Skip to content

Commit 589bdf3

Browse files
GirgiasGirgias
committed
Implement Explicit octal notation for integers RFC
RFC: https://wiki.php.net/rfc/explicit_octal_notation Add an extensive test suits for other variants of integer literals Closes GH-6360
1 parent ab9f497 commit 589bdf3

20 files changed

+872
-4
lines changed

‎UPGRADING‎

Lines changed: 14 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -65,6 +65,11 @@ PHP 8.1 UPGRADE NOTES
6565
2. New Features
6666
========================================
6767

68+
- Core:
69+
. It is now possible to specify octal integer by using the explicit "0o"/"0O"
70+
prefix similar to hexadecimal ("0x"/"0X) and binary ("0b"/"0B") integer literals
71+
RFC: https://wiki.php.net/rfc/explicit_octal_notation
72+
6873
- hash:
6974
. The following functions have changed signatures:
7075

@@ -101,6 +106,15 @@ PHP 8.1 UPGRADE NOTES
101106
5. Changed Functions
102107
========================================
103108

109+
- Filter:
110+
. The FILTER_FLAG_ALLOW_OCTAL flag of the FILTER_VALIDATE_INT filter now accept
111+
octal string with the leading octal prefix ("0o"/"0O")
112+
RFC: https://wiki.php.net/rfc/explicit_octal_notation
113+
114+
- GMP:
115+
. All GMP function now accept octal string with the leading octal prefix ("0o"/"0O")
116+
RFC: https://wiki.php.net/rfc/explicit_octal_notation
117+
104118
========================================
105119
6. New Functions
106120
========================================

‎Zend/zend_language_scanner.l‎

Lines changed: 47 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1366,6 +1366,7 @@ DNUM ({LNUM}?"."{LNUM})|({LNUM}"."{LNUM}?)
13661366
EXPONENT_DNUM (({LNUM}|{DNUM})[eE][+-]?{LNUM})
13671367
HNUM "0x"[0-9a-fA-F]+(_[0-9a-fA-F]+)*
13681368
BNUM "0b"[01]+(_[01]+)*
1369+
ONUM "0o"[0-7]+(_[0-7]+)*
13691370
LABEL [a-zA-Z_\x80-\xff][a-zA-Z0-9_\x80-\xff]*
13701371
WHITESPACE [ \n\r\t]+
13711372
TABS_AND_SPACES [ \t]*
@@ -1950,6 +1951,51 @@ NEWLINE ("\r"|"\n"|"\r\n")
19501951
}
19511952
}
19521953

1954+
<ST_IN_SCRIPTING>{ONUM} {
1955+
/* The +/- 2 skips "0o" */
1956+
size_t len = yyleng - 2;
1957+
char *end, *octal = yytext + 2;
1958+
zend_bool contains_underscores = (memchr(octal, '_', len) != NULL);
1959+
1960+
/* Skip any leading 0s */
1961+
while (len > 0 && (*octal == '0' || *octal == '_')) {
1962+
++octal;
1963+
--len;
1964+
}
1965+
1966+
if (contains_underscores) {
1967+
octal = estrndup(octal, len);
1968+
strip_underscores(octal, &len);
1969+
}
1970+
1971+
errno = 0;
1972+
1973+
ZVAL_LONG(zendlval, ZEND_STRTOL(octal, &end, 8));
1974+
1975+
ZEND_ASSERT(end == octal + len);
1976+
1977+
if (!errno) {
1978+
if (contains_underscores) {
1979+
efree(octal);
1980+
}
1981+
RETURN_TOKEN_WITH_VAL(T_LNUMBER);
1982+
}
1983+
1984+
/* Overflow */
1985+
ZEND_ASSERT(errno == ERANGE);
1986+
/* Reset errno */
1987+
errno = 0;
1988+
1989+
/* zend_oct_strtod skips leading '0' */
1990+
ZVAL_DOUBLE(zendlval, zend_oct_strtod(octal, (const char **)&end));
1991+
ZEND_ASSERT(!errno);
1992+
ZEND_ASSERT(end == octal + len);
1993+
if (contains_underscores) {
1994+
efree(octal);
1995+
}
1996+
RETURN_TOKEN_WITH_VAL(T_DNUMBER);
1997+
}
1998+
19531999
<ST_IN_SCRIPTING>{LNUM} {
19542000
size_t len = yyleng;
19552001
char *end, *lnum = yytext;
@@ -2071,7 +2117,7 @@ string:
20712117
RETURN_TOKEN_WITH_VAL(T_NUM_STRING);
20722118
}
20732119

2074-
<ST_VAR_OFFSET>{LNUM}|{HNUM}|{BNUM} { /* Offset must be treated as a string */
2120+
<ST_VAR_OFFSET>{LNUM}|{HNUM}|{BNUM}|{ONUM} { /* Offset must be treated as a string */
20752121
if (yyleng == 1) {
20762122
ZVAL_INTERNED_STR(zendlval, ZSTR_CHAR((zend_uchar)*(yytext)));
20772123
} else {

‎Zend/zend_strtod.c‎

Lines changed: 0 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -4467,9 +4467,6 @@ ZEND_API double zend_oct_strtod(const char *str, const char **endptr)
44674467
return 0.0;
44684468
}
44694469

4470-
/* skip leading zero */
4471-
s++;
4472-
44734470
while ((c = *s++)) {
44744471
if (c < '0' || c > '7') {
44754472
/* break and return the current value if the number is not well-formed

‎ext/filter/logical_filters.c‎

Lines changed: 7 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -240,6 +240,13 @@ void php_filter_int(PHP_INPUT_FILTER_PARAM_DECL) /* {{{ */
240240
error = 1;
241241
}
242242
} else if (allow_octal) {
243+
/* Support explicit octal prefix notation */
244+
if (*p == 'o' || *p == 'O') {
245+
p++; len--;
246+
if (len == 0) {
247+
RETURN_VALIDATION_FAILED
248+
}
249+
}
243250
if (php_filter_parse_octal(p, len, &ctx_value) < 0) {
244251
error = 1;
245252
}

‎ext/filter/tests/013.phpt‎

Lines changed: 8 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -38,6 +38,11 @@ var_dump(filter_var(345, FILTER_VALIDATE_INT, array("options" => array("min_rang
3838
var_dump(filter_var("0ff", FILTER_VALIDATE_INT));
3939
var_dump(filter_var("010", FILTER_VALIDATE_INT));
4040

41+
// Explicit octal prefix
42+
var_dump(filter_var("0o16", FILTER_VALIDATE_INT, array("flags"=>FILTER_FLAG_ALLOW_OCTAL)));
43+
var_dump(filter_var("0O16", FILTER_VALIDATE_INT, array("flags"=>FILTER_FLAG_ALLOW_OCTAL)));
44+
var_dump(filter_var("0o016", FILTER_VALIDATE_INT, array("flags"=>FILTER_FLAG_ALLOW_OCTAL)));
45+
4146
echo "Done\n";
4247
?>
4348
--EXPECT--
@@ -71,4 +76,7 @@ bool(false)
7176
bool(false)
7277
bool(false)
7378
bool(false)
79+
int(14)
80+
int(14)
81+
int(14)
7482
Done

‎ext/filter/tests/surprising_integer_literals.phpt‎

Lines changed: 36 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,36 @@
1+
--TEST--
2+
Surprising result with integer literals (hex/octal)
3+
--SKIPIF--
4+
<?php if (!extension_loaded("filter")) die("skip"); ?>
5+
--FILE--
6+
<?php
7+
echo 'Hex', \PHP_EOL;
8+
var_dump(filter_var('0x', FILTER_VALIDATE_INT, FILTER_FLAG_ALLOW_HEX));
9+
var_dump(filter_var('0xg', FILTER_VALIDATE_INT, FILTER_FLAG_ALLOW_HEX));
10+
var_dump(filter_var('0X', FILTER_VALIDATE_INT, FILTER_FLAG_ALLOW_HEX));
11+
var_dump(filter_var('0Xg', FILTER_VALIDATE_INT, FILTER_FLAG_ALLOW_HEX));
12+
var_dump(filter_var('', FILTER_VALIDATE_INT, FILTER_FLAG_ALLOW_HEX));
13+
echo 'Octal', \PHP_EOL;
14+
var_dump(filter_var('0o', FILTER_VALIDATE_INT, FILTER_FLAG_ALLOW_OCTAL));
15+
var_dump(filter_var('0og', FILTER_VALIDATE_INT, FILTER_FLAG_ALLOW_OCTAL));
16+
var_dump(filter_var('0O', FILTER_VALIDATE_INT, FILTER_FLAG_ALLOW_OCTAL));
17+
var_dump(filter_var('0Og', FILTER_VALIDATE_INT, FILTER_FLAG_ALLOW_OCTAL));
18+
var_dump(filter_var('O', FILTER_VALIDATE_INT, FILTER_FLAG_ALLOW_OCTAL));
19+
var_dump(filter_var('Og', FILTER_VALIDATE_INT, FILTER_FLAG_ALLOW_OCTAL));
20+
var_dump(filter_var('', FILTER_VALIDATE_INT, FILTER_FLAG_ALLOW_OCTAL));
21+
?>
22+
--EXPECT--
23+
Hex
24+
bool(false)
25+
bool(false)
26+
bool(false)
27+
bool(false)
28+
bool(false)
29+
Octal
30+
bool(false)
31+
bool(false)
32+
bool(false)
33+
bool(false)
34+
bool(false)
35+
bool(false)
36+
bool(false)

‎ext/gmp/gmp.c‎

Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -605,6 +605,9 @@ static zend_result convert_to_gmp(mpz_t gmpnumber, zval *val, zend_long base, ui
605605
if ((base == 0 || base == 16) && (numstr[1] == 'x' || numstr[1] == 'X')) {
606606
base = 16;
607607
skip_lead = 1;
608+
} else if ((base == 0 || base == 8) && (numstr[1] == 'o' || numstr[1] == 'O')) {
609+
base = 8;
610+
skip_lead = 1;
608611
} else if ((base == 0 || base == 2) && (numstr[1] == 'b' || numstr[1] == 'B')) {
609612
base = 2;
610613
skip_lead = 1;

‎ext/gmp/tests/gmp_init_integer_notations.phpt‎

Lines changed: 50 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,50 @@
1+
--TEST--
2+
gmp_init() with various integer notations
3+
--SKIPIF--
4+
<?php if (!extension_loaded("gmp")) print "skip"; ?>
5+
--FILE--
6+
<?php
7+
8+
var_dump(gmp_init("0x16"));
9+
var_dump(gmp_init("0X16"));
10+
var_dump(gmp_init("0o16"));
11+
var_dump(gmp_init("0o16"));
12+
var_dump(gmp_init("016"));
13+
var_dump(gmp_init("016"));
14+
var_dump(gmp_init("0b11"));
15+
var_dump(gmp_init("0b11"));
16+
17+
?>
18+
--EXPECT--
19+
object(GMP)#1 (1) {
20+
["num"]=>
21+
string(2) "22"
22+
}
23+
object(GMP)#1 (1) {
24+
["num"]=>
25+
string(2) "22"
26+
}
27+
object(GMP)#1 (1) {
28+
["num"]=>
29+
string(2) "14"
30+
}
31+
object(GMP)#1 (1) {
32+
["num"]=>
33+
string(2) "14"
34+
}
35+
object(GMP)#1 (1) {
36+
["num"]=>
37+
string(2) "14"
38+
}
39+
object(GMP)#1 (1) {
40+
["num"]=>
41+
string(2) "14"
42+
}
43+
object(GMP)#1 (1) {
44+
["num"]=>
45+
string(1) "3"
46+
}
47+
object(GMP)#1 (1) {
48+
["num"]=>
49+
string(1) "3"
50+
}

‎ext/gmp/tests/surprising_integer_literals.phpt‎

Lines changed: 33 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,33 @@
1+
--TEST--
2+
Surprising result with integer literals (hex/binary/octal)
3+
--SKIPIF--
4+
<?php if (!extension_loaded("gmp")) print "skip"; ?>
5+
--FILE--
6+
<?php
7+
8+
$values = [
9+
'0x',
10+
'0X',
11+
'0b',
12+
'0B',
13+
'0o',
14+
'0O',
15+
''
16+
];
17+
18+
foreach ($values as $value) {
19+
try {
20+
var_dump(gmp_init($value));
21+
} catch (\ValueError $e) {
22+
echo $e->getMessage(), \PHP_EOL;
23+
}
24+
}
25+
?>
26+
--EXPECT--
27+
gmp_init(): Argument #1 ($num) is not an integer string
28+
gmp_init(): Argument #1 ($num) is not an integer string
29+
gmp_init(): Argument #1 ($num) is not an integer string
30+
gmp_init(): Argument #1 ($num) is not an integer string
31+
gmp_init(): Argument #1 ($num) is not an integer string
32+
gmp_init(): Argument #1 ($num) is not an integer string
33+
gmp_init(): Argument #1 ($num) is not an integer string

‎ext/standard/tests/math/base_convert_variation2.phpt‎

Lines changed: 32 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,32 @@
1+
--TEST--
2+
Test base_convert() function : strange literals
3+
--FILE--
4+
<?php
5+
echo 'Binary to decimal:', \PHP_EOL;
6+
var_dump(base_convert('0b', 2, 10));
7+
var_dump(base_convert('0B', 2, 10));
8+
var_dump(base_convert('', 2, 10));
9+
echo 'Octal to decimal:', \PHP_EOL;
10+
var_dump(base_convert('0o', 8, 10));
11+
var_dump(base_convert('0O', 8, 10));
12+
var_dump(base_convert('0', 8, 10));
13+
var_dump(base_convert('', 8, 10));
14+
echo 'Hexadecimal to decimal:', \PHP_EOL;
15+
var_dump(base_convert('0x', 16, 10));
16+
var_dump(base_convert('0X', 16, 10));
17+
var_dump(base_convert('', 16, 10));
18+
?>
19+
--EXPECT--
20+
Binary to decimal:
21+
string(1) "0"
22+
string(1) "0"
23+
string(1) "0"
24+
Octal to decimal:
25+
string(1) "0"
26+
string(1) "0"
27+
string(1) "0"
28+
string(1) "0"
29+
Hexadecimal to decimal:
30+
string(1) "0"
31+
string(1) "0"
32+
string(1) "0"

0 commit comments

Comments
 (0)