/* FLEX lexer for Ada expressions, for GDB.
- Copyright (C) 1994, 1997, 1998, 2000, 2001, 2002, 2003
- Free Software Foundation, Inc.
+ Copyright (C) 1994-2019 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/>. */
/*----------------------------------------------------------------------*/
%{
+#include "diagnostics.h"
+
+/* Some old versions of flex generate code that uses the "register" keyword,
+ which clang warns about. This was observed for example with flex 2.5.35,
+ as shipped with macOS 10.12. */
+DIAGNOSTIC_PUSH
+DIAGNOSTIC_IGNORE_DEPRECATED_REGISTER
+
#define NUMERAL_WIDTH 256
#define LONGEST_SIGN ((ULONGEST) 1 << (sizeof(LONGEST) * HOST_CHAR_BIT - 1))
static char numbuf[NUMERAL_WIDTH];
static void canonicalizeNumeral (char *s1, const char *);
static struct stoken processString (const char*, int);
-static int processInt (const char *, const char *, const char *);
-static int processReal (const char *);
+static int processInt (struct parser_state *, const char *, const char *,
+ const char *);
+static int processReal (struct parser_state *, const char *);
static struct stoken processId (const char *, int);
static int processAttribute (const char *);
static int find_dot_all (const char *);
+static void rewind_to_char (int);
#undef YY_DECL
#define YY_DECL static int yylex ( void )
+/* Flex generates a static function "input" which is not used.
+ Defining YY_NO_INPUT comments it out. */
+#define YY_NO_INPUT
+
#undef YY_INPUT
#define YY_INPUT(BUF, RESULT, MAX_SIZE) \
- if ( *lexptr == '\000' ) \
+ if ( *pstate->lexptr == '\000' ) \
(RESULT) = YY_NULL; \
else \
{ \
- *(BUF) = *lexptr; \
+ *(BUF) = *pstate->lexptr; \
(RESULT) = 1; \
- lexptr += 1; \
+ pstate->lexptr += 1; \
}
static int find_dot_all (const char *);
+/* Depth of parentheses. */
+static int paren_depth;
+
%}
%option case-insensitive interactive nodefault
{NUM10}{POSEXP} {
canonicalizeNumeral (numbuf, yytext);
- return processInt (NULL, numbuf, strrchr(numbuf, 'e')+1);
+ return processInt (pstate, NULL, numbuf,
+ strrchr (numbuf, 'e') + 1);
}
{NUM10} {
canonicalizeNumeral (numbuf, yytext);
- return processInt (NULL, numbuf, NULL);
+ return processInt (pstate, NULL, numbuf, NULL);
}
{NUM10}"#"{HEXDIG}({HEXDIG}|_)*"#"{POSEXP} {
canonicalizeNumeral (numbuf, yytext);
- return processInt (numbuf,
+ return processInt (pstate, numbuf,
strchr (numbuf, '#') + 1,
strrchr(numbuf, '#') + 1);
}
{NUM10}"#"{HEXDIG}({HEXDIG}|_)*"#" {
canonicalizeNumeral (numbuf, yytext);
- return processInt (numbuf, strchr (numbuf, '#') + 1, NULL);
+ return processInt (pstate, numbuf, strchr (numbuf, '#') + 1,
+ NULL);
}
"0x"{HEXDIG}+ {
canonicalizeNumeral (numbuf, yytext+2);
- return processInt ("16#", numbuf, NULL);
+ return processInt (pstate, "16#", numbuf, NULL);
}
{NUM10}"."{NUM10}{EXP} {
canonicalizeNumeral (numbuf, yytext);
- return processReal (numbuf);
+ return processReal (pstate, numbuf);
}
{NUM10}"."{NUM10} {
canonicalizeNumeral (numbuf, yytext);
- return processReal (numbuf);
+ return processReal (pstate, numbuf);
}
{NUM10}"#"{NUM16}"."{NUM16}"#"{EXP} {
}
<INITIAL>"'"({GRAPHIC}|\")"'" {
- yylval.typed_val.type = type_char ();
+ yylval.typed_val.type = type_char (pstate);
yylval.typed_val.val = yytext[1];
return CHARLIT;
}
<INITIAL>"'[\""{HEXDIG}{2}"\"]'" {
int v;
- yylval.typed_val.type = type_char ();
+ yylval.typed_val.type = type_char (pstate);
sscanf (yytext+3, "%2x", &v);
yylval.typed_val.val = v;
return CHARLIT;
if {
- while (*lexptr != 'i' && *lexptr != 'I')
- lexptr -= 1;
- yyrestart(NULL);
+ rewind_to_char ('i');
+ return 0;
+ }
+
+task {
+ rewind_to_char ('t');
+ return 0;
+ }
+
+thread{WHITE}+{DIG} {
+ /* This keyword signals the end of the expression and
+ will be processed separately. */
+ rewind_to_char ('t');
return 0;
}
then { return THEN; }
xor { return XOR; }
+ /* BOOLEAN "KEYWORDS" */
+
+ /* True and False are not keywords in Ada, but rather enumeration constants.
+ However, the boolean type is no longer represented as an enum, so True
+ and False are no longer defined in symbol tables. We compromise by
+ making them keywords (when bare). */
+
+true { return TRUEKEYWORD; }
+false { return FALSEKEYWORD; }
+
/* ATTRIBUTES */
-{TICK}[a-zA-Z][a-zA-Z]+ { return processAttribute (yytext+1); }
+{TICK}[a-zA-Z][a-zA-Z]+ { BEGIN INITIAL; return processAttribute (yytext+1); }
/* PUNCTUATION */
[-&*+./:<>=|;\[\]] { return yytext[0]; }
-"," { if (paren_depth == 0 && comma_terminates)
+"," { if (paren_depth == 0 && pstate->comma_terminates)
{
- lexptr -= 1;
- yyrestart(NULL);
+ rewind_to_char (',');
return 0;
}
else
"(" { paren_depth += 1; return '('; }
")" { if (paren_depth == 0)
{
- lexptr -= 1;
- yyrestart(NULL);
+ rewind_to_char (')');
return 0;
}
else
"$"({LETTER}|{DIG}|"$")* {
yylval.sval.ptr = yytext;
yylval.sval.length = yyleng;
- return SPECIAL_VARIABLE;
+ return DOLLAR_VARIABLE;
}
/* CATCH-ALL ERROR CASE */
%%
#include <ctype.h>
-#include "gdb_string.h"
-
/* Initialize the lexer for processing new expression. */
-void
+static void
lexer_init (FILE *inp)
{
BEGIN INITIAL;
+ paren_depth = 0;
yyrestart (inp);
}
s1[0] = '\000';
}
-#define HIGH_BYTE_POSN ((sizeof (ULONGEST) - 1) * HOST_CHAR_BIT)
-
-/* True (non-zero) iff DIGIT is a valid digit in radix BASE,
- where 2 <= BASE <= 16. */
-
-static int
-is_digit_in_base (unsigned char digit, int base)
-{
- if (!isxdigit (digit))
- return 0;
- if (base <= 10)
- return (isdigit (digit) && digit < base + '0');
- else
- return (isdigit (digit) || tolower (digit) < base - 10 + 'a');
-}
-
-static int
-digit_to_int (unsigned char c)
-{
- if (isdigit (c))
- return c - '0';
- else
- return tolower (c) - 'a' + 10;
-}
-
-/* As for strtoul, but for ULONGEST results. */
-
-ULONGEST
-strtoulst (const char *num, const char **trailer, int base)
-{
- unsigned int high_part;
- ULONGEST result;
- int i;
- unsigned char lim;
-
- if (base < 2 || base > 16)
- {
- errno = EINVAL;
- return 0;
- }
- lim = base - 1 + '0';
-
- result = high_part = 0;
- for (i = 0; is_digit_in_base (num[i], base); i += 1)
- {
- result = result*base + digit_to_int (num[i]);
- high_part = high_part*base + (unsigned int) (result >> HIGH_BYTE_POSN);
- result &= ((ULONGEST) 1 << HIGH_BYTE_POSN) - 1;
- if (high_part > 0xff)
- {
- errno = ERANGE;
- result = high_part = 0;
- break;
- }
- }
-
- if (trailer != NULL)
- *trailer = &num[i];
-
- return result + ((ULONGEST) high_part << HIGH_BYTE_POSN);
-}
-
/* Interprets the prefix of NUM that consists of digits of the given BASE
as an integer of that BASE, with the string EXP as an exponent.
Puts value in yylval, and returns INT, if the string is valid. Causes
*/
static int
-processInt (const char *base0, const char *num0, const char *exp0)
+processInt (struct parser_state *par_state, const char *base0,
+ const char *num0, const char *exp0)
{
ULONGEST result;
long exp;
int base;
-
- char *trailer;
+ const char *trailer;
if (base0 == NULL)
base = 10;
exp = strtol(exp0, (char **) NULL, 10);
errno = 0;
- result = strtoulst (num0, (const char **) &trailer, base);
+ result = strtoulst (num0, &trailer, base);
if (errno == ERANGE)
error (_("Integer literal out of range"));
if (isxdigit(*trailer))
exp -= 1;
}
- if ((result >> (TARGET_INT_BIT-1)) == 0)
- yylval.typed_val.type = type_int ();
- else if ((result >> (TARGET_LONG_BIT-1)) == 0)
- yylval.typed_val.type = type_long ();
- else if (((result >> (TARGET_LONG_BIT-1)) >> 1) == 0)
+ if ((result >> (gdbarch_int_bit (par_state->gdbarch ())-1)) == 0)
+ yylval.typed_val.type = type_int (par_state);
+ else if ((result >> (gdbarch_long_bit (par_state->gdbarch ())-1)) == 0)
+ yylval.typed_val.type = type_long (par_state);
+ else if (((result >> (gdbarch_long_bit (par_state->gdbarch ())-1)) >> 1) == 0)
{
/* We have a number representable as an unsigned integer quantity.
For consistency with the C treatment, we will treat it as an
for the mess, but C doesn't officially guarantee that a simple
assignment does the trick (no, it doesn't; read the reference manual).
*/
- yylval.typed_val.type = builtin_type_unsigned_long;
+ yylval.typed_val.type
+ = builtin_type (par_state->gdbarch ())->builtin_unsigned_long;
if (result & LONGEST_SIGN)
yylval.typed_val.val =
(LONGEST) (result & ~LONGEST_SIGN)
return INT;
}
else
- yylval.typed_val.type = type_long_long ();
+ yylval.typed_val.type = type_long_long (par_state);
yylval.typed_val.val = (LONGEST) result;
return INT;
}
static int
-processReal (const char *num0)
+processReal (struct parser_state *par_state, const char *num0)
{
- sscanf (num0, DOUBLEST_SCAN_FORMAT, &yylval.typed_val_float.dval);
-
- yylval.typed_val_float.type = type_float ();
- if (sizeof(DOUBLEST) >= TARGET_DOUBLE_BIT / TARGET_CHAR_BIT)
- yylval.typed_val_float.type = type_double ();
- if (sizeof(DOUBLEST) >= TARGET_LONG_DOUBLE_BIT / TARGET_CHAR_BIT)
- yylval.typed_val_float.type = type_long_double ();
+ yylval.typed_val_float.type = type_long_double (par_state);
+ bool parsed = parse_float (num0, strlen (num0),
+ yylval.typed_val_float.type,
+ yylval.typed_val_float.val);
+ gdb_assert (parsed);
return FLOAT;
}
/* Store a canonicalized version of NAME0[0..LEN-1] in yylval.ssym. The
- resulting string is valid until the next call to ada_parse. It differs
+ resulting string is valid until the next call to ada_parse. If
+ NAME0 contains the substring "___", it is assumed to be already
+ encoded and the resulting name is equal to it. Similarly, if the name
+ starts with '<', it is copied verbatim. Otherwise, it differs
from NAME0 in that:
- + Characters between '...' or <...> are transfered verbatim to
- yylval.ssym.
- + <, >, and trailing "'" characters in quoted sequences are removed
- (a leading quote is preserved to indicate that the name is not to be
- GNAT-encoded).
+ + Characters between '...' are transfered verbatim to yylval.ssym.
+ + Trailing "'" characters in quoted sequences are removed (a leading quote is
+ preserved to indicate that the name is not to be GNAT-encoded).
+ Unquoted whitespace is removed.
+ Unquoted alphabetic characters are mapped to lower case.
Result is returned as a struct stoken, but for convenience, the string
static struct stoken
processId (const char *name0, int len)
{
- char *name = obstack_alloc (&temp_parse_space, len + 11);
+ char *name = (char *) obstack_alloc (&temp_parse_space, len + 11);
int i0, i;
struct stoken result;
+ result.ptr = name;
while (len > 0 && isspace (name0[len-1]))
len -= 1;
+
+ if (name0[0] == '<' || strstr (name0, "___") != NULL)
+ {
+ strncpy (name, name0, len);
+ name[len] = '\000';
+ result.length = len;
+ return result;
+ }
+
i = i0 = 0;
while (i0 < len)
{
while (i0 < len && name0[i0] != '\'');
i0 += 1;
break;
- case '<':
- i0 += 1;
- while (i0 < len && name0[i0] != '>')
- {
- name[i] = name0[i0];
- i += 1; i0 += 1;
- }
- i0 += 1;
- break;
}
}
name[i] = '\000';
- result.ptr = name;
result.length = i;
return result;
}
const char *lim = text + len;
struct stoken result;
- q = result.ptr = obstack_alloc (&temp_parse_space, len);
+ q = (char *) obstack_alloc (&temp_parse_space, len);
+ result.ptr = q;
p = text;
while (p < lim)
{
find_dot_all (const char *str)
{
int i;
- for (i = 0; str[i] != '\000'; i += 1)
- {
- if (str[i] == '.')
- {
- int i0 = i;
- do
- i += 1;
- while (isspace (str[i]));
- if (strncmp (str+i, "all", 3) == 0
- && ! isalnum (str[i+3]) && str[i+3] != '_')
- return i0;
- }
- }
+
+ for (i = 0; str[i] != '\000'; i++)
+ if (str[i] == '.')
+ {
+ int i0 = i;
+
+ do
+ i += 1;
+ while (isspace (str[i]));
+
+ if (strncasecmp (str + i, "all", 3) == 0
+ && !isalnum (str[i + 3]) && str[i + 3] != '_')
+ return i0;
+ }
return -1;
}
return attributes[k].code;
}
+/* Back up lexptr by yyleng and then to the rightmost occurrence of
+ character CH, case-folded (there must be one). WARNING: since
+ lexptr points to the next input character that Flex has not yet
+ transferred to its internal buffer, the use of this function
+ depends on the assumption that Flex calls YY_INPUT only when it is
+ logically necessary to do so (thus, there is no reading ahead
+ farther than needed to identify the next token.) */
+
+static void
+rewind_to_char (int ch)
+{
+ pstate->lexptr -= yyleng;
+ while (toupper (*pstate->lexptr) != toupper (ch))
+ pstate->lexptr -= 1;
+ yyrestart (NULL);
+}
+
int
yywrap(void)
{
{
(dummy_function) yyunput
};
+
+DIAGNOSTIC_POP