/* YACC parser for C expressions, for GDB. Copyright (C) 1986, 1989, 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2003, 2004, 2006, 2007, 2008 Free Software Foundation, Inc. This file is part of GDB. 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 (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 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. */ /* Parse a C expression from text in a string, and return the result as a struct expression pointer. That structure contains arithmetic operations in reverse polish, with constants represented by operations that are followed by special data. See expression.h for the details of the format. What is important here is that it can be built up sequentially during the process of parsing; the lower levels of the tree always come first in the result. Note that malloc's and realloc's in this file are transformed to xmalloc and xrealloc respectively by the same sed command in the makefile that remaps any other malloc/realloc inserted by the parser generator. Doing this with #defines and trying to control the interaction with include files ( and for example) just became too messy, particularly when such includes can be inserted at random times by the parser generator. */ %{ #include "defs.h" #include "gdb_string.h" #include #include "expression.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" #include "dfp.h" #include "gdb_assert.h" #include "macroscope.h" #define parse_type builtin_type (parse_gdbarch) /* Remap normal yacc parser interface names (yyparse, yylex, yyerror, etc), as well as gratuitiously global symbol names, so we can have multiple yacc generated parsers in gdb. Note that these are only the variables produced by yacc. If other parser generators (bison, byacc, etc) produce additional global names that conflict at link time, then those parser generators need to be fixed instead of adding those names to this list. */ #define yymaxdepth c_maxdepth #define yyparse c_parse_internal #define yylex c_lex #define yyerror c_error #define yylval c_lval #define yychar c_char #define yydebug c_debug #define yypact c_pact #define yyr1 c_r1 #define yyr2 c_r2 #define yydef c_def #define yychk c_chk #define yypgo c_pgo #define yyact c_act #define yyexca c_exca #define yyerrflag c_errflag #define yynerrs c_nerrs #define yyps c_ps #define yypv c_pv #define yys c_s #define yy_yys c_yys #define yystate c_state #define yytmp c_tmp #define yyv c_v #define yy_yyv c_yyv #define yyval c_val #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 1 /* Default to yydebug support */ #endif #define YYFPRINTF parser_fprintf int yyparse (void); static int yylex (void); void yyerror (char *); %} /* Although the yacc "value" of an expression is not used, since the result is stored in the structure being created, other node types do have values. */ %union { LONGEST lval; struct { LONGEST val; struct type *type; } typed_val_int; struct { DOUBLEST dval; struct type *type; } typed_val_float; struct { gdb_byte val[16]; struct type *type; } typed_val_decfloat; struct symbol *sym; struct type *tval; struct stoken sval; struct ttype tsym; struct symtoken ssym; int voidval; struct block *bval; enum exp_opcode opcode; struct internalvar *ivar; struct type **tvec; int *ivec; } %{ /* YYSTYPE gets defined by %union */ static int parse_number (char *, int, int, YYSTYPE *); %} %type exp exp1 type_exp start variable qualified_name lcurly %type rcurly %type type typebase qualified_type %type nonempty_typelist /* %type block */ /* Fancy type parsing. */ %type func_mod direct_abs_decl abs_decl %type ptype %type array_mod %token INT %token FLOAT %token DECFLOAT /* Both NAME and TYPENAME tokens represent symbols in the input, and both convey their data as strings. But a TYPENAME is a string that happens to be defined as a typedef or builtin type name (such as int or char) and a NAME is any other symbol. Contexts where this distinction is not important can use the nonterminal "name", which matches either NAME or TYPENAME. */ %token STRING %token NAME /* BLOCKNAME defined below to give it higher precedence. */ %token COMPLETE %token TYPENAME %type name string_exp %type name_not_typename %type typename /* A NAME_OR_INT is a symbol which is not known in the symbol table, but which would parse as a valid number in the current input radix. E.g. "c" when input_radix==16. Depending on the parse, it will be turned into a name or into a number. */ %token NAME_OR_INT %token STRUCT CLASS UNION ENUM SIZEOF UNSIGNED COLONCOLON %token TEMPLATE %token ERROR /* 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 DOUBLE_KEYWORD %token VARIABLE %token ASSIGN_MODIFY /* C++ */ %token TRUEKEYWORD %token FALSEKEYWORD %left ',' %left ABOVE_COMMA %right '=' ASSIGN_MODIFY %right '?' %left OROR %left ANDAND %left '|' %left '^' %left '&' %left EQUAL NOTEQUAL %left '<' '>' LEQ GEQ %left LSH RSH %left '@' %left '+' '-' %left '*' '/' '%' %right UNARY INCREMENT DECREMENT %right ARROW '.' '[' '(' %token BLOCKNAME %token FILENAME %type block %left COLONCOLON %% start : exp1 | type_exp ; type_exp: type { write_exp_elt_opcode(OP_TYPE); write_exp_elt_type($1); write_exp_elt_opcode(OP_TYPE);} ; /* Expressions, including the comma operator. */ exp1 : exp | exp1 ',' exp { write_exp_elt_opcode (BINOP_COMMA); } ; /* 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); } ; exp : '~' exp %prec UNARY { write_exp_elt_opcode (UNOP_COMPLEMENT); } ; exp : INCREMENT exp %prec UNARY { write_exp_elt_opcode (UNOP_PREINCREMENT); } ; exp : DECREMENT exp %prec UNARY { write_exp_elt_opcode (UNOP_PREDECREMENT); } ; exp : exp INCREMENT %prec UNARY { write_exp_elt_opcode (UNOP_POSTINCREMENT); } ; exp : exp DECREMENT %prec UNARY { write_exp_elt_opcode (UNOP_POSTDECREMENT); } ; exp : SIZEOF exp %prec UNARY { write_exp_elt_opcode (UNOP_SIZEOF); } ; exp : exp ARROW name { write_exp_elt_opcode (STRUCTOP_PTR); write_exp_string ($3); write_exp_elt_opcode (STRUCTOP_PTR); } ; exp : exp ARROW name COMPLETE { mark_struct_expression (); write_exp_elt_opcode (STRUCTOP_PTR); write_exp_string ($3); write_exp_elt_opcode (STRUCTOP_PTR); } ; exp : exp ARROW COMPLETE { struct stoken s; mark_struct_expression (); write_exp_elt_opcode (STRUCTOP_PTR); s.ptr = ""; s.length = 0; write_exp_string (s); write_exp_elt_opcode (STRUCTOP_PTR); } ; exp : exp ARROW qualified_name { /* exp->type::name becomes exp->*(&type::name) */ /* Note: this doesn't work if name is a static member! FIXME */ write_exp_elt_opcode (UNOP_ADDR); write_exp_elt_opcode (STRUCTOP_MPTR); } ; exp : exp ARROW '*' exp { write_exp_elt_opcode (STRUCTOP_MPTR); } ; exp : exp '.' name { write_exp_elt_opcode (STRUCTOP_STRUCT); write_exp_string ($3); write_exp_elt_opcode (STRUCTOP_STRUCT); } ; exp : exp '.' name COMPLETE { mark_struct_expression (); write_exp_elt_opcode (STRUCTOP_STRUCT); write_exp_string ($3); write_exp_elt_opcode (STRUCTOP_STRUCT); } ; exp : exp '.' COMPLETE { struct stoken s; mark_struct_expression (); write_exp_elt_opcode (STRUCTOP_STRUCT); s.ptr = ""; s.length = 0; write_exp_string (s); write_exp_elt_opcode (STRUCTOP_STRUCT); } ; exp : exp '.' qualified_name { /* exp.type::name becomes exp.*(&type::name) */ /* Note: this doesn't work if name is a static member! FIXME */ write_exp_elt_opcode (UNOP_ADDR); write_exp_elt_opcode (STRUCTOP_MEMBER); } ; exp : exp '.' '*' exp { write_exp_elt_opcode (STRUCTOP_MEMBER); } ; exp : exp '[' exp1 ']' { write_exp_elt_opcode (BINOP_SUBSCRIPT); } ; exp : exp '(' /* This is to save the value of arglist_len being accumulated by an outer function call. */ { start_arglist (); } arglist ')' %prec ARROW { write_exp_elt_opcode (OP_FUNCALL); write_exp_elt_longcst ((LONGEST) end_arglist ()); write_exp_elt_opcode (OP_FUNCALL); } ; lcurly : '{' { start_arglist (); } ; arglist : ; arglist : exp { arglist_len = 1; } ; arglist : arglist ',' exp %prec ABOVE_COMMA { arglist_len++; } ; rcurly : '}' { $$ = end_arglist () - 1; } ; exp : lcurly arglist rcurly %prec ARROW { write_exp_elt_opcode (OP_ARRAY); write_exp_elt_longcst ((LONGEST) 0); write_exp_elt_longcst ((LONGEST) $3); write_exp_elt_opcode (OP_ARRAY); } ; exp : lcurly type rcurly exp %prec UNARY { write_exp_elt_opcode (UNOP_MEMVAL); write_exp_elt_type ($2); write_exp_elt_opcode (UNOP_MEMVAL); } ; exp : '(' type ')' exp %prec UNARY { write_exp_elt_opcode (UNOP_CAST); write_exp_elt_type ($2); write_exp_elt_opcode (UNOP_CAST); } ; exp : '(' exp1 ')' { } ; /* Binary operators in order of decreasing precedence. */ exp : exp '@' exp { write_exp_elt_opcode (BINOP_REPEAT); } ; exp : exp '*' exp { write_exp_elt_opcode (BINOP_MUL); } ; exp : exp '/' exp { write_exp_elt_opcode (BINOP_DIV); } ; exp : exp '%' exp { write_exp_elt_opcode (BINOP_REM); } ; exp : exp '+' exp { write_exp_elt_opcode (BINOP_ADD); } ; exp : exp '-' exp { write_exp_elt_opcode (BINOP_SUB); } ; exp : exp LSH exp { write_exp_elt_opcode (BINOP_LSH); } ; exp : exp RSH exp { write_exp_elt_opcode (BINOP_RSH); } ; exp : exp EQUAL exp { write_exp_elt_opcode (BINOP_EQUAL); } ; exp : exp NOTEQUAL exp { write_exp_elt_opcode (BINOP_NOTEQUAL); } ; exp : exp LEQ exp { write_exp_elt_opcode (BINOP_LEQ); } ; exp : exp GEQ exp { write_exp_elt_opcode (BINOP_GEQ); } ; exp : exp '<' exp { write_exp_elt_opcode (BINOP_LESS); } ; exp : exp '>' exp { write_exp_elt_opcode (BINOP_GTR); } ; exp : exp '&' exp { write_exp_elt_opcode (BINOP_BITWISE_AND); } ; exp : exp '^' exp { write_exp_elt_opcode (BINOP_BITWISE_XOR); } ; exp : exp '|' exp { write_exp_elt_opcode (BINOP_BITWISE_IOR); } ; exp : exp ANDAND exp { write_exp_elt_opcode (BINOP_LOGICAL_AND); } ; exp : exp OROR exp { write_exp_elt_opcode (BINOP_LOGICAL_OR); } ; exp : exp '?' exp ':' exp %prec '?' { write_exp_elt_opcode (TERNOP_COND); } ; exp : exp '=' exp { write_exp_elt_opcode (BINOP_ASSIGN); } ; exp : exp ASSIGN_MODIFY exp { write_exp_elt_opcode (BINOP_ASSIGN_MODIFY); write_exp_elt_opcode ($2); write_exp_elt_opcode (BINOP_ASSIGN_MODIFY); } ; exp : INT { write_exp_elt_opcode (OP_LONG); write_exp_elt_type ($1.type); write_exp_elt_longcst ((LONGEST)($1.val)); write_exp_elt_opcode (OP_LONG); } ; 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_int.type); write_exp_elt_longcst ((LONGEST)val.typed_val_int.val); write_exp_elt_opcode (OP_LONG); } ; exp : FLOAT { write_exp_elt_opcode (OP_DOUBLE); write_exp_elt_type ($1.type); write_exp_elt_dblcst ($1.dval); write_exp_elt_opcode (OP_DOUBLE); } ; exp : DECFLOAT { write_exp_elt_opcode (OP_DECFLOAT); write_exp_elt_type ($1.type); write_exp_elt_decfloatcst ($1.val); write_exp_elt_opcode (OP_DECFLOAT); } ; exp : variable ; exp : VARIABLE /* Already written by write_dollar_variable. */ ; exp : SIZEOF '(' type ')' %prec UNARY { write_exp_elt_opcode (OP_LONG); write_exp_elt_type (parse_type->builtin_int); CHECK_TYPEDEF ($3); write_exp_elt_longcst ((LONGEST) TYPE_LENGTH ($3)); write_exp_elt_opcode (OP_LONG); } ; string_exp: STRING { /* We copy the string here, and not in the lexer, to guarantee that we do not leak a string. Note that we follow the NUL-termination convention of the lexer. */ $$.length = $1.length; $$.ptr = malloc ($1.length + 1); memcpy ($$.ptr, $1.ptr, $1.length + 1); } | string_exp STRING { /* Note that we NUL-terminate here, but just for convenience. */ struct stoken t; t.length = $1.length + $2.length; t.ptr = malloc (t.length + 1); memcpy (t.ptr, $1.ptr, $1.length); memcpy (t.ptr + $1.length, $2.ptr, $2.length + 1); free ($1.ptr); $$ = t; } ; exp : string_exp { /* C strings are converted into array constants with an explicit null byte added at the end. Thus the array upper bound is the string length. There is no such thing in C as a completely empty string. */ char *sp = $1.ptr; int count = $1.length; while (count-- > 0) { write_exp_elt_opcode (OP_LONG); write_exp_elt_type (parse_type->builtin_char); write_exp_elt_longcst ((LONGEST)(*sp++)); write_exp_elt_opcode (OP_LONG); } write_exp_elt_opcode (OP_LONG); write_exp_elt_type (parse_type->builtin_char); write_exp_elt_longcst ((LONGEST)'\0'); write_exp_elt_opcode (OP_LONG); write_exp_elt_opcode (OP_ARRAY); write_exp_elt_longcst ((LONGEST) 0); write_exp_elt_longcst ((LONGEST) ($1.length)); write_exp_elt_opcode (OP_ARRAY); free ($1.ptr); } ; /* C++. */ exp : TRUEKEYWORD { write_exp_elt_opcode (OP_LONG); write_exp_elt_type (parse_type->builtin_bool); write_exp_elt_longcst ((LONGEST) 1); write_exp_elt_opcode (OP_LONG); } ; exp : FALSEKEYWORD { write_exp_elt_opcode (OP_LONG); write_exp_elt_type (parse_type->builtin_bool); write_exp_elt_longcst ((LONGEST) 0); write_exp_elt_opcode (OP_LONG); } ; /* end of C++. */ block : BLOCKNAME { if ($1.sym) $$ = SYMBOL_BLOCK_VALUE ($1.sym); else error ("No file or function \"%s\".", copy_name ($1.stoken)); } | FILENAME { $$ = $1; } ; block : block COLONCOLON name { struct symbol *tem = lookup_symbol (copy_name ($3), $1, VAR_DOMAIN, (int *) NULL); if (!tem || SYMBOL_CLASS (tem) != LOC_BLOCK) error ("No function \"%s\" in specified context.", copy_name ($3)); $$ = SYMBOL_BLOCK_VALUE (tem); } ; variable: block COLONCOLON name { struct symbol *sym; sym = lookup_symbol (copy_name ($3), $1, VAR_DOMAIN, (int *) NULL); if (sym == 0) error ("No symbol \"%s\" in specified context.", copy_name ($3)); write_exp_elt_opcode (OP_VAR_VALUE); /* block_found is set by lookup_symbol. */ write_exp_elt_block (block_found); write_exp_elt_sym (sym); write_exp_elt_opcode (OP_VAR_VALUE); } ; 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_NAMESPACE) error ("`%s' is not defined as an aggregate type.", TYPE_NAME (type)); write_exp_elt_opcode (OP_SCOPE); write_exp_elt_type (type); write_exp_string ($3); write_exp_elt_opcode (OP_SCOPE); } | 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_NAMESPACE) error ("`%s' is not defined as an aggregate type.", TYPE_NAME (type)); 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); write_exp_elt_opcode (OP_SCOPE); } ; variable: qualified_name | COLONCOLON name { char *name = copy_name ($2); struct symbol *sym; struct minimal_symbol *msymbol; sym = lookup_symbol (name, (const struct block *) NULL, VAR_DOMAIN, (int *) NULL); if (sym) { write_exp_elt_opcode (OP_VAR_VALUE); write_exp_elt_block (NULL); write_exp_elt_sym (sym); write_exp_elt_opcode (OP_VAR_VALUE); break; } msymbol = lookup_minimal_symbol (name, NULL, NULL); if (msymbol != NULL) write_exp_msymbol (msymbol); else if (!have_full_symbols () && !have_partial_symbols ()) error ("No symbol table is loaded. Use the \"file\" command."); else error ("No symbol \"%s\" in current context.", name); } ; variable: name_not_typename { struct symbol *sym = $1.sym; if (sym) { if (symbol_read_needs_frame (sym)) { if (innermost_block == 0 || contained_in (block_found, innermost_block)) innermost_block = block_found; } write_exp_elt_opcode (OP_VAR_VALUE); /* We want to use the selected frame, not another more inner frame which happens to be in the same block. */ write_exp_elt_block (NULL); write_exp_elt_sym (sym); write_exp_elt_opcode (OP_VAR_VALUE); } else if ($1.is_a_field_of_this) { /* C++: it hangs off of `this'. Must not inadvertently convert from a method call to data ref. */ if (innermost_block == 0 || contained_in (block_found, innermost_block)) innermost_block = block_found; write_exp_elt_opcode (OP_THIS); write_exp_elt_opcode (OP_THIS); write_exp_elt_opcode (STRUCTOP_PTR); write_exp_string ($1.stoken); write_exp_elt_opcode (STRUCTOP_PTR); } else { struct minimal_symbol *msymbol; char *arg = copy_name ($1.stoken); msymbol = lookup_minimal_symbol (arg, NULL, NULL); if (msymbol != NULL) write_exp_msymbol (msymbol); else if (!have_full_symbols () && !have_partial_symbols ()) error ("No symbol table is loaded. Use the \"file\" command."); else error ("No symbol \"%s\" in current context.", copy_name ($1.stoken)); } } ; space_identifier : '@' NAME { push_type_address_space (copy_name ($2.stoken)); push_type (tp_space_identifier); } ; 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: '*' { push_type (tp_pointer); $$ = 0; } | '*' abs_decl { push_type (tp_pointer); $$ = $2; } | '&' { push_type (tp_reference); $$ = 0; } | '&' abs_decl { push_type (tp_reference); $$ = $2; } | direct_abs_decl ; direct_abs_decl: '(' abs_decl ')' { $$ = $2; } | direct_abs_decl array_mod { push_type_int ($2); push_type (tp_array); } | array_mod { push_type_int ($1); push_type (tp_array); $$ = 0; } | direct_abs_decl func_mod { push_type (tp_function); } | func_mod { push_type (tp_function); } ; array_mod: '[' ']' { $$ = -1; } | '[' INT ']' { $$ = $2.val; } ; func_mod: '(' ')' { $$ = 0; } | '(' nonempty_typelist ')' { free ($2); $$ = 0; } ; /* 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 /* Implements (approximately): (type-qualifier)* type-specifier */ : TYPENAME { $$ = $1.type; } | INT_KEYWORD { $$ = parse_type->builtin_int; } | LONG { $$ = parse_type->builtin_long; } | SHORT { $$ = parse_type->builtin_short; } | LONG INT_KEYWORD { $$ = parse_type->builtin_long; } | LONG SIGNED_KEYWORD INT_KEYWORD { $$ = parse_type->builtin_long; } | LONG SIGNED_KEYWORD { $$ = parse_type->builtin_long; } | SIGNED_KEYWORD LONG INT_KEYWORD { $$ = parse_type->builtin_long; } | UNSIGNED LONG INT_KEYWORD { $$ = parse_type->builtin_unsigned_long; } | LONG UNSIGNED INT_KEYWORD { $$ = parse_type->builtin_unsigned_long; } | LONG UNSIGNED { $$ = parse_type->builtin_unsigned_long; } | LONG LONG { $$ = parse_type->builtin_long_long; } | LONG LONG INT_KEYWORD { $$ = parse_type->builtin_long_long; } | LONG LONG SIGNED_KEYWORD INT_KEYWORD { $$ = parse_type->builtin_long_long; } | LONG LONG SIGNED_KEYWORD { $$ = parse_type->builtin_long_long; } | SIGNED_KEYWORD LONG LONG { $$ = parse_type->builtin_long_long; } | SIGNED_KEYWORD LONG LONG INT_KEYWORD { $$ = parse_type->builtin_long_long; } | UNSIGNED LONG LONG { $$ = parse_type->builtin_unsigned_long_long; } | UNSIGNED LONG LONG INT_KEYWORD { $$ = parse_type->builtin_unsigned_long_long; } | LONG LONG UNSIGNED { $$ = parse_type->builtin_unsigned_long_long; } | LONG LONG UNSIGNED INT_KEYWORD { $$ = parse_type->builtin_unsigned_long_long; } | SHORT INT_KEYWORD { $$ = parse_type->builtin_short; } | SHORT SIGNED_KEYWORD INT_KEYWORD { $$ = parse_type->builtin_short; } | SHORT SIGNED_KEYWORD { $$ = parse_type->builtin_short; } | UNSIGNED SHORT INT_KEYWORD { $$ = parse_type->builtin_unsigned_short; } | SHORT UNSIGNED { $$ = parse_type->builtin_unsigned_short; } | SHORT UNSIGNED INT_KEYWORD { $$ = parse_type->builtin_unsigned_short; } | DOUBLE_KEYWORD { $$ = parse_type->builtin_double; } | LONG DOUBLE_KEYWORD { $$ = parse_type->builtin_long_double; } | STRUCT name { $$ = lookup_struct (copy_name ($2), expression_context_block); } | CLASS name { $$ = lookup_struct (copy_name ($2), expression_context_block); } | UNION name { $$ = lookup_union (copy_name ($2), expression_context_block); } | ENUM name { $$ = lookup_enum (copy_name ($2), expression_context_block); } | UNSIGNED typename { $$ = lookup_unsigned_typename (TYPE_NAME($2.type)); } | UNSIGNED { $$ = parse_type->builtin_unsigned_int; } | SIGNED_KEYWORD typename { $$ = lookup_signed_typename (TYPE_NAME($2.type)); } | SIGNED_KEYWORD { $$ = parse_type->builtin_int; } /* It appears that this rule for templates is never reduced; template recognition happens by lookahead in the token processing code in yylex. */ | TEMPLATE name '<' type '>' { $$ = lookup_template_type(copy_name($2), $4, expression_context_block); } | const_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)); $$ = new_type; } ; typename: TYPENAME | INT_KEYWORD { $$.stoken.ptr = "int"; $$.stoken.length = 3; $$.type = parse_type->builtin_int; } | LONG { $$.stoken.ptr = "long"; $$.stoken.length = 4; $$.type = parse_type->builtin_long; } | SHORT { $$.stoken.ptr = "short"; $$.stoken.length = 5; $$.type = parse_type->builtin_short; } ; nonempty_typelist : type { $$ = (struct type **) malloc (sizeof (struct type *) * 2); $$[0] = 1; /* Number of types in vector */ $$[1] = $1; } | nonempty_typelist ',' type { int len = sizeof (struct type *) * (++($1[0]) + 1); $$ = (struct type **) realloc ((char *) $1, len); $$[$$[0]] = $3; } ; 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; } | NAME_OR_INT { $$ = $1.stoken; } ; name_not_typename : NAME | BLOCKNAME /* These would be useful if name_not_typename was useful, but it is just a fake for "variable", so these cause reduce/reduce conflicts because the parser can't tell whether NAME_OR_INT is a name_not_typename (=variable, =exp) or just an exp. If name_not_typename was ever used in an lvalue context where only a name could occur, this might be useful. | NAME_OR_INT */ ; %% /* Take care of parsing a number (anything that starts with a digit). Set yylval and return the token type; update lexptr. LEN is the number of characters in it. */ /*** Needs some error checking for the float case ***/ static int parse_number (p, len, parsed_float, putithere) char *p; int len; int parsed_float; YYSTYPE *putithere; { /* 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; /* 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. */ char *s; int num; /* number of tokens scanned by scanf */ char saved_char; /* If it ends at "df", "dd" or "dl", take it as type of decimal floating point. Return DECFLOAT. */ if (len >= 2 && p[len - 2] == 'd' && p[len - 1] == 'f') { p[len - 2] = '\0'; putithere->typed_val_decfloat.type = parse_type->builtin_decfloat; decimal_from_string (putithere->typed_val_decfloat.val, 4, p); p[len - 2] = 'd'; return DECFLOAT; } if (len >= 2 && p[len - 2] == 'd' && p[len - 1] == 'd') { p[len - 2] = '\0'; putithere->typed_val_decfloat.type = parse_type->builtin_decdouble; decimal_from_string (putithere->typed_val_decfloat.val, 8, p); p[len - 2] = 'd'; return DECFLOAT; } if (len >= 2 && p[len - 2] == 'd' && p[len - 1] == 'l') { p[len - 2] = '\0'; putithere->typed_val_decfloat.type = parse_type->builtin_declong; decimal_from_string (putithere->typed_val_decfloat.val, 16, p); p[len - 2] = 'd'; return DECFLOAT; } s = malloc (len); 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 = parse_type->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 = parse_type->builtin_float; else if (!strcasecmp (s, "l")) putithere->typed_val_float.type = parse_type->builtin_long_double; else { free (s); return ERROR; } } free (s); return FLOAT; } /* Handle base-switching prefixes 0x, 0t, 0d, 0 */ if (p[0] == '0') switch (p[1]) { case 'x': case 'X': if (len >= 3) { p += 2; base = 16; len -= 2; } break; case 't': case 'T': case 'd': case 'D': if (len >= 3) { p += 2; base = 10; len -= 2; } break; default: base = 8; break; } while (len-- > 0) { c = *p++; if (c >= 'A' && c <= 'Z') c += 'a' - 'A'; if (c != 'l' && c != 'u') n *= base; if (c >= '0' && c <= '9') { if (found_suffix) return ERROR; n += i = c - '0'; } else { if (base > 10 && c >= 'a' && c <= 'f') { 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 */ } if (i >= base) return ERROR; /* Invalid digit in this base */ /* Portably test for overflow (only works for nonzero values, so make 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 gdbarch_int_bit or gdbarch_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 >> (gdbarch_int_bit (parse_gdbarch) - 2)) == 0) { high_bit = ((ULONGEST)1) << (gdbarch_int_bit (parse_gdbarch) - 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 = parse_type->builtin_unsigned_int; signed_type = parse_type->builtin_int; } else if (long_p <= 1 && (un >> (gdbarch_long_bit (parse_gdbarch) - 2)) == 0) { high_bit = ((ULONGEST)1) << (gdbarch_long_bit (parse_gdbarch) - 1); unsigned_type = parse_type->builtin_unsigned_long; signed_type = parse_type->builtin_long; } else { int shift; if (sizeof (ULONGEST) * HOST_CHAR_BIT < gdbarch_long_long_bit (parse_gdbarch)) /* A long long does not fit in a LONGEST. */ shift = (sizeof (ULONGEST) * HOST_CHAR_BIT - 1); else shift = (gdbarch_long_long_bit (parse_gdbarch) - 1); high_bit = (ULONGEST) 1 << shift; unsigned_type = parse_type->builtin_unsigned_long_long; signed_type = parse_type->builtin_long_long; } 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_int.type = unsigned_type; } else { putithere->typed_val_int.type = signed_type; } return INT; } struct token { char *operator; int token; enum exp_opcode opcode; int cxx_only; }; static const struct token tokentab3[] = { {">>=", ASSIGN_MODIFY, BINOP_RSH, 0}, {"<<=", ASSIGN_MODIFY, BINOP_LSH, 0} }; static const struct token tokentab2[] = { {"+=", ASSIGN_MODIFY, BINOP_ADD, 0}, {"-=", ASSIGN_MODIFY, BINOP_SUB, 0}, {"*=", ASSIGN_MODIFY, BINOP_MUL, 0}, {"/=", ASSIGN_MODIFY, BINOP_DIV, 0}, {"%=", ASSIGN_MODIFY, BINOP_REM, 0}, {"|=", ASSIGN_MODIFY, BINOP_BITWISE_IOR, 0}, {"&=", ASSIGN_MODIFY, BINOP_BITWISE_AND, 0}, {"^=", ASSIGN_MODIFY, BINOP_BITWISE_XOR, 0}, {"++", INCREMENT, BINOP_END, 0}, {"--", DECREMENT, BINOP_END, 0}, {"->", ARROW, BINOP_END, 0}, {"&&", ANDAND, BINOP_END, 0}, {"||", OROR, BINOP_END, 0}, {"::", COLONCOLON, BINOP_END, 0}, {"<<", LSH, BINOP_END, 0}, {">>", RSH, BINOP_END, 0}, {"==", EQUAL, BINOP_END, 0}, {"!=", NOTEQUAL, BINOP_END, 0}, {"<=", LEQ, BINOP_END, 0}, {">=", GEQ, BINOP_END, 0} }; /* Identifier-like tokens. */ static const struct token ident_tokens[] = { {"unsigned", UNSIGNED, OP_NULL, 0}, {"template", TEMPLATE, OP_NULL, 1}, {"volatile", VOLATILE_KEYWORD, OP_NULL, 0}, {"struct", STRUCT, OP_NULL, 0}, {"signed", SIGNED_KEYWORD, OP_NULL, 0}, {"sizeof", SIZEOF, OP_NULL, 0}, {"double", DOUBLE_KEYWORD, OP_NULL, 0}, {"false", FALSEKEYWORD, OP_NULL, 1}, {"class", CLASS, OP_NULL, 1}, {"union", UNION, OP_NULL, 0}, {"short", SHORT, OP_NULL, 0}, {"const", CONST_KEYWORD, OP_NULL, 0}, {"enum", ENUM, OP_NULL, 0}, {"long", LONG, OP_NULL, 0}, {"true", TRUEKEYWORD, OP_NULL, 1}, {"int", INT_KEYWORD, OP_NULL, 0}, {"and", ANDAND, BINOP_END, 1}, {"and_eq", ASSIGN_MODIFY, BINOP_BITWISE_AND, 1}, {"bitand", '&', OP_NULL, 1}, {"bitor", '|', OP_NULL, 1}, {"compl", '~', OP_NULL, 1}, {"not", '!', OP_NULL, 1}, {"not_eq", NOTEQUAL, BINOP_END, 1}, {"or", OROR, BINOP_END, 1}, {"or_eq", ASSIGN_MODIFY, BINOP_BITWISE_IOR, 1}, {"xor", '^', OP_NULL, 1}, {"xor_eq", ASSIGN_MODIFY, BINOP_BITWISE_XOR, 1} }; /* When we find that lexptr (the global var defined in parse.c) is pointing at a macro invocation, we expand the invocation, and call scan_macro_expansion to save the old lexptr here and point lexptr into the expanded text. When we reach the end of that, we call end_macro_expansion to pop back to the value we saved here. The macro expansion code promises to return only fully-expanded text, so we don't need to "push" more than one level. This is disgusting, of course. It would be cleaner to do all macro expansion beforehand, and then hand that to lexptr. But we don't really know where the expression ends. Remember, in a command like (gdb) break *ADDRESS if CONDITION we evaluate ADDRESS in the scope of the current frame, but we evaluate CONDITION in the scope of the breakpoint's location. So it's simply wrong to try to macro-expand the whole thing at once. */ static char *macro_original_text; /* We save all intermediate macro expansions on this obstack for the duration of a single parse. The expansion text may sometimes have to live past the end of the expansion, due to yacc lookahead. Rather than try to be clever about saving the data for a single token, we simply keep it all and delete it after parsing has completed. */ static struct obstack expansion_obstack; static void scan_macro_expansion (char *expansion) { char *copy; /* We'd better not be trying to push the stack twice. */ gdb_assert (! macro_original_text); /* Copy to the obstack, and then free the intermediate expansion. */ copy = obstack_copy0 (&expansion_obstack, expansion, strlen (expansion)); xfree (expansion); /* Save the old lexptr value, so we can return to it when we're done parsing the expanded text. */ macro_original_text = lexptr; lexptr = copy; } static int scanning_macro_expansion (void) { return macro_original_text != 0; } static void finished_macro_expansion (void) { /* There'd better be something to pop back to. */ gdb_assert (macro_original_text); /* Pop back to the original text. */ lexptr = macro_original_text; macro_original_text = 0; } static void scan_macro_cleanup (void *dummy) { if (macro_original_text) finished_macro_expansion (); obstack_free (&expansion_obstack, NULL); } /* The scope used for macro expansion. */ static struct macro_scope *expression_macro_scope; /* This is set if a NAME token appeared at the very end of the input string, with no whitespace separating the name from the EOF. This is used only when parsing to do field name completion. */ static int saw_name_at_eof; /* This is set if the previously-returned token was a structure operator -- either '.' or ARROW. This is used only when parsing to do field name completion. */ static int last_was_structop; /* Read one token, getting characters through lexptr. */ static int yylex () { int c; int namelen; unsigned int i; char *tokstart; char *tokptr; int tempbufindex; static char *tempbuf; static int tempbufsize; char * token_string = NULL; int class_prefix = 0; int saw_structop = last_was_structop; char *copy; last_was_structop = 0; retry: /* Check if this is a macro invocation that we need to expand. */ if (! scanning_macro_expansion ()) { char *expanded = macro_expand_next (&lexptr, standard_macro_lookup, expression_macro_scope); if (expanded) scan_macro_expansion (expanded); } prev_lexptr = lexptr; tokstart = lexptr; /* See if it is a special token of length 3. */ for (i = 0; i < sizeof tokentab3 / sizeof tokentab3[0]; i++) if (strncmp (tokstart, tokentab3[i].operator, 3) == 0) { lexptr += 3; yylval.opcode = tokentab3[i].opcode; return tokentab3[i].token; } /* See if it is a special token of length 2. */ for (i = 0; i < sizeof tokentab2 / sizeof tokentab2[0]; i++) if (strncmp (tokstart, tokentab2[i].operator, 2) == 0) { lexptr += 2; yylval.opcode = tokentab2[i].opcode; if (in_parse_field && tokentab2[i].token == ARROW) last_was_structop = 1; return tokentab2[i].token; } switch (c = *tokstart) { case 0: /* If we were just scanning the result of a macro expansion, then we need to resume scanning the original text. If we're parsing for field name completion, and the previous token allows such completion, return a COMPLETE token. Otherwise, we were already scanning the original text, and we're really done. */ if (scanning_macro_expansion ()) { finished_macro_expansion (); goto retry; } else if (saw_name_at_eof) { saw_name_at_eof = 0; return COMPLETE; } else if (saw_structop) return COMPLETE; else return 0; case ' ': case '\t': case '\n': lexptr++; goto retry; case '\'': /* We either have a character constant ('0' or '\177' for example) or we have a quoted symbol reference ('foo(int,int)' in C++ for example). */ lexptr++; 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_int.val = c; yylval.typed_val_int.type = parse_type->builtin_char; c = *lexptr++; if (c != '\'') { namelen = skip_quoted (tokstart) - tokstart; if (namelen > 2) { lexptr = tokstart + namelen; if (lexptr[-1] != '\'') error ("Unmatched single quote."); namelen -= 2; tokstart++; goto tryname; } error ("Invalid character constant."); } return INT; case '(': paren_depth++; lexptr++; return c; case ')': if (paren_depth == 0) return 0; paren_depth--; lexptr++; return c; case ',': if (comma_terminates && paren_depth == 0 && ! scanning_macro_expansion ()) return 0; lexptr++; return c; case '.': /* Might be a floating point number. */ if (lexptr[1] < '0' || lexptr[1] > '9') { if (in_parse_field) last_was_structop = 1; goto symbol; /* Nope, must be a symbol. */ } /* FALL THRU into number case. */ case '0': case '1': case '2': case '3': case '4': case '5': case '6': case '7': case '8': case '9': { /* It's a number. */ int got_dot = 0, got_e = 0, toktype; char *p = tokstart; int hex = input_radix > 10; if (c == '0' && (p[1] == 'x' || p[1] == 'X')) { p += 2; hex = 1; } else if (c == '0' && (p[1]=='t' || p[1]=='T' || p[1]=='d' || p[1]=='D')) { p += 2; hex = 0; } for (;; ++p) { /* This test includes !hex because 'e' is a valid hex digit and thus does not indicate a floating point number when the radix is hex. */ if (!hex && !got_e && (*p == 'e' || *p == 'E')) got_dot = got_e = 1; /* This test does not include !hex, because a '.' always indicates a decimal floating point number regardless of the radix. */ else if (!got_dot && *p == '.') got_dot = 1; else if (got_e && (p[-1] == 'e' || p[-1] == 'E') && (*p == '-' || *p == '+')) /* This is the sign of the exponent, not the end of the number. */ continue; /* We will take any letters or digits. parse_number will complain if past the radix, or if L or U are not final. */ else if ((*p < '0' || *p > '9') && ((*p < 'a' || *p > 'z') && (*p < 'A' || *p > 'Z'))) break; } toktype = parse_number (tokstart, p - tokstart, got_dot|got_e, &yylval); if (toktype == ERROR) { char *err_copy = (char *) alloca (p - tokstart + 1); memcpy (err_copy, tokstart, p - tokstart); err_copy[p - tokstart] = 0; error ("Invalid number \"%s\".", err_copy); } lexptr = p; return toktype; } case '+': case '-': case '*': case '/': case '%': case '|': case '&': case '^': case '~': case '!': case '@': case '<': case '>': case '[': case ']': case '?': case ':': case '=': case '{': case '}': symbol: lexptr++; return c; case '"': /* Build the gdb internal form of the input string in tempbuf, translating any standard C escape forms seen. Note that the buffer is null byte terminated *only* for the convenience of debugging gdb itself and printing the buffer contents when the buffer contains no embedded nulls. Gdb does not depend upon the buffer being null byte terminated, it uses the length string instead. This allows gdb to handle C strings (as well as strings in other languages) with embedded null bytes */ tokptr = ++tokstart; 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) { tempbuf = (char *) realloc (tempbuf, tempbufsize += 64); } switch (*tokptr) { case '\0': case '"': /* Do nothing, loop will terminate. */ break; case '\\': tokptr++; c = parse_escape (&tokptr); if (c == -1) { continue; } tempbuf[tempbufindex++] = c; break; default: 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')); if (*tokptr++ != '"') { error ("Unterminated string in expression."); } tempbuf[tempbufindex] = '\0'; /* See note above */ yylval.sval.ptr = tempbuf; yylval.sval.length = tempbufindex; lexptr = tokptr; return (STRING); } if (!(c == '_' || c == '$' || (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z'))) /* We must have come across a bad character (e.g. ';'). */ error ("Invalid character '%c' in expression.", c); /* It's a name. See how long it is. */ namelen = 0; for (c = tokstart[namelen]; (c == '_' || c == '$' || (c >= '0' && c <= '9') || (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. 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; tryname: yylval.sval.ptr = tokstart; yylval.sval.length = namelen; /* Catch specific keywords. */ copy = copy_name (yylval.sval); for (i = 0; i < sizeof ident_tokens / sizeof ident_tokens[0]; i++) if (strcmp (copy, ident_tokens[i].operator) == 0) { if (ident_tokens[i].cxx_only && parse_language->la_language != language_cplus) break; /* It is ok to always set this, even though we don't always strictly need to. */ yylval.opcode = ident_tokens[i].opcode; return ident_tokens[i].token; } if (*tokstart == '$') { write_dollar_variable (yylval.sval); return VARIABLE; } /* 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 currently as names of types; NAME for other symbols. The caller is not constrained to care about the distinction. */ { struct symbol *sym; int is_a_field_of_this = 0; int hextype; sym = lookup_symbol (copy, expression_context_block, VAR_DOMAIN, parse_language->la_language == language_cplus ? &is_a_field_of_this : (int *) NULL); /* 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 (copy); if (symtab) { yylval.bval = BLOCKVECTOR_BLOCK (BLOCKVECTOR (symtab), STATIC_BLOCK); return FILENAME; } } if (sym && SYMBOL_CLASS (sym) == LOC_TYPEDEF) { /* 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; } yylval.tsym.type = language_lookup_primitive_type_by_name (parse_language, parse_gdbarch, copy); 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 depending on the parse. Note we support radixes > 16 here. */ if (!sym && ((tokstart[0] >= 'a' && tokstart[0] < 'a' + input_radix - 10) || (tokstart[0] >= 'A' && tokstart[0] < 'A' + input_radix - 10))) { YYSTYPE newlval; /* Its value is ignored. */ hextype = parse_number (tokstart, namelen, 0, &newlval); if (hextype == INT) { yylval.ssym.sym = sym; yylval.ssym.is_a_field_of_this = is_a_field_of_this; return NAME_OR_INT; } } /* Any other kind of symbol */ yylval.ssym.sym = sym; yylval.ssym.is_a_field_of_this = is_a_field_of_this; if (in_parse_field && *lexptr == '\0') saw_name_at_eof = 1; return NAME; } } int c_parse (void) { int result; struct cleanup *back_to = make_cleanup (free_current_contents, &expression_macro_scope); /* Set up the scope for macro expansion. */ expression_macro_scope = NULL; if (expression_context_block) expression_macro_scope = sal_macro_scope (find_pc_line (expression_context_pc, 0)); else expression_macro_scope = default_macro_scope (); if (! expression_macro_scope) expression_macro_scope = user_macro_scope (); /* Initialize macro expansion code. */ obstack_init (&expansion_obstack); gdb_assert (! macro_original_text); make_cleanup (scan_macro_cleanup, 0); /* Initialize some state used by the lexer. */ last_was_structop = 0; saw_name_at_eof = 0; result = yyparse (); do_cleanups (back_to); return result; } void yyerror (msg) char *msg; { if (prev_lexptr) lexptr = prev_lexptr; error ("A %s in expression, near `%s'.", (msg ? msg : "error"), lexptr); }