | CONST_KEYWORD typebase { $$ = $2; } - | VOLATILE_KEYWORD typebase { $$ = $2; } + | const_or_volatile_or_space_identifier_noopt typebase + { $$ = follow_types ($2); } + | typebase const_or_volatile_or_space_identifier_noopt + { $$ = follow_types ($1); } + | qualified_type + ; + +/* FIXME: carlton/2003-09-25: This next bit leads to lots of + reduce-reduce conflicts, because the parser doesn't know whether or + not to use qualified_name or qualified_type: the rules are + identical. If the parser is parsing 'A::B::x', then, when it sees + the second '::', it knows that the expression to the left of it has + to be a type, so it uses qualified_type. But if it is parsing just + 'A::B', then it doesn't have any way of knowing which rule to use, + so there's a reduce-reduce conflict; it picks qualified_name, since + that occurs earlier in this file than qualified_type. + + There's no good way to fix this with the grammar as it stands; as + far as I can tell, some of the problems arise from ambiguities that + GDB introduces ('start' can be either an expression or a type), but + some of it is inherent to the nature of C++ (you want to treat the + input "(FOO)" fairly differently depending on whether FOO is an + expression or a type, and if FOO is a complex expression, this can + be hard to determine at the right time). Fortunately, it works + pretty well in most cases. For example, if you do 'ptype A::B', + where A::B is a nested type, then the parser will mistakenly + misidentify it as an expression; but evaluate_subexp will get + called with 'noside' set to EVAL_AVOID_SIDE_EFFECTS, and everything + will work out anyways. But there are situations where the parser + will get confused: the most common one that I've run into is when + you want to do + + print *((A::B *) x)" + + where the parser doesn't realize that A::B has to be a type until + it hits the first right paren, at which point it's too late. (The + workaround is to type "print *(('A::B' *) x)" instead.) (And + another solution is to fix our symbol-handling code so that the + user never wants to type something like that in the first place, + because we get all the types right without the user's help!) + + Perhaps we could fix this by making the lexer smarter. Some of + this functionality used to be in the lexer, but in a way that + worked even less well than the current solution: that attempt + involved having the parser sometimes handle '::' and having the + lexer sometimes handle it, and without a clear division of + responsibility, it quickly degenerated into a big mess. Probably + the eventual correct solution will give more of a role to the lexer + (ideally via code that is shared between the lexer and + decode_line_1), but I'm not holding my breath waiting for somebody + to get around to cleaning this up... */ + +qualified_type: typebase COLONCOLON name + { + struct type *type = $1; + struct type *new_type; + char *ncopy = alloca ($3.length + 1); + + memcpy (ncopy, $3.ptr, $3.length); + ncopy[$3.length] = '\0'; + + if (TYPE_CODE (type) != TYPE_CODE_STRUCT + && TYPE_CODE (type) != TYPE_CODE_UNION + && TYPE_CODE (type) != TYPE_CODE_NAMESPACE) + error ("`%s' is not defined as an aggregate type.", + TYPE_NAME (type)); + + new_type = cp_lookup_nested_type (type, ncopy, + expression_context_block); + if (new_type == NULL) + error ("No type \"%s\" within class or namespace \"%s\".", + ncopy, TYPE_NAME (type)); + + $$ = new_type; + } ; typename: TYPENAME @@ -864,19 +968,19 @@ typename: TYPENAME { $$.stoken.ptr = "int"; $$.stoken.length = 3; - $$.type = builtin_type_int; + $$.type = builtin_type (current_gdbarch)->builtin_int; } | LONG { $$.stoken.ptr = "long"; $$.stoken.length = 4; - $$.type = builtin_type_long; + $$.type = builtin_type (current_gdbarch)->builtin_long; } | SHORT { $$.stoken.ptr = "short"; $$.stoken.length = 5; - $$.type = builtin_type_short; + $$.type = builtin_type (current_gdbarch)->builtin_short; } ; @@ -893,6 +997,25 @@ nonempty_typelist } ; +ptype : typebase + | ptype const_or_volatile_or_space_identifier abs_decl const_or_volatile_or_space_identifier + { $$ = follow_types ($1); } + ; + +const_and_volatile: CONST_KEYWORD VOLATILE_KEYWORD + | VOLATILE_KEYWORD CONST_KEYWORD + ; + +const_or_volatile_noopt: const_and_volatile + { push_type (tp_const); + push_type (tp_volatile); + } + | CONST_KEYWORD + { push_type (tp_const); } + | VOLATILE_KEYWORD + { push_type (tp_volatile); } + ; + name : NAME { $$ = $1.stoken; } | BLOCKNAME { $$ = $1.stoken; } | TYPENAME { $$ = $1.stoken; } @@ -920,26 +1043,62 @@ name_not_typename : NAME static int parse_number (p, len, parsed_float, putithere) - register char *p; - register int len; + char *p; + int len; int parsed_float; YYSTYPE *putithere; { - register LONGEST n = 0; - register LONGEST prevn = 0; - register int i = 0; - register int c; - register int base = input_radix; + /* FIXME: Shouldn't these be unsigned? We don't deal with negative values + here, and we do kind of silly things like cast to unsigned. */ + LONGEST n = 0; + LONGEST prevn = 0; + ULONGEST un; + + int i = 0; + int c; + int base = input_radix; int unsigned_p = 0; + + /* Number of "L" suffixes encountered. */ int long_p = 0; - unsigned LONGEST high_bit; + + /* We have found a "L" or "U" suffix. */ + int found_suffix = 0; + + ULONGEST high_bit; struct type *signed_type; struct type *unsigned_type; if (parsed_float) { /* It's a float since it contains a point or an exponent. */ - putithere->dval = atof (p); + char *s = malloc (len); + int num = 0; /* number of tokens scanned by scanf */ + char saved_char = p[len]; + + p[len] = 0; /* null-terminate the token */ + num = sscanf (p, DOUBLEST_SCAN_FORMAT "%s", + &putithere->typed_val_float.dval, s); + p[len] = saved_char; /* restore the input stream */ + + if (num == 1) + putithere->typed_val_float.type = + builtin_type (current_gdbarch)->builtin_double; + + if (num == 2 ) + { + /* See if it has any float suffix: 'f' for float, 'l' for long + double. */ + if (!strcasecmp (s, "f")) + putithere->typed_val_float.type = + builtin_type (current_gdbarch)->builtin_float; + else if (!strcasecmp (s, "l")) + putithere->typed_val_float.type = + builtin_type (current_gdbarch)->builtin_long_double; + else + return ERROR; + } + return FLOAT; } @@ -982,15 +1141,29 @@ parse_number (p, len, parsed_float, putithere) if (c != 'l' && c != 'u') n *= base; if (c >= '0' && c <= '9') - n += i = c - '0'; + { + if (found_suffix) + return ERROR; + n += i = c - '0'; + } else { if (base > 10 && c >= 'a' && c <= 'f') - n += i = c - 'a' + 10; - else if (len == 0 && c == 'l') - long_p = 1; - else if (len == 0 && c == 'u') - unsigned_p = 1; + { + if (found_suffix) + return ERROR; + n += i = c - 'a' + 10; + } + else if (c == 'l') + { + ++long_p; + found_suffix = 1; + } + else if (c == 'u') + { + unsigned_p = 1; + found_suffix = 1; + } else return ERROR; /* Char not a digit */ } @@ -998,57 +1171,81 @@ parse_number (p, len, parsed_float, putithere) return ERROR; /* Invalid digit in this base */ /* Portably test for overflow (only works for nonzero values, so make - a second check for zero). */ - if((prevn >= n) && n != 0) - unsigned_p=1; /* Try something unsigned */ - /* If range checking enabled, portably test for unsigned overflow. */ - if(RANGE_CHECK && n!=0) - { - if((unsigned_p && (unsigned)prevn >= (unsigned)n)) - range_error("Overflow on numeric constant."); - } - prevn=n; + a second check for zero). FIXME: Can't we just make n and prevn + unsigned and avoid this? */ + if (c != 'l' && c != 'u' && (prevn >= n) && n != 0) + unsigned_p = 1; /* Try something unsigned */ + + /* Portably test for unsigned overflow. + FIXME: This check is wrong; for example it doesn't find overflow + on 0x123456789 when LONGEST is 32 bits. */ + if (c != 'l' && c != 'u' && n != 0) + { + if ((unsigned_p && (ULONGEST) prevn >= (ULONGEST) n)) + error ("Numeric constant too large."); + } + prevn = n; + } + + /* An integer constant is an int, a long, or a long long. An L + suffix forces it to be long; an LL suffix forces it to be long + long. If not forced to a larger size, it gets the first type of + the above that it fits in. To figure out whether it fits, we + shift it right and see whether anything remains. Note that we + can't shift sizeof (LONGEST) * HOST_CHAR_BIT bits or more in one + operation, because many compilers will warn about such a shift + (which always produces a zero result). Sometimes TARGET_INT_BIT + or TARGET_LONG_BIT will be that big, sometimes not. To deal with + the case where it is we just always shift the value more than + once, with fewer bits each time. */ + + un = (ULONGEST)n >> 2; + if (long_p == 0 + && (un >> (TARGET_INT_BIT - 2)) == 0) + { + high_bit = ((ULONGEST)1) << (TARGET_INT_BIT-1); + + /* A large decimal (not hex or octal) constant (between INT_MAX + and UINT_MAX) is a long or unsigned long, according to ANSI, + never an unsigned int, but this code treats it as unsigned + int. This probably should be fixed. GCC gives a warning on + such constants. */ + + unsigned_type = builtin_type (current_gdbarch)->builtin_unsigned_int; + signed_type = builtin_type (current_gdbarch)->builtin_int; + } + else if (long_p <= 1 + && (un >> (TARGET_LONG_BIT - 2)) == 0) + { + high_bit = ((ULONGEST)1) << (TARGET_LONG_BIT-1); + unsigned_type = builtin_type (current_gdbarch)->builtin_unsigned_long; + signed_type = builtin_type (current_gdbarch)->builtin_long; + } + else + { + int shift; + if (sizeof (ULONGEST) * HOST_CHAR_BIT < TARGET_LONG_LONG_BIT) + /* A long long does not fit in a LONGEST. */ + shift = (sizeof (ULONGEST) * HOST_CHAR_BIT - 1); + else + shift = (TARGET_LONG_LONG_BIT - 1); + high_bit = (ULONGEST) 1 << shift; + unsigned_type = builtin_type (current_gdbarch)->builtin_unsigned_long_long; + signed_type = builtin_type (current_gdbarch)->builtin_long_long; } - - /* If the number is too big to be an int, or it's got an l suffix - then it's a long. Work out if this has to be a long by - shifting right and and seeing if anything remains, and the - target int size is different to the target long size. - - In the expression below, we could have tested - (n >> TARGET_INT_BIT) - to see if it was zero, - but too many compilers warn about that, when ints and longs - are the same size. So we shift it twice, with fewer bits - each time, for the same result. */ - - if ( (TARGET_INT_BIT != TARGET_LONG_BIT - && ((n >> 2) >> (TARGET_INT_BIT-2))) /* Avoid shift warning */ - || long_p) - { - high_bit = ((unsigned LONGEST)1) << (TARGET_LONG_BIT-1); - unsigned_type = builtin_type_unsigned_long; - signed_type = builtin_type_long; - } - else - { - high_bit = ((unsigned LONGEST)1) << (TARGET_INT_BIT-1); - unsigned_type = builtin_type_unsigned_int; - signed_type = builtin_type_int; - } - putithere->typed_val.val = n; + putithere->typed_val_int.val = n; /* If the high bit of the worked out type is set then this number has to be unsigned. */ if (unsigned_p || (n & high_bit)) { - putithere->typed_val.type = unsigned_type; + putithere->typed_val_int.type = unsigned_type; } else { - putithere->typed_val.type = signed_type; + putithere->typed_val_int.type = signed_type; } return INT; @@ -1104,13 +1301,31 @@ yylex () int tempbufindex; static char *tempbuf; static int tempbufsize; - + struct symbol * sym_class = NULL; + char * token_string = NULL; + int class_prefix = 0; + int unquoted_expr; + retry: + /* Check if this is a macro invocation that we need to expand. */ + if (! scanning_macro_expansion ()) + { + char *expanded = macro_expand_next (&lexptr, + expression_macro_lookup_func, + expression_macro_lookup_baton); + + if (expanded) + scan_macro_expansion (expanded); + } + + prev_lexptr = lexptr; + unquoted_expr = 1; + tokstart = lexptr; /* See if it is a special token of length 3. */ for (i = 0; i < sizeof tokentab3 / sizeof tokentab3[0]; i++) - if (STREQN (tokstart, tokentab3[i].operator, 3)) + if (strncmp (tokstart, tokentab3[i].operator, 3) == 0) { lexptr += 3; yylval.opcode = tokentab3[i].opcode; @@ -1119,7 +1334,7 @@ yylex () /* See if it is a special token of length 2. */ for (i = 0; i < sizeof tokentab2 / sizeof tokentab2[0]; i++) - if (STREQN (tokstart, tokentab2[i].operator, 2)) + if (strncmp (tokstart, tokentab2[i].operator, 2) == 0) { lexptr += 2; yylval.opcode = tokentab2[i].opcode; @@ -1129,7 +1344,17 @@ yylex () switch (c = *tokstart) { case 0: - return 0; + /* If we were just scanning the result of a macro expansion, + then we need to resume scanning the original text. + Otherwise, we were already scanning the original text, and + we're really done. */ + if (scanning_macro_expansion ()) + { + finished_macro_expansion (); + goto retry; + } + else + return 0; case ' ': case '\t': @@ -1145,9 +1370,20 @@ yylex () c = *lexptr++; if (c == '\\') c = parse_escape (&lexptr); + else if (c == '\'') + error ("Empty character constant."); + else if (! host_char_to_target (c, &c)) + { + int toklen = lexptr - tokstart + 1; + char *tok = alloca (toklen + 1); + memcpy (tok, tokstart, toklen); + tok[toklen] = '\0'; + error ("There is no character corresponding to %s in the target " + "character set `%s'.", tok, target_charset ()); + } - yylval.typed_val.val = c; - yylval.typed_val.type = builtin_type_char; + yylval.typed_val_int.val = c; + yylval.typed_val_int.type = builtin_type (current_gdbarch)->builtin_char; c = *lexptr++; if (c != '\'') @@ -1156,6 +1392,7 @@ yylex () if (namelen > 2) { lexptr = tokstart + namelen; + unquoted_expr = 0; if (lexptr[-1] != '\'') error ("Unmatched single quote."); namelen -= 2; @@ -1179,7 +1416,9 @@ yylex () return c; case ',': - if (comma_terminates && paren_depth == 0) + if (comma_terminates + && paren_depth == 0 + && ! scanning_macro_expansion ()) return 0; lexptr++; return c; @@ -1203,7 +1442,7 @@ yylex () { /* It's a number. */ int got_dot = 0, got_e = 0, toktype; - register char *p = tokstart; + char *p = tokstart; int hex = input_radix > 10; if (c == '0' && (p[1] == 'x' || p[1] == 'X')) @@ -1292,6 +1531,8 @@ yylex () tempbufindex = 0; do { + char *char_start_pos = tokptr; + /* Grow the static temp buffer if necessary, including allocating the first one on demand. */ if (tempbufindex + 1 >= tempbufsize) @@ -1314,7 +1555,19 @@ yylex () tempbuf[tempbufindex++] = c; break; default: - tempbuf[tempbufindex++] = *tokptr++; + c = *tokptr++; + if (! host_char_to_target (c, &c)) + { + int len = tokptr - char_start_pos; + char *copy = alloca (len + 1); + memcpy (copy, char_start_pos, len); + copy[len] = '\0'; + + error ("There is no character corresponding to `%s' " + "in the target character set `%s'.", + copy, target_charset ()); + } + tempbuf[tempbufindex++] = c; break; } } while ((*tokptr != '"') && (*tokptr != '\0')); @@ -1338,123 +1591,91 @@ yylex () namelen = 0; for (c = tokstart[namelen]; (c == '_' || c == '$' || (c >= '0' && c <= '9') - || (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z')); - c = tokstart[++namelen]) - ; + || (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z') || c == '<');) + { + /* Template parameter lists are part of the name. + FIXME: This mishandles `print $a<4&&$a>3'. */ + + if (c == '<') + { + /* Scan ahead to get rest of the template specification. Note + that we look ahead only when the '<' adjoins non-whitespace + characters; for comparison expressions, e.g. "a < b > c", + there must be spaces before the '<', etc. */ + + char * p = find_template_name_end (tokstart + namelen); + if (p) + namelen = p - tokstart; + break; + } + c = tokstart[++namelen]; + } - /* The token "if" terminates the expression and is NOT - removed from the input stream. */ - if (namelen == 2 && tokstart[0] == 'i' && tokstart[1] == 'f') + /* The token "if" terminates the expression and is NOT removed from + the input stream. It doesn't count if it appears in the + expansion of a macro. */ + if (namelen == 2 + && tokstart[0] == 'i' + && tokstart[1] == 'f' + && ! scanning_macro_expansion ()) { return 0; } lexptr += namelen; - /* Handle the tokens $digits; also $ (short for $0) and $$ (short for $$1) - and $$digits (equivalent to $<-digits> if you could type that). - Make token type LAST, and put the number (the digits) in yylval. */ - tryname: - if (*tokstart == '$') - { - register int negate = 0; - c = 1; - /* Double dollar means negate the number and add -1 as well. - Thus $$ alone means -1. */ - if (namelen >= 2 && tokstart[1] == '$') - { - negate = 1; - c = 2; - } - if (c == namelen) - { - /* Just dollars (one or two) */ - yylval.lval = - negate; - return LAST; - } - /* Is the rest of the token digits? */ - for (; c < namelen; c++) - if (!(tokstart[c] >= '0' && tokstart[c] <= '9')) - break; - if (c == namelen) - { - yylval.lval = atoi (tokstart + 1 + negate); - if (negate) - yylval.lval = - yylval.lval; - return LAST; - } - } - /* Handle tokens that refer to machine registers: - $ followed by a register name. */ - - if (*tokstart == '$') { - for (c = 0; c < NUM_REGS; c++) - if (namelen - 1 == strlen (reg_names[c]) - && STREQN (tokstart + 1, reg_names[c], namelen - 1)) - { - yylval.lval = c; - return REGNAME; - } - for (c = 0; c < num_std_regs; c++) - if (namelen - 1 == strlen (std_regs[c].name) - && STREQN (tokstart + 1, std_regs[c].name, namelen - 1)) - { - yylval.lval = std_regs[c].regnum; - return REGNAME; - } - } /* Catch specific keywords. Should be done with a data structure. */ switch (namelen) { case 8: - if (STREQN (tokstart, "unsigned", 8)) + if (strncmp (tokstart, "unsigned", 8) == 0) return UNSIGNED; if (current_language->la_language == language_cplus - && STREQN (tokstart, "template", 8)) + && strncmp (tokstart, "template", 8) == 0) return TEMPLATE; - if (STREQN (tokstart, "volatile", 8)) + if (strncmp (tokstart, "volatile", 8) == 0) return VOLATILE_KEYWORD; break; case 6: - if (STREQN (tokstart, "struct", 6)) + if (strncmp (tokstart, "struct", 6) == 0) return STRUCT; - if (STREQN (tokstart, "signed", 6)) + if (strncmp (tokstart, "signed", 6) == 0) return SIGNED_KEYWORD; - if (STREQN (tokstart, "sizeof", 6)) + if (strncmp (tokstart, "sizeof", 6) == 0) return SIZEOF; + if (strncmp (tokstart, "double", 6) == 0) + return DOUBLE_KEYWORD; break; case 5: - if (current_language->la_language == language_cplus - && STREQN (tokstart, "class", 5)) - return CLASS; - if (STREQN (tokstart, "union", 5)) + if (current_language->la_language == language_cplus) + { + if (strncmp (tokstart, "false", 5) == 0) + return FALSEKEYWORD; + if (strncmp (tokstart, "class", 5) == 0) + return CLASS; + } + if (strncmp (tokstart, "union", 5) == 0) return UNION; - if (STREQN (tokstart, "short", 5)) + if (strncmp (tokstart, "short", 5) == 0) return SHORT; - if (STREQN (tokstart, "const", 5)) + if (strncmp (tokstart, "const", 5) == 0) return CONST_KEYWORD; break; case 4: - if (STREQN (tokstart, "enum", 4)) + if (strncmp (tokstart, "enum", 4) == 0) return ENUM; - if (STREQN (tokstart, "long", 4)) + if (strncmp (tokstart, "long", 4) == 0) return LONG; - if (current_language->la_language == language_cplus - && STREQN (tokstart, "this", 4)) - { - static const char this_name[] = - { CPLUS_MARKER, 't', 'h', 'i', 's', '\0' }; - - if (lookup_symbol (this_name, expression_context_block, - VAR_NAMESPACE, (int *) NULL, - (struct symtab **) NULL)) - return THIS; - } + if (current_language->la_language == language_cplus) + { + if (strncmp (tokstart, "true", 4) == 0) + return TRUEKEYWORD; + } break; case 3: - if (STREQN (tokstart, "int", 3)) + if (strncmp (tokstart, "int", 3) == 0) return INT_KEYWORD; break; default: @@ -1464,14 +1685,36 @@ yylex () yylval.sval.ptr = tokstart; yylval.sval.length = namelen; - /* Any other names starting in $ are debugger internal variables. */ - if (*tokstart == '$') { - yylval.ivar = lookup_internalvar (copy_name (yylval.sval) + 1); + write_dollar_variable (yylval.sval); return VARIABLE; } - + + /* Look ahead and see if we can consume more of the input + string to get a reasonable class/namespace spec or a + fully-qualified name. This is a kludge to get around the + HP aCC compiler's generation of symbol names with embedded + colons for namespace and nested classes. */ + + /* NOTE: carlton/2003-09-24: I don't entirely understand the + HP-specific code, either here or in linespec. Having said that, + I suspect that we're actually moving towards their model: we want + symbols whose names are fully qualified, which matches the + description above. */ + if (unquoted_expr) + { + /* Only do it if not inside single quotes */ + sym_class = parse_nested_classes_for_hpacc (yylval.sval.ptr, yylval.sval.length, + &token_string, &class_prefix, &lexptr); + if (sym_class) + { + /* Replace the current token with the bigger one we found */ + yylval.sval.ptr = token_string; + yylval.sval.length = strlen (token_string); + } + } + /* Use token-type BLOCKNAME for symbols that happen to be defined as functions or symtabs. If this is not so, then ... Use token-type TYPENAME for symbols that happen to be defined @@ -1484,90 +1727,45 @@ yylex () int hextype; sym = lookup_symbol (tmp, expression_context_block, - VAR_NAMESPACE, + VAR_DOMAIN, current_language->la_language == language_cplus ? &is_a_field_of_this : (int *) NULL, (struct symtab **) NULL); - if ((sym && SYMBOL_CLASS (sym) == LOC_BLOCK) || - lookup_partial_symtab (tmp)) + /* Call lookup_symtab, not lookup_partial_symtab, in case there are + no psymtabs (coff, xcoff, or some future change to blow away the + psymtabs once once symbols are read). */ + if (sym && SYMBOL_CLASS (sym) == LOC_BLOCK) { yylval.ssym.sym = sym; yylval.ssym.is_a_field_of_this = is_a_field_of_this; return BLOCKNAME; } + else if (!sym) + { /* See if it's a file name. */ + struct symtab *symtab; + + symtab = lookup_symtab (tmp); + + if (symtab) + { + yylval.bval = BLOCKVECTOR_BLOCK (BLOCKVECTOR (symtab), STATIC_BLOCK); + return FILENAME; + } + } + if (sym && SYMBOL_CLASS (sym) == LOC_TYPEDEF) { - char *p; - char *namestart; - struct symbol *best_sym; - - /* Look ahead to detect nested types. This probably should be - done in the grammar, but trying seemed to introduce a lot - of shift/reduce and reduce/reduce conflicts. It's possible - that it could be done, though. Or perhaps a non-grammar, but - less ad hoc, approach would work well. */ - - /* Since we do not currently have any way of distinguishing - a nested type from a non-nested one (the stabs don't tell - us whether a type is nested), we just ignore the - containing type. */ - - p = lexptr; - best_sym = sym; - while (1) - { - /* Skip whitespace. */ - while (*p == ' ' || *p == '\t' || *p == '\n') - ++p; - if (*p == ':' && p[1] == ':') - { - /* Skip the `::'. */ - p += 2; - /* Skip whitespace. */ - while (*p == ' ' || *p == '\t' || *p == '\n') - ++p; - namestart = p; - while (*p == '_' || *p == '$' || (*p >= '0' && *p <= '9') - || (*p >= 'a' && *p <= 'z') - || (*p >= 'A' && *p <= 'Z')) - ++p; - if (p != namestart) - { - struct symbol *cur_sym; - /* As big as the whole rest of the expression, which is - at least big enough. */ - char *tmp = alloca (strlen (namestart)); - - memcpy (tmp, namestart, p - namestart); - tmp[p - namestart] = '\0'; - cur_sym = lookup_symbol (tmp, expression_context_block, - VAR_NAMESPACE, (int *) NULL, - (struct symtab **) NULL); - if (cur_sym) - { - if (SYMBOL_CLASS (cur_sym) == LOC_TYPEDEF) - { - best_sym = cur_sym; - lexptr = p; - } - else - break; - } - else - break; - } - else - break; - } - else - break; - } - - yylval.tsym.type = SYMBOL_TYPE (best_sym); + /* NOTE: carlton/2003-09-25: There used to be code here to + handle nested types. It didn't work very well. See the + comment before qualified_type for more info. */ + yylval.tsym.type = SYMBOL_TYPE (sym); return TYPENAME; } - if ((yylval.tsym.type = lookup_primitive_typename (tmp)) != 0) - return TYPENAME; + yylval.tsym.type + = language_lookup_primitive_type_by_name (current_language, + current_gdbarch, tmp); + if (yylval.tsym.type != NULL) + return TYPENAME; /* Input names that aren't symbols but ARE valid hex numbers, when the input radix permits them, can be names or numbers @@ -1597,5 +1795,8 @@ void yyerror (msg) char *msg; { - error (msg ? msg : "Invalid syntax in expression."); + if (prev_lexptr) + lexptr = prev_lexptr; + + error ("A %s in expression, near `%s'.", (msg ? msg : "error"), lexptr); }