Skip to content

optimize unpack() for named fields #6958

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
May 14, 2021
Merged

optimize unpack() for named fields #6958

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
7c95b78
optimize unpack() for named fields
kaja47 May 7, 2021
b312d97
further improvements
kaja47 May 13, 2021
4094495
remove memory leak and make code little bit nicer
kaja47 May 13, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
76 changes: 41 additions & 35 deletions ext/standard/pack.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -739,32 +739,32 @@ PHP_FUNCTION(unpack)
while (formatlen-- > 0) {
char type = *(format++);
char c;
int arg = 1, argb;
int repetitions = 1, argb;
char *name;
int namelen;
int size=0;
int size = 0;

/* Handle format arguments if any */
if (formatlen > 0) {
c = *format;

if (c >= '0' && c <= '9') {
arg = atoi(format);
repetitions = atoi(format);

while (formatlen > 0 && *format >= '0' && *format <= '9') {
format++;
formatlen--;
}
} else if (c == '*') {
arg = -1;
repetitions = -1;
format++;
formatlen--;
}
}

/* Get of new value in array */
name = format;
argb = arg;
argb = repetitions;

while (formatlen > 0 && *format != '/') {
formatlen--;
Expand All @@ -780,9 +780,9 @@ PHP_FUNCTION(unpack)
/* Never use any input */
case 'X':
size = -1;
if (arg < 0) {
if (repetitions < 0) {
php_error_docref(NULL, E_WARNING, "Type %c: '*' ignored", type);
arg = 1;
repetitions = 1;
}
break;

Expand All @@ -793,14 +793,14 @@ PHP_FUNCTION(unpack)
case 'a':
case 'A':
case 'Z':
size = arg;
arg = 1;
size = repetitions;
repetitions = 1;
break;

case 'h':
case 'H':
size = (arg > 0) ? (arg + (arg % 2)) / 2 : arg;
arg = 1;
size = (repetitions > 0) ? (repetitions + (repetitions % 2)) / 2 : repetitions;
repetitions = 1;
break;

/* Use 1 byte of input */
Expand Down Expand Up @@ -871,16 +871,21 @@ PHP_FUNCTION(unpack)
}

/* Do actual unpacking */
for (i = 0; i != arg; i++ ) {
for (i = 0; i != repetitions; i++ ) {
/* Space for name + number, safe as namelen is ensured <= 200 */
char n[256];
char name_buffer[256];

if (arg != 1 || namelen == 0) {
/* Need to add element number to name */
snprintf(n, sizeof(n), "%.*s%d", namelen, name, i + 1);
char* real_name;
size_t real_name_length;

if (repetitions == 1 && namelen > 0) {
/* Use a part of the formatarg argument directly as the name. */
real_name_length = namelen;
real_name = name;
} else {
/* Truncate name to next format code or end of string */
snprintf(n, sizeof(n), "%.*s", namelen, name);
/* Need to add the 1-based element number to the name */
real_name_length = snprintf(name_buffer, sizeof(name_buffer), "%.*s%d", namelen, name, i + 1);
real_name = name_buffer;
}

if (size != 0 && size != -1 && INT_MAX - size + 1 < inputpos) {
Expand All @@ -902,7 +907,7 @@ PHP_FUNCTION(unpack)

size = len;

add_assoc_stringl(return_value, n, &input[inputpos], len);
add_assoc_stringl_ex(return_value, real_name, real_name_length, &input[inputpos], len);
break;
}
case 'A': {
Expand All @@ -928,7 +933,7 @@ PHP_FUNCTION(unpack)
break;
}

add_assoc_stringl(return_value, n, &input[inputpos], len + 1);
add_assoc_stringl_ex(return_value, real_name, real_name_length, &input[inputpos], len + 1);
break;
}
/* New option added for Z to remain in-line with the Perl implementation */
Expand All @@ -952,7 +957,7 @@ PHP_FUNCTION(unpack)
}
len = s;

add_assoc_stringl(return_value, n, &input[inputpos], len);
add_assoc_stringl_ex(return_value, real_name, real_name_length, &input[inputpos], len);
break;
}

Expand Down Expand Up @@ -995,15 +1000,15 @@ PHP_FUNCTION(unpack)
}

ZSTR_VAL(buf)[len] = '\0';
add_assoc_str(return_value, n, buf);
add_assoc_str_ex(return_value, real_name, real_name_length, buf);
break;
}

case 'c': /* signed */
case 'C': { /* unsigned */
uint8_t x = input[inputpos];
zend_long v = (type == 'c') ? (int8_t) x : x;
add_assoc_long(return_value, n, v);
add_assoc_long_ex(return_value, real_name, real_name_length, v);
break;
}

Expand All @@ -1022,15 +1027,15 @@ PHP_FUNCTION(unpack)
v = x;
}

add_assoc_long(return_value, n, v);
add_assoc_long_ex(return_value, real_name, real_name_length, v);
break;
}

case 'i': /* signed integer, machine size, machine endian */
case 'I': { /* unsigned integer, machine size, machine endian */
unsigned int x = *((unaligned_uint*) &input[inputpos]);
zend_long v = (type == 'i') ? (int) x : x;
add_assoc_long(return_value, n, v);
add_assoc_long_ex(return_value, real_name, real_name_length, v);
break;
}

Expand All @@ -1049,7 +1054,8 @@ PHP_FUNCTION(unpack)
v = x;
}

add_assoc_long(return_value, n, v);
add_assoc_long_ex(return_value, real_name, real_name_length, v);

break;
}

Expand All @@ -1069,7 +1075,7 @@ PHP_FUNCTION(unpack)
v = x;
}

add_assoc_long(return_value, n, v);
add_assoc_long_ex(return_value, real_name, real_name_length, v);
break;
}
#endif
Expand All @@ -1088,7 +1094,7 @@ PHP_FUNCTION(unpack)
memcpy(&v, &input[inputpos], sizeof(float));
}

add_assoc_double(return_value, n, (double)v);
add_assoc_double_ex(return_value, real_name, real_name_length, (double)v);
break;
}

Expand All @@ -1105,7 +1111,7 @@ PHP_FUNCTION(unpack)
} else {
memcpy(&v, &input[inputpos], sizeof(double));
}
add_assoc_double(return_value, n, v);
add_assoc_double_ex(return_value, real_name, real_name_length, v);
break;
}

Expand All @@ -1116,22 +1122,22 @@ PHP_FUNCTION(unpack)
case 'X':
if (inputpos < size) {
inputpos = -size;
i = arg - 1; /* Break out of for loop */
i = repetitions - 1; /* Break out of for loop */

if (arg >= 0) {
if (repetitions >= 0) {
php_error_docref(NULL, E_WARNING, "Type %c: outside of string", type);
}
}
break;

case '@':
if (arg <= inputlen) {
inputpos = arg;
if (repetitions <= inputlen) {
inputpos = repetitions;
} else {
php_error_docref(NULL, E_WARNING, "Type %c: outside of string", type);
}

i = arg - 1; /* Done, break out of for loop */
i = repetitions - 1; /* Done, break out of for loop */
break;
}

Expand All @@ -1142,7 +1148,7 @@ PHP_FUNCTION(unpack)
}
inputpos = 0;
}
} else if (arg < 0) {
} else if (repetitions < 0) {
/* Reached end of input for '*' repeater */
break;
} else {
Expand Down
42 changes: 42 additions & 0 deletions ext/standard/tests/strings/pack_arrays.phpt
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,42 @@
--TEST--
test unpack() to array with named keys
--FILE--
<?php
$str = pack('VVV', 0x00010203, 0x04050607, 0x08090a0b);
print_r(unpack('Vaa/Vbb/Vcc', $str));
print_r(unpack('V2aa/Vcc', $str));
print_r(unpack('V3aa', $str));
print_r(unpack('V*aa', $str));
print_r(unpack('V*', $str));
?>
--EXPECT--
Array
(
[aa] => 66051
[bb] => 67438087
[cc] => 134810123
)
Array
(
[aa1] => 66051
[aa2] => 67438087
[cc] => 134810123
)
Array
(
[aa1] => 66051
[aa2] => 67438087
[aa3] => 134810123
)
Array
(
[aa1] => 66051
[aa2] => 67438087
[aa3] => 134810123
)
Array
(
[1] => 66051
[2] => 67438087
[3] => 134810123
)