/* 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
+ 1998, 1999, 2000, 2003, 2004, 2006, 2007, 2008, 2009, 2010
Free Software Foundation, Inc.
-This file is part of GDB.
+ 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 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 3 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.
+ 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. */
+ You should have received a copy of the GNU General Public License
+ along with this program. If not, see <http://www.gnu.org/licenses/>. */
/* Parse a C expression from text in a string,
and return the result as a struct expression pointer.
struct symbol *sym;
struct type *tval;
struct stoken sval;
+ struct typed_stoken tsval;
struct ttype tsym;
struct symtoken ssym;
int voidval;
enum exp_opcode opcode;
struct internalvar *ivar;
+ struct stoken_vector svec;
struct type **tvec;
int *ivec;
}
%{
/* YYSTYPE gets defined by %union */
static int parse_number (char *, int, int, YYSTYPE *);
+static struct stoken operator_stoken (const char *);
%}
%type <voidval> exp exp1 type_exp start variable qualified_name lcurly
%type <lval> rcurly
-%type <tval> type typebase qualified_type
+%type <tval> type typebase
%type <tvec> nonempty_typelist
/* %type <bval> block */
Contexts where this distinction is not important can use the
nonterminal "name", which matches either NAME or TYPENAME. */
-%token <sval> STRING
+%token <tsval> STRING
+%token <tsval> CHAR
%token <ssym> NAME /* BLOCKNAME defined below to give it higher precedence. */
+%token <ssym> UNKNOWN_CPP_NAME
%token <voidval> COMPLETE
%token <tsym> TYPENAME
-%type <sval> name string_exp
+%type <sval> name
+%type <svec> string_exp
%type <ssym> name_not_typename
%type <tsym> typename
%token <ssym> NAME_OR_INT
+%token OPERATOR
%token STRUCT CLASS UNION ENUM SIZEOF UNSIGNED COLONCOLON
%token TEMPLATE
%token ERROR
+%token NEW DELETE
+%type <sval> operator
+%token REINTERPRET_CAST DYNAMIC_CAST STATIC_CAST CONST_CAST
/* 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 <voidval> VARIABLE
+%token <sval> VARIABLE
%token <opcode> ASSIGN_MODIFY
%left '+' '-'
%left '*' '/' '%'
%right UNARY INCREMENT DECREMENT
-%right ARROW '.' '[' '('
+%right ARROW ARROW_STAR '.' DOT_STAR '[' '('
%token <ssym> BLOCKNAME
%token <bval> FILENAME
%type <bval> block
write_exp_elt_opcode (STRUCTOP_MPTR); }
;
-exp : exp ARROW '*' exp
+exp : exp ARROW_STAR exp
{ write_exp_elt_opcode (STRUCTOP_MPTR); }
;
write_exp_elt_opcode (STRUCTOP_MEMBER); }
;
-exp : exp '.' '*' exp
+exp : exp DOT_STAR exp
{ write_exp_elt_opcode (STRUCTOP_MEMBER); }
;
write_exp_elt_opcode (OP_FUNCALL); }
;
+exp : UNKNOWN_CPP_NAME '('
+ {
+ /* This could potentially be a an argument defined
+ lookup function (Koenig). */
+ write_exp_elt_opcode (OP_ADL_FUNC);
+ write_exp_elt_block (expression_context_block);
+ write_exp_elt_sym (NULL); /* Placeholder. */
+ write_exp_string ($1.stoken);
+ write_exp_elt_opcode (OP_ADL_FUNC);
+
+ /* 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_len++; }
;
+exp : exp '(' nonempty_typelist ')' const_or_volatile
+ { int i;
+ write_exp_elt_opcode (TYPE_INSTANCE);
+ write_exp_elt_longcst ((LONGEST) $<ivec>3[0]);
+ for (i = 0; i < $<ivec>3[0]; ++i)
+ write_exp_elt_type ($<tvec>3[i + 1]);
+ write_exp_elt_longcst((LONGEST) $<ivec>3[0]);
+ write_exp_elt_opcode (TYPE_INSTANCE);
+ free ($3);
+ }
+ ;
+
rcurly : '}'
{ $$ = end_arglist () - 1; }
;
write_exp_elt_opcode (OP_LONG); }
;
+exp : CHAR
+ {
+ struct stoken_vector vec;
+ vec.len = 1;
+ vec.tokens = &$1;
+ write_exp_string_vector ($1.type, &vec);
+ }
+ ;
+
exp : NAME_OR_INT
{ YYSTYPE val;
parse_number ($1.stoken.ptr, $1.stoken.length, 0, &val);
;
exp : VARIABLE
- /* Already written by write_dollar_variable. */
+ {
+ write_dollar_variable ($1);
+ }
;
exp : SIZEOF '(' type ')' %prec UNARY
write_exp_elt_opcode (OP_LONG); }
;
+exp : REINTERPRET_CAST '<' type '>' '(' exp ')' %prec UNARY
+ { write_exp_elt_opcode (UNOP_REINTERPRET_CAST);
+ write_exp_elt_type ($3);
+ write_exp_elt_opcode (UNOP_REINTERPRET_CAST); }
+ ;
+
+exp : STATIC_CAST '<' type '>' '(' exp ')' %prec UNARY
+ { write_exp_elt_opcode (UNOP_CAST);
+ write_exp_elt_type ($3);
+ write_exp_elt_opcode (UNOP_CAST); }
+ ;
+
+exp : DYNAMIC_CAST '<' type '>' '(' exp ')' %prec UNARY
+ { write_exp_elt_opcode (UNOP_DYNAMIC_CAST);
+ write_exp_elt_type ($3);
+ write_exp_elt_opcode (UNOP_DYNAMIC_CAST); }
+ ;
+
+exp : CONST_CAST '<' type '>' '(' exp ')' %prec UNARY
+ { /* We could do more error checking here, but
+ it doesn't seem worthwhile. */
+ write_exp_elt_opcode (UNOP_CAST);
+ write_exp_elt_type ($3);
+ write_exp_elt_opcode (UNOP_CAST); }
+ ;
+
string_exp:
STRING
{
string. Note that we follow the
NUL-termination convention of the
lexer. */
- $$.length = $1.length;
- $$.ptr = malloc ($1.length + 1);
- memcpy ($$.ptr, $1.ptr, $1.length + 1);
+ struct typed_stoken *vec = XNEW (struct typed_stoken);
+ $$.len = 1;
+ $$.tokens = vec;
+
+ vec->type = $1.type;
+ vec->length = $1.length;
+ vec->ptr = malloc ($1.length + 1);
+ memcpy (vec->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);
- $$ = t;
+ char *p;
+ ++$$.len;
+ $$.tokens = realloc ($$.tokens,
+ $$.len * sizeof (struct typed_stoken));
+
+ p = malloc ($2.length + 1);
+ memcpy (p, $2.ptr, $2.length + 1);
+
+ $$.tokens[$$.len - 1].type = $2.type;
+ $$.tokens[$$.len - 1].length = $2.length;
+ $$.tokens[$$.len - 1].ptr = p;
}
;
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)
+ {
+ int i;
+ enum c_string_type type = C_STRING;
+
+ for (i = 0; i < $1.len; ++i)
{
- 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);
+ switch ($1.tokens[i].type)
+ {
+ case C_STRING:
+ break;
+ case C_WIDE_STRING:
+ case C_STRING_16:
+ case C_STRING_32:
+ if (type != C_STRING
+ && type != $1.tokens[i].type)
+ error ("Undefined string concatenation.");
+ type = $1.tokens[i].type;
+ break;
+ default:
+ /* internal error */
+ internal_error (__FILE__, __LINE__,
+ "unrecognized type in string concatenation");
+ }
}
- 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);
+
+ write_exp_string_vector (type, &$1);
+ for (i = 0; i < $1.len; ++i)
+ free ($1.tokens[i].ptr);
+ free ($1.tokens);
}
;
write_exp_elt_opcode (OP_VAR_VALUE); }
;
-qualified_name: typebase COLONCOLON name
+qualified_name: TYPENAME COLONCOLON name
{
- struct type *type = $1;
+ struct type *type = $1.type;
+ CHECK_TYPEDEF (type);
if (TYPE_CODE (type) != TYPE_CODE_STRUCT
&& TYPE_CODE (type) != TYPE_CODE_UNION
&& TYPE_CODE (type) != TYPE_CODE_NAMESPACE)
write_exp_string ($3);
write_exp_elt_opcode (OP_SCOPE);
}
- | typebase COLONCOLON '~' name
+ | TYPENAME COLONCOLON '~' name
{
- struct type *type = $1;
+ struct type *type = $1.type;
struct stoken tmp_token;
+ CHECK_TYPEDEF (type);
if (TYPE_CODE (type) != TYPE_CODE_STRUCT
&& TYPE_CODE (type) != TYPE_CODE_UNION
&& TYPE_CODE (type) != TYPE_CODE_NAMESPACE)
write_exp_string (tmp_token);
write_exp_elt_opcode (OP_SCOPE);
}
+ | TYPENAME COLONCOLON name COLONCOLON name
+ {
+ char *copy = copy_name ($3);
+ error (_("No type \"%s\" within class "
+ "or namespace \"%s\"."),
+ copy, TYPE_NAME ($1.type));
+ }
;
variable: qualified_name
- | COLONCOLON name
+ | COLONCOLON name_not_typename
{
- char *name = copy_name ($2);
+ char *name = copy_name ($2.stoken);
struct symbol *sym;
struct minimal_symbol *msymbol;
{
if (symbol_read_needs_frame (sym))
{
- if (innermost_block == 0 ||
- contained_in (block_found,
- innermost_block))
+ if (innermost_block == 0
+ || contained_in (block_found,
+ innermost_block))
innermost_block = block_found;
}
/* 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))
+ 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);
{ $$ = lookup_enum (copy_name ($2),
expression_context_block); }
| UNSIGNED typename
- { $$ = lookup_unsigned_typename (TYPE_NAME($2.type)); }
+ { $$ = lookup_unsigned_typename (parse_language,
+ parse_gdbarch,
+ TYPE_NAME($2.type)); }
| UNSIGNED
{ $$ = parse_type->builtin_unsigned_int; }
| SIGNED_KEYWORD typename
- { $$ = lookup_signed_typename (TYPE_NAME($2.type)); }
+ { $$ = lookup_signed_typename (parse_language,
+ parse_gdbarch,
+ TYPE_NAME($2.type)); }
| SIGNED_KEYWORD
{ $$ = parse_type->builtin_int; }
/* It appears that this rule for templates is never
{ $$ = follow_types ($2); }
| typebase const_or_volatile_or_space_identifier_noopt
{ $$ = 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));
-
- $$ = new_type;
- }
;
typename: TYPENAME
{ push_type (tp_volatile); }
;
+operator: OPERATOR NEW
+ { $$ = operator_stoken (" new"); }
+ | OPERATOR DELETE
+ { $$ = operator_stoken (" delete"); }
+ | OPERATOR NEW '[' ']'
+ { $$ = operator_stoken (" new[]"); }
+ | OPERATOR DELETE '[' ']'
+ { $$ = operator_stoken (" delete[]"); }
+ | OPERATOR '+'
+ { $$ = operator_stoken ("+"); }
+ | OPERATOR '-'
+ { $$ = operator_stoken ("-"); }
+ | OPERATOR '*'
+ { $$ = operator_stoken ("*"); }
+ | OPERATOR '/'
+ { $$ = operator_stoken ("/"); }
+ | OPERATOR '%'
+ { $$ = operator_stoken ("%"); }
+ | OPERATOR '^'
+ { $$ = operator_stoken ("^"); }
+ | OPERATOR '&'
+ { $$ = operator_stoken ("&"); }
+ | OPERATOR '|'
+ { $$ = operator_stoken ("|"); }
+ | OPERATOR '~'
+ { $$ = operator_stoken ("~"); }
+ | OPERATOR '!'
+ { $$ = operator_stoken ("!"); }
+ | OPERATOR '='
+ { $$ = operator_stoken ("="); }
+ | OPERATOR '<'
+ { $$ = operator_stoken ("<"); }
+ | OPERATOR '>'
+ { $$ = operator_stoken (">"); }
+ | OPERATOR ASSIGN_MODIFY
+ { const char *op = "unknown";
+ switch ($2)
+ {
+ case BINOP_RSH:
+ op = ">>=";
+ break;
+ case BINOP_LSH:
+ op = "<<=";
+ break;
+ case BINOP_ADD:
+ op = "+=";
+ break;
+ case BINOP_SUB:
+ op = "-=";
+ break;
+ case BINOP_MUL:
+ op = "*=";
+ break;
+ case BINOP_DIV:
+ op = "/=";
+ break;
+ case BINOP_REM:
+ op = "%=";
+ break;
+ case BINOP_BITWISE_IOR:
+ op = "|=";
+ break;
+ case BINOP_BITWISE_AND:
+ op = "&=";
+ break;
+ case BINOP_BITWISE_XOR:
+ op = "^=";
+ break;
+ default:
+ break;
+ }
+
+ $$ = operator_stoken (op);
+ }
+ | OPERATOR LSH
+ { $$ = operator_stoken ("<<"); }
+ | OPERATOR RSH
+ { $$ = operator_stoken (">>"); }
+ | OPERATOR EQUAL
+ { $$ = operator_stoken ("=="); }
+ | OPERATOR NOTEQUAL
+ { $$ = operator_stoken ("!="); }
+ | OPERATOR LEQ
+ { $$ = operator_stoken ("<="); }
+ | OPERATOR GEQ
+ { $$ = operator_stoken (">="); }
+ | OPERATOR ANDAND
+ { $$ = operator_stoken ("&&"); }
+ | OPERATOR OROR
+ { $$ = operator_stoken ("||"); }
+ | OPERATOR INCREMENT
+ { $$ = operator_stoken ("++"); }
+ | OPERATOR DECREMENT
+ { $$ = operator_stoken ("--"); }
+ | OPERATOR ','
+ { $$ = operator_stoken (","); }
+ | OPERATOR ARROW_STAR
+ { $$ = operator_stoken ("->*"); }
+ | OPERATOR ARROW
+ { $$ = operator_stoken ("->"); }
+ | OPERATOR '(' ')'
+ { $$ = operator_stoken ("()"); }
+ | OPERATOR '[' ']'
+ { $$ = operator_stoken ("[]"); }
+ | OPERATOR ptype
+ { char *name;
+ long length;
+ struct ui_file *buf = mem_fileopen ();
+
+ c_print_type ($2, NULL, buf, -1, 0);
+ name = ui_file_xstrdup (buf, &length);
+ ui_file_delete (buf);
+ $$ = operator_stoken (name);
+ free (name);
+ }
+ ;
+
+
+
name : NAME { $$ = $1.stoken; }
| BLOCKNAME { $$ = $1.stoken; }
| TYPENAME { $$ = $1.stoken; }
| NAME_OR_INT { $$ = $1.stoken; }
+ | UNKNOWN_CPP_NAME { $$ = $1.stoken; }
+ | operator { $$ = $1; }
;
name_not_typename : NAME
context where only a name could occur, this might be useful.
| NAME_OR_INT
*/
+ | operator
+ {
+ $$.stoken = $1;
+ $$.sym = lookup_symbol ($1.ptr,
+ expression_context_block,
+ VAR_DOMAIN,
+ &$$.is_a_field_of_this);
+ }
+ | UNKNOWN_CPP_NAME
;
%%
+/* Returns a stoken of the operator name given by OP (which does not
+ include the string "operator"). */
+static struct stoken
+operator_stoken (const char *op)
+{
+ static const char *operator_string = "operator";
+ struct stoken st = { NULL, 0 };
+ st.length = strlen (operator_string) + strlen (op);
+ st.ptr = malloc (st.length + 1);
+ strcpy (st.ptr, operator_string);
+ strcat (st.ptr, op);
+
+ /* The toplevel (c_parse) will free the memory allocated here. */
+ make_cleanup (free, st.ptr);
+ return st;
+};
+
/* 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;
+parse_number (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. */
p[len - 2] = '\0';
putithere->typed_val_decfloat.type
= parse_type->builtin_decfloat;
- decimal_from_string (putithere->typed_val_decfloat.val, 4, p);
+ decimal_from_string (putithere->typed_val_decfloat.val, 4,
+ gdbarch_byte_order (parse_gdbarch), p);
p[len - 2] = 'd';
return DECFLOAT;
}
p[len - 2] = '\0';
putithere->typed_val_decfloat.type
= parse_type->builtin_decdouble;
- decimal_from_string (putithere->typed_val_decfloat.val, 8, p);
+ decimal_from_string (putithere->typed_val_decfloat.val, 8,
+ gdbarch_byte_order (parse_gdbarch), p);
p[len - 2] = 'd';
return DECFLOAT;
}
p[len - 2] = '\0';
putithere->typed_val_decfloat.type
= parse_type->builtin_declong;
- decimal_from_string (putithere->typed_val_decfloat.val, 16, p);
+ decimal_from_string (putithere->typed_val_decfloat.val, 16,
+ gdbarch_byte_order (parse_gdbarch), p);
p[len - 2] = 'd';
return DECFLOAT;
}
}
break;
+ case 'b':
+ case 'B':
+ if (len >= 3)
+ {
+ p += 2;
+ base = 2;
+ len -= 2;
+ }
+ break;
+
case 't':
case 'T':
case 'd':
return INT;
}
+/* Temporary obstack used for holding strings. */
+static struct obstack tempbuf;
+static int tempbuf_init;
+
+/* Parse a C escape sequence. The initial backslash of the sequence
+ is at (*PTR)[-1]. *PTR will be updated to point to just after the
+ last character of the sequence. If OUTPUT is not NULL, the
+ translated form of the escape sequence will be written there. If
+ OUTPUT is NULL, no output is written and the call will only affect
+ *PTR. If an escape sequence is expressed in target bytes, then the
+ entire sequence will simply be copied to OUTPUT. Return 1 if any
+ character was emitted, 0 otherwise. */
+
+int
+c_parse_escape (char **ptr, struct obstack *output)
+{
+ char *tokptr = *ptr;
+ int result = 1;
+
+ /* Some escape sequences undergo character set conversion. Those we
+ translate here. */
+ switch (*tokptr)
+ {
+ /* Hex escapes do not undergo character set conversion, so keep
+ the escape sequence for later. */
+ case 'x':
+ if (output)
+ obstack_grow_str (output, "\\x");
+ ++tokptr;
+ if (!isxdigit (*tokptr))
+ error (_("\\x escape without a following hex digit"));
+ while (isxdigit (*tokptr))
+ {
+ if (output)
+ obstack_1grow (output, *tokptr);
+ ++tokptr;
+ }
+ break;
+
+ /* Octal escapes do not undergo character set conversion, so
+ keep the escape sequence for later. */
+ case '0':
+ case '1':
+ case '2':
+ case '3':
+ case '4':
+ case '5':
+ case '6':
+ case '7':
+ {
+ int i;
+ if (output)
+ obstack_grow_str (output, "\\");
+ for (i = 0;
+ i < 3 && isdigit (*tokptr) && *tokptr != '8' && *tokptr != '9';
+ ++i)
+ {
+ if (output)
+ obstack_1grow (output, *tokptr);
+ ++tokptr;
+ }
+ }
+ break;
+
+ /* We handle UCNs later. We could handle them here, but that
+ would mean a spurious error in the case where the UCN could
+ be converted to the target charset but not the host
+ charset. */
+ case 'u':
+ case 'U':
+ {
+ char c = *tokptr;
+ int i, len = c == 'U' ? 8 : 4;
+ if (output)
+ {
+ obstack_1grow (output, '\\');
+ obstack_1grow (output, *tokptr);
+ }
+ ++tokptr;
+ if (!isxdigit (*tokptr))
+ error (_("\\%c escape without a following hex digit"), c);
+ for (i = 0; i < len && isxdigit (*tokptr); ++i)
+ {
+ if (output)
+ obstack_1grow (output, *tokptr);
+ ++tokptr;
+ }
+ }
+ break;
+
+ /* We must pass backslash through so that it does not
+ cause quoting during the second expansion. */
+ case '\\':
+ if (output)
+ obstack_grow_str (output, "\\\\");
+ ++tokptr;
+ break;
+
+ /* Escapes which undergo conversion. */
+ case 'a':
+ if (output)
+ obstack_1grow (output, '\a');
+ ++tokptr;
+ break;
+ case 'b':
+ if (output)
+ obstack_1grow (output, '\b');
+ ++tokptr;
+ break;
+ case 'f':
+ if (output)
+ obstack_1grow (output, '\f');
+ ++tokptr;
+ break;
+ case 'n':
+ if (output)
+ obstack_1grow (output, '\n');
+ ++tokptr;
+ break;
+ case 'r':
+ if (output)
+ obstack_1grow (output, '\r');
+ ++tokptr;
+ break;
+ case 't':
+ if (output)
+ obstack_1grow (output, '\t');
+ ++tokptr;
+ break;
+ case 'v':
+ if (output)
+ obstack_1grow (output, '\v');
+ ++tokptr;
+ break;
+
+ /* GCC extension. */
+ case 'e':
+ if (output)
+ obstack_1grow (output, HOST_ESCAPE_CHAR);
+ ++tokptr;
+ break;
+
+ /* Backslash-newline expands to nothing at all. */
+ case '\n':
+ ++tokptr;
+ result = 0;
+ break;
+
+ /* A few escapes just expand to the character itself. */
+ case '\'':
+ case '\"':
+ case '?':
+ /* GCC extensions. */
+ case '(':
+ case '{':
+ case '[':
+ case '%':
+ /* Unrecognized escapes turn into the character itself. */
+ default:
+ if (output)
+ obstack_1grow (output, *tokptr);
+ ++tokptr;
+ break;
+ }
+ *ptr = tokptr;
+ return result;
+}
+
+/* Parse a string or character literal from TOKPTR. The string or
+ character may be wide or unicode. *OUTPTR is set to just after the
+ end of the literal in the input string. The resulting token is
+ stored in VALUE. This returns a token value, either STRING or
+ CHAR, depending on what was parsed. *HOST_CHARS is set to the
+ number of host characters in the literal. */
+static int
+parse_string_or_char (char *tokptr, char **outptr, struct typed_stoken *value,
+ int *host_chars)
+{
+ int quote;
+ enum c_string_type type;
+
+ /* Build the gdb internal form of the input string in tempbuf. 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 */
+
+ if (!tempbuf_init)
+ tempbuf_init = 1;
+ else
+ obstack_free (&tempbuf, NULL);
+ obstack_init (&tempbuf);
+
+ /* Record the string type. */
+ if (*tokptr == 'L')
+ {
+ type = C_WIDE_STRING;
+ ++tokptr;
+ }
+ else if (*tokptr == 'u')
+ {
+ type = C_STRING_16;
+ ++tokptr;
+ }
+ else if (*tokptr == 'U')
+ {
+ type = C_STRING_32;
+ ++tokptr;
+ }
+ else
+ type = C_STRING;
+
+ /* Skip the quote. */
+ quote = *tokptr;
+ if (quote == '\'')
+ type |= C_CHAR;
+ ++tokptr;
+
+ *host_chars = 0;
+
+ while (*tokptr)
+ {
+ char c = *tokptr;
+ if (c == '\\')
+ {
+ ++tokptr;
+ *host_chars += c_parse_escape (&tokptr, &tempbuf);
+ }
+ else if (c == quote)
+ break;
+ else
+ {
+ obstack_1grow (&tempbuf, c);
+ ++tokptr;
+ /* FIXME: this does the wrong thing with multi-byte host
+ characters. We could use mbrlen here, but that would
+ make "set host-charset" a bit less useful. */
+ ++*host_chars;
+ }
+ }
+
+ if (*tokptr != quote)
+ {
+ if (quote == '"')
+ error ("Unterminated string in expression.");
+ else
+ error ("Unmatched single quote.");
+ }
+ ++tokptr;
+
+ value->type = type;
+ value->ptr = obstack_base (&tempbuf);
+ value->length = obstack_object_size (&tempbuf);
+
+ *outptr = tokptr;
+
+ return quote == '"' ? STRING : CHAR;
+}
+
struct token
{
char *operator;
int token;
enum exp_opcode opcode;
+ int cxx_only;
};
static const struct token tokentab3[] =
{
- {">>=", ASSIGN_MODIFY, BINOP_RSH},
- {"<<=", ASSIGN_MODIFY, BINOP_LSH}
+ {">>=", ASSIGN_MODIFY, BINOP_RSH, 0},
+ {"<<=", ASSIGN_MODIFY, BINOP_LSH, 0},
+ {"->*", ARROW_STAR, BINOP_END, 1}
};
static const struct token tokentab2[] =
{
- {"+=", ASSIGN_MODIFY, BINOP_ADD},
- {"-=", ASSIGN_MODIFY, BINOP_SUB},
- {"*=", ASSIGN_MODIFY, BINOP_MUL},
- {"/=", ASSIGN_MODIFY, BINOP_DIV},
- {"%=", ASSIGN_MODIFY, BINOP_REM},
- {"|=", ASSIGN_MODIFY, BINOP_BITWISE_IOR},
- {"&=", ASSIGN_MODIFY, BINOP_BITWISE_AND},
- {"^=", ASSIGN_MODIFY, BINOP_BITWISE_XOR},
- {"++", INCREMENT, BINOP_END},
- {"--", DECREMENT, BINOP_END},
- {"->", ARROW, BINOP_END},
- {"&&", ANDAND, BINOP_END},
- {"||", OROR, BINOP_END},
- {"::", COLONCOLON, BINOP_END},
- {"<<", LSH, BINOP_END},
- {">>", RSH, BINOP_END},
- {"==", EQUAL, BINOP_END},
- {"!=", NOTEQUAL, BINOP_END},
- {"<=", LEQ, BINOP_END},
- {">=", GEQ, BINOP_END}
+ {"+=", 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},
+ /* "::" is *not* only C++: gdb overrides its meaning in several
+ different ways, e.g., 'filename'::func, function::variable. */
+ {"::", 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},
+ {".*", DOT_STAR, BINOP_END, 1}
+ };
+
+/* 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},
+ {"new", NEW, OP_NULL, 1},
+ {"delete", DELETE, OP_NULL, 1},
+ {"operator", OPERATOR, OP_NULL, 1},
+
+ {"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},
+
+ {"const_cast", CONST_CAST, OP_NULL, 1 },
+ {"dynamic_cast", DYNAMIC_CAST, OP_NULL, 1 },
+ {"static_cast", STATIC_CAST, OP_NULL, 1 },
+ {"reinterpret_cast", REINTERPRET_CAST, OP_NULL, 1 }
};
/* When we find that lexptr (the global var defined in parse.c) is
obstack_free (&expansion_obstack, NULL);
}
+/* Return true iff the token represents a C++ cast operator. */
+
+static int
+is_cast_operator (const char *token, int len)
+{
+ return (! strncmp (token, "dynamic_cast", len)
+ || ! strncmp (token, "static_cast", len)
+ || ! strncmp (token, "reinterpret_cast", len)
+ || ! strncmp (token, "const_cast", len));
+}
/* The scope used for macro expansion. */
static struct macro_scope *expression_macro_scope;
/* Read one token, getting characters through lexptr. */
static int
-yylex ()
+lex_one_token (void)
{
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;
for (i = 0; i < sizeof tokentab3 / sizeof tokentab3[0]; i++)
if (strncmp (tokstart, tokentab3[i].operator, 3) == 0)
{
+ if (tokentab3[i].cxx_only
+ && parse_language->la_language != language_cplus)
+ break;
+
lexptr += 3;
yylval.opcode = tokentab3[i].opcode;
return tokentab3[i].token;
for (i = 0; i < sizeof tokentab2 / sizeof tokentab2[0]; i++)
if (strncmp (tokstart, tokentab2[i].operator, 2) == 0)
{
+ if (tokentab2[i].cxx_only
+ && parse_language->la_language != language_cplus)
+ break;
+
lexptr += 2;
yylval.opcode = tokentab2[i].opcode;
if (in_parse_field && tokentab2[i].token == ARROW)
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 '[':
case '(':
paren_depth++;
lexptr++;
return c;
+ case ']':
case ')':
if (paren_depth == 0)
return 0;
case '@':
case '<':
case '>':
- case '[':
- case ']':
case '?':
case ':':
case '=':
lexptr++;
return c;
+ case 'L':
+ case 'u':
+ case 'U':
+ if (tokstart[1] != '"' && tokstart[1] != '\'')
+ break;
+ /* Fall through. */
+ case '\'':
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)
+ {
+ int host_len;
+ int result = parse_string_or_char (tokstart, &lexptr, &yylval.tsval,
+ &host_len);
+ if (result == CHAR)
{
- case '\0':
- case '"':
- /* Do nothing, loop will terminate. */
- break;
- case '\\':
- tokptr++;
- c = parse_escape (&tokptr);
- if (c == -1)
+ if (host_len == 0)
+ error ("Empty character constant.");
+ else if (host_len > 2 && c == '\'')
{
- continue;
+ ++tokstart;
+ namelen = lexptr - tokstart - 1;
+ goto tryname;
}
- 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;
+ else if (host_len > 1)
+ error ("Invalid character constant.");
}
- } 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);
+ return result;
+ }
}
if (!(c == '_' || c == '$'
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. */
+ {
+ if (! is_cast_operator (tokstart, namelen))
+ {
+ /* 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;
+ char * p = find_template_name_end (tokstart + namelen);
+ if (p)
+ namelen = p - tokstart;
+ }
+ break;
}
c = tokstart[++namelen];
}
return 0;
}
+ /* For the same reason (breakpoint conditions), "thread N"
+ terminates the expression. "thread" could be an identifier, but
+ an identifier is never followed by a number without intervening
+ punctuation. "task" is similar. Handle abbreviations of these,
+ similarly to breakpoint.c:find_condition_and_thread. */
+ if (namelen >= 1
+ && (strncmp (tokstart, "thread", namelen) == 0
+ || strncmp (tokstart, "task", namelen) == 0)
+ && (tokstart[namelen] == ' ' || tokstart[namelen] == '\t')
+ && ! scanning_macro_expansion ())
+ {
+ char *p = tokstart + namelen + 1;
+ while (*p == ' ' || *p == '\t')
+ p++;
+ if (*p >= '0' && *p <= '9')
+ return 0;
+ }
+
lexptr += namelen;
tryname:
- /* Catch specific keywords. Should be done with a data structure. */
- switch (namelen)
- {
- case 8:
- if (strncmp (tokstart, "unsigned", 8) == 0)
- return UNSIGNED;
- if (parse_language->la_language == language_cplus
- && strncmp (tokstart, "template", 8) == 0)
- return TEMPLATE;
- if (strncmp (tokstart, "volatile", 8) == 0)
- return VOLATILE_KEYWORD;
- break;
- case 6:
- if (strncmp (tokstart, "struct", 6) == 0)
- return STRUCT;
- if (strncmp (tokstart, "signed", 6) == 0)
- return SIGNED_KEYWORD;
- if (strncmp (tokstart, "sizeof", 6) == 0)
- return SIZEOF;
- if (strncmp (tokstart, "double", 6) == 0)
- return DOUBLE_KEYWORD;
- break;
- case 5:
- if (parse_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 (strncmp (tokstart, "short", 5) == 0)
- return SHORT;
- if (strncmp (tokstart, "const", 5) == 0)
- return CONST_KEYWORD;
- break;
- case 4:
- if (strncmp (tokstart, "enum", 4) == 0)
- return ENUM;
- if (strncmp (tokstart, "long", 4) == 0)
- return LONG;
- if (parse_language->la_language == language_cplus)
- {
- if (strncmp (tokstart, "true", 4) == 0)
- return TRUEKEYWORD;
- }
- break;
- case 3:
- if (strncmp (tokstart, "int", 3) == 0)
- return INT_KEYWORD;
- break;
- default:
- break;
- }
-
yylval.sval.ptr = tokstart;
yylval.sval.length = namelen;
- if (*tokstart == '$')
- {
- write_dollar_variable (yylval.sval);
- return VARIABLE;
- }
-
- /* Use token-type BLOCKNAME for symbols that happen to be defined as
- functions or symtabs. If this is not so, then ...
- Use token-type TYPENAME for symbols that happen to be defined
- currently as names of types; NAME for other symbols.
- The caller is not constrained to care about the distinction. */
- {
- char *tmp = copy_name (yylval.sval);
- struct symbol *sym;
- int is_a_field_of_this = 0;
- int hextype;
-
- sym = lookup_symbol (tmp, expression_context_block,
- VAR_DOMAIN,
- parse_language->la_language == language_cplus
- ? &is_a_field_of_this : (int *) NULL);
- /* Call lookup_symtab, not lookup_partial_symtab, in case there are
- no psymtabs (coff, xcoff, or some future change to blow away the
- psymtabs once once symbols are read). */
- if (sym && SYMBOL_CLASS (sym) == LOC_BLOCK)
+ /* Catch specific keywords. */
+ copy = copy_name (yylval.sval);
+ for (i = 0; i < sizeof ident_tokens / sizeof ident_tokens[0]; i++)
+ if (strcmp (copy, ident_tokens[i].operator) == 0)
{
- yylval.ssym.sym = sym;
- yylval.ssym.is_a_field_of_this = is_a_field_of_this;
- return BLOCKNAME;
+ if (ident_tokens[i].cxx_only
+ && parse_language->la_language != language_cplus)
+ break;
+
+ /* It is ok to always set this, even though we don't always
+ strictly need to. */
+ yylval.opcode = ident_tokens[i].opcode;
+ return ident_tokens[i].token;
}
- else if (!sym)
- { /* See if it's a file name. */
- struct symtab *symtab;
- symtab = lookup_symtab (tmp);
+ if (*tokstart == '$')
+ return VARIABLE;
- if (symtab)
- {
- yylval.bval = BLOCKVECTOR_BLOCK (BLOCKVECTOR (symtab), STATIC_BLOCK);
- return FILENAME;
- }
- }
+ if (in_parse_field && *lexptr == '\0')
+ saw_name_at_eof = 1;
+ return NAME;
+}
- 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, tmp);
- if (yylval.tsym.type != NULL)
+/* An object of this type is pushed on a FIFO by the "outer" lexer. */
+typedef struct
+{
+ int token;
+ YYSTYPE value;
+} token_and_value;
+
+DEF_VEC_O (token_and_value);
+
+/* A FIFO of tokens that have been read but not yet returned to the
+ parser. */
+static VEC (token_and_value) *token_fifo;
+
+/* Non-zero if the lexer should return tokens from the FIFO. */
+static int popping;
+
+/* Temporary storage for c_lex; this holds symbol names as they are
+ built up. */
+static struct obstack name_obstack;
+
+/* Classify a NAME token. The contents of the token are in `yylval'.
+ Updates yylval and returns the new token type. BLOCK is the block
+ in which lookups start; this can be NULL to mean the global
+ scope. */
+static int
+classify_name (struct block *block)
+{
+ struct symbol *sym;
+ char *copy;
+ int is_a_field_of_this = 0;
+
+ copy = copy_name (yylval.sval);
+
+ sym = lookup_symbol (copy, block, VAR_DOMAIN,
+ parse_language->la_language == language_cplus
+ ? &is_a_field_of_this : (int *) NULL);
+
+ 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)
+ {
+ yylval.tsym.type = SYMBOL_TYPE (sym);
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;
- }
- }
+ 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
+ && ((copy[0] >= 'a' && copy[0] < 'a' + input_radix - 10)
+ || (copy[0] >= 'A' && copy[0] < 'A' + input_radix - 10)))
+ {
+ YYSTYPE newlval; /* Its value is ignored. */
+ int hextype = parse_number (copy, yylval.sval.length, 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 (sym == NULL
+ && parse_language->la_language == language_cplus
+ && !lookup_minimal_symbol (copy, NULL, NULL))
+ return UNKNOWN_CPP_NAME;
+
+ return NAME;
+}
+
+/* Like classify_name, but used by the inner loop of the lexer, when a
+ name might have already been seen. FIRST_NAME is true if the token
+ in `yylval' is the first component of a name, false otherwise. If
+ this function returns NAME, it might not have updated `yylval'.
+ This is ok because the caller only cares about TYPENAME. */
+static int
+classify_inner_name (struct block *block, int first_name)
+{
+ struct type *type, *new_type;
+ char *copy;
+
+ if (first_name)
+ return classify_name (block);
- /* 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;
+ type = check_typedef (yylval.tsym.type);
+ if (TYPE_CODE (type) != TYPE_CODE_STRUCT
+ && TYPE_CODE (type) != TYPE_CODE_UNION
+ && TYPE_CODE (type) != TYPE_CODE_NAMESPACE)
+ /* We know the caller won't expect us to update yylval. */
return NAME;
- }
+
+ copy = copy_name (yylval.tsym.stoken);
+ new_type = cp_lookup_nested_type (type, copy, block);
+
+ if (new_type == NULL)
+ /* We know the caller won't expect us to update yylval. */
+ return NAME;
+
+ yylval.tsym.type = new_type;
+ return TYPENAME;
+}
+
+/* The outer level of a two-level lexer. This calls the inner lexer
+ to return tokens. It then either returns these tokens, or
+ aggregates them into a larger token. This lets us work around a
+ problem in our parsing approach, where the parser could not
+ distinguish between qualified names and qualified types at the
+ right point.
+
+ This approach is still not ideal, because it mishandles template
+ types. See the comment in lex_one_token for an example. However,
+ this is still an improvement over the earlier approach, and will
+ suffice until we move to better parsing technology. */
+static int
+yylex (void)
+{
+ token_and_value current;
+ int first_was_coloncolon, last_was_coloncolon, first_iter;
+
+ if (popping && !VEC_empty (token_and_value, token_fifo))
+ {
+ token_and_value tv = *VEC_index (token_and_value, token_fifo, 0);
+ VEC_ordered_remove (token_and_value, token_fifo, 0);
+ yylval = tv.value;
+ return tv.token;
+ }
+ popping = 0;
+
+ current.token = lex_one_token ();
+ if (current.token == NAME)
+ current.token = classify_name (expression_context_block);
+ if (parse_language->la_language != language_cplus
+ || (current.token != TYPENAME && current.token != COLONCOLON))
+ return current.token;
+
+ first_was_coloncolon = current.token == COLONCOLON;
+ last_was_coloncolon = first_was_coloncolon;
+ obstack_free (&name_obstack, obstack_base (&name_obstack));
+ if (!last_was_coloncolon)
+ obstack_grow (&name_obstack, yylval.sval.ptr, yylval.sval.length);
+ current.value = yylval;
+ first_iter = 1;
+ while (1)
+ {
+ token_and_value next;
+
+ next.token = lex_one_token ();
+ next.value = yylval;
+
+ if (next.token == NAME && last_was_coloncolon)
+ {
+ int classification;
+
+ classification = classify_inner_name (first_was_coloncolon
+ ? NULL
+ : expression_context_block,
+ first_iter);
+ /* We keep going until we either run out of names, or until
+ we have a qualified name which is not a type. */
+ if (classification != TYPENAME)
+ {
+ /* Push the final component and leave the loop. */
+ VEC_safe_push (token_and_value, token_fifo, &next);
+ break;
+ }
+
+ /* Update the partial name we are constructing. */
+ if (!first_iter)
+ {
+ /* We don't want to put a leading "::" into the name. */
+ obstack_grow_str (&name_obstack, "::");
+ }
+ obstack_grow (&name_obstack, next.value.sval.ptr,
+ next.value.sval.length);
+
+ yylval.sval.ptr = obstack_base (&name_obstack);
+ yylval.sval.length = obstack_object_size (&name_obstack);
+ current.value = yylval;
+ current.token = classification;
+
+ last_was_coloncolon = 0;
+ }
+ else if (next.token == COLONCOLON && !last_was_coloncolon)
+ last_was_coloncolon = 1;
+ else
+ {
+ /* We've reached the end of the name. */
+ VEC_safe_push (token_and_value, token_fifo, &next);
+ break;
+ }
+
+ first_iter = 0;
+ }
+
+ popping = 1;
+
+ /* If we ended with a "::", insert it too. */
+ if (last_was_coloncolon)
+ {
+ token_and_value cc;
+ memset (&cc, 0, sizeof (token_and_value));
+ if (first_was_coloncolon)
+ {
+ yylval = cc.value;
+ return COLONCOLON;
+ }
+ cc.token = COLONCOLON;
+ VEC_safe_insert (token_and_value, token_fifo, 0, &cc);
+ }
+
+ yylval = current.value;
+ yylval.sval.ptr = obstack_copy0 (&expansion_obstack,
+ yylval.sval.ptr,
+ yylval.sval.length);
+ return current.token;
}
int
gdb_assert (! macro_original_text);
make_cleanup (scan_macro_cleanup, 0);
+ make_cleanup_restore_integer (&yydebug);
+ yydebug = parser_debug;
+
/* Initialize some state used by the lexer. */
last_was_structop = 0;
saw_name_at_eof = 0;
+ VEC_free (token_and_value, token_fifo);
+ popping = 0;
+ obstack_init (&name_obstack);
+ make_cleanup_obstack_free (&name_obstack);
+
result = yyparse ();
do_cleanups (back_to);
return result;
void
-yyerror (msg)
- char *msg;
+yyerror (char *msg)
{
if (prev_lexptr)
lexptr = prev_lexptr;