1
0
mirror of https://github.com/php/php-src.git synced 2026-03-24 00:02:20 +01:00

Improve performance of urldecode() and rawurldecode()

There are two hot spots on my machines:
1. We copy the string because the internal PHP API works in-place.
2. The conversion of hex characters is slow due to going through the C
   locale handling.

This patch resolves the first hot spots by introducing 2 new internal
APIs that avoid the redundant copy and allocate an empty string upfront.
The second hotspot is resolved by having a specialised htoi handler.

For the following benchmark:
```php
$encoded = "Hello%20World%21+This%20is%20a%20test%3A%20%40%23%24%25%5E%26*%28%29";
for ($i=0;$i<2000000;$i++) {
  rawurldecode($encoded);
  urldecode($encoded);
}
```

On an i7-4790:
```
Benchmark 1: ./sapi/cli/php x.php
  Time (mean ± σ):     364.8 ms ±   3.7 ms    [User: 359.9 ms, System: 3.3 ms]
  Range (min … max):   359.9 ms … 372.0 ms    10 runs

Benchmark 2: ./sapi/cli/php_old x.php
  Time (mean ± σ):     565.5 ms ±   4.9 ms    [User: 561.8 ms, System: 2.5 ms]
  Range (min … max):   560.7 ms … 578.2 ms    10 runs

Summary
  ./sapi/cli/php x.php ran
    1.55 ± 0.02 times faster than ./sapi/cli/php_old x.php
```

On an i7-1185G7:
```
Benchmark 1: ./sapi/cli/php x.php
  Time (mean ± σ):     708.8 ms ±   6.1 ms    [User: 701.4 ms, System: 6.3 ms]
  Range (min … max):   701.9 ms … 722.3 ms    10 runs

Benchmark 2: ./sapi/cli/php_old x.php
  Time (mean ± σ):      1.311 s ±  0.019 s    [User: 1.300 s, System: 0.008 s]
  Range (min … max):    1.281 s …  1.348 s    10 runs

Summary
  ./sapi/cli/php x.php ran
    1.85 ± 0.03 times faster than ./sapi/cli/php_old x.php
```

Closes GH-18378.
This commit is contained in:
Niels Dossche
2025-04-21 13:32:38 +02:00
parent c0da1bc44d
commit 6c0578d31c
4 changed files with 48 additions and 28 deletions

View File

@@ -476,6 +476,7 @@ PHP 8.5 UPGRADE NOTES
- Standard:
. Improved performance of array functions with callbacks
(array_find, array_filter, array_map, usort, ...).
. Improved performance of urlencode() and rawurlencode().
- XMLReader:
. Improved property access performance.

View File

@@ -61,6 +61,10 @@ PHP 8.5 INTERNALS UPGRADE NOTES
is still valid. This is useful when a GC cycle is collected and the
database object can be destroyed prior to destroying the statement.
- ext/standard
. Added php_url_decode_ex() and php_raw_url_decode_ex() that unlike their
non-ex counterparts do not work in-place.
========================
4. OpCode changes
========================

View File

@@ -411,21 +411,24 @@ done:
}
/* }}} */
/* https://stackoverflow.com/questions/34365746/whats-the-fastest-way-to-convert-hex-to-integer-in-c */
static unsigned int php_htoi_single(unsigned char x)
{
ZEND_ASSERT((x >= 'a' && x <= 'f') || (x >= 'A' && x <= 'F') || (x >= '0' && x <= '9'));
return 9 * (x >> 6) + (x & 0xf);
}
/* {{{ php_htoi */
static int php_htoi(char *s)
static int php_htoi(const char *s)
{
int value;
int c;
unsigned char c;
c = ((unsigned char *)s)[0];
if (isupper(c))
c = tolower(c);
value = (c >= '0' && c <= '9' ? c - '0' : c - 'a' + 10) * 16;
value = php_htoi_single(c) * 16;
c = ((unsigned char *)s)[1];
if (isupper(c))
c = tolower(c);
value += c >= '0' && c <= '9' ? c - '0' : c - 'a' + 10;
value += php_htoi_single(c);
return (value);
}
@@ -572,28 +575,27 @@ PHP_FUNCTION(urldecode)
Z_PARAM_STR(in_str)
ZEND_PARSE_PARAMETERS_END();
out_str = zend_string_init(ZSTR_VAL(in_str), ZSTR_LEN(in_str), 0);
ZSTR_LEN(out_str) = php_url_decode(ZSTR_VAL(out_str), ZSTR_LEN(out_str));
out_str = zend_string_alloc(ZSTR_LEN(in_str), false);
ZSTR_LEN(out_str) = php_url_decode_ex(ZSTR_VAL(out_str), ZSTR_VAL(in_str), ZSTR_LEN(in_str));
RETURN_NEW_STR(out_str);
}
/* }}} */
/* {{{ php_url_decode */
PHPAPI size_t php_url_decode(char *str, size_t len)
PHPAPI size_t php_url_decode_ex(char *dest, const char *src, size_t src_len)
{
char *dest = str;
char *data = str;
char *dest_start = dest;
const char *data = src;
while (len--) {
while (src_len--) {
if (*data == '+') {
*dest = ' ';
}
else if (*data == '%' && len >= 2 && isxdigit((int) *(data + 1))
else if (*data == '%' && src_len >= 2 && isxdigit((int) *(data + 1))
&& isxdigit((int) *(data + 2))) {
*dest = (char) php_htoi(data + 1);
data += 2;
len -= 2;
src_len -= 2;
} else {
*dest = *data;
}
@@ -601,7 +603,13 @@ PHPAPI size_t php_url_decode(char *str, size_t len)
dest++;
}
*dest = '\0';
return dest - str;
return dest - dest_start;
}
/* {{{ php_url_decode */
PHPAPI size_t php_url_decode(char *str, size_t len)
{
return php_url_decode_ex(str, str, len);
}
/* }}} */
@@ -634,25 +642,24 @@ PHP_FUNCTION(rawurldecode)
Z_PARAM_STR(in_str)
ZEND_PARSE_PARAMETERS_END();
out_str = zend_string_init(ZSTR_VAL(in_str), ZSTR_LEN(in_str), 0);
ZSTR_LEN(out_str) = php_raw_url_decode(ZSTR_VAL(out_str), ZSTR_LEN(out_str));
out_str = zend_string_alloc(ZSTR_LEN(in_str), false);
ZSTR_LEN(out_str) = php_raw_url_decode_ex(ZSTR_VAL(out_str), ZSTR_VAL(in_str), ZSTR_LEN(in_str));
RETURN_NEW_STR(out_str);
}
/* }}} */
/* {{{ php_raw_url_decode */
PHPAPI size_t php_raw_url_decode(char *str, size_t len)
PHPAPI size_t php_raw_url_decode_ex(char *dest, const char *src, size_t src_len)
{
char *dest = str;
char *data = str;
char *dest_start = dest;
const char *data = src;
while (len--) {
if (*data == '%' && len >= 2 && isxdigit((int) *(data + 1))
while (src_len--) {
if (*data == '%' && src_len >= 2 && isxdigit((int) *(data + 1))
&& isxdigit((int) *(data + 2))) {
*dest = (char) php_htoi(data + 1);
data += 2;
len -= 2;
src_len -= 2;
} else {
*dest = *data;
}
@@ -660,7 +667,13 @@ PHPAPI size_t php_raw_url_decode(char *str, size_t len)
dest++;
}
*dest = '\0';
return dest - str;
return dest - dest_start;
}
/* {{{ php_raw_url_decode */
PHPAPI size_t php_raw_url_decode(char *str, size_t len)
{
return php_raw_url_decode_ex(str, str, len);
}
/* }}} */

View File

@@ -33,7 +33,9 @@ PHPAPI php_url *php_url_parse(char const *str);
PHPAPI php_url *php_url_parse_ex(char const *str, size_t length);
PHPAPI php_url *php_url_parse_ex2(char const *str, size_t length, bool *has_port);
PHPAPI size_t php_url_decode(char *str, size_t len); /* return value: length of decoded string */
PHPAPI size_t php_url_decode_ex(char *dest, const char *src, size_t src_len);
PHPAPI size_t php_raw_url_decode(char *str, size_t len); /* return value: length of decoded string */
PHPAPI size_t php_raw_url_decode_ex(char *dest, const char *src, size_t src_len);
PHPAPI zend_string *php_url_encode(char const *s, size_t len);
PHPAPI zend_string *php_raw_url_encode(char const *s, size_t len);