PHP 源碼探秘 - 為什么 trim 會導致亂碼
作者：佚名
本文通過對PHP源碼的探秘，揭示了 trim 會導致亂碼的原因。
運行以下代碼：
$tag = '互聯網產品、'; $text = rtrim($tag, "、"); print_r($text);
運行，我們可能以為會得到的結果是 互聯網產品 ，實際結果是 互聯網產� 。為什么會這樣呢？
原理
trim 函數文檔
string trim ( string $str [, string $character_mask = " \t\n\r\0\x0B" ] )
該函數不是多字節函數，也就是說，漢字這樣的多字節字符，會拿其頭或尾的單字節來和后面的 $character_mask 對應的char數組進行匹配，如果在后面的數組中，則刪掉，繼續匹配。比如：
echo ltrim("bcdf","abc"); // df
如下面的 demo 中的函數 string_print_char 所示：
、 由 0xe3 0x80 0x81 三字節組成，
品 由 0xe5 0x93 0x81 三字節組成。
所以在執行 rtrim 的時候，通過字節比對，會將 0x81 去掉，導致了***出現了亂碼。
源碼精簡版演示
查看 PHP7 的源碼，然后提煉出下面的小 demo ，方便大家一起學習，其實PHP源碼的學習并不難，每天進步一點點。
//
// main.c
// trim
//
// Created by 周夢康 on 2017/10/18.
// Copyright © 2017年 周夢康. All rights reserved.
//
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
void string_print_char(char *str);
void php_charmask(unsigned char *input, size_t len, char *mask);
char *ltrim(char *str,char *character_mask);
char *rtrim(char *str,char *character_mask);
int main(int argc, char const *argv[])
{
printf("%s\n",ltrim("bcdf","abc"));
string_print_char("品"); // e5 93 81
string_print_char("、"); // e3 80 81
printf("%s\n",rtrim("互聯網產品、","、"));
return 0;
}
char *ltrim(char *str,char *character_mask)
{
char *res;
char mask[256];
register size_t i;
int trimmed = 0;
size_t len = strlen(str);
php_charmask((unsigned char*)character_mask, strlen(character_mask), mask);
for (i = 0; i < len; i++) {
if (mask[(unsigned char)str[i]]) {
trimmed++;
} else {
break;
}
}
len -= trimmed;
str += trimmed;
res = (char *) malloc(sizeof(char) * (len+1));
memcpy(res,str,len);
return res;
}
char *rtrim(char *str,char *character_mask)
{
char *res;
char mask[256];
register size_t i;
size_t len = strlen(str);
php_charmask((unsigned char*)character_mask, strlen(character_mask), mask);
if (len > 0) {
i = len - 1;
do {
if (mask[(unsigned char)str[i]]) {
len--;
} else {
break;
}
} while (i-- != 0);
}
res = (char *) malloc(sizeof(char) * (len+1));
memcpy(res,str,len);
return res;
}
void string_print_char(char *str)
{
unsigned long l = strlen(str);
for (int i=0; i < l; i++) {
printf("%02hhx\t",str[i]);
}
printf("\n");
}
void php_charmask(unsigned char *input, size_t len, char *mask)
{
unsigned char *end;
unsigned char c;
memset(mask, 0, 256);
for (end = input+len; input < end; input++) {
c = *input;
mask[c]= 1;
}
}
PHP7 相關源碼
PHP_FUNCTION(trim)
{
php_do_trim(INTERNAL_FUNCTION_PARAM_PASSTHRU, 3);
}
PHP_FUNCTION(rtrim)
{
php_do_trim(INTERNAL_FUNCTION_PARAM_PASSTHRU, 2);
}
PHP_FUNCTION(ltrim)
{
php_do_trim(INTERNAL_FUNCTION_PARAM_PASSTHRU, 1);
}
static void php_do_trim(INTERNAL_FUNCTION_PARAMETERS, int mode)
{
zend_string *str;
zend_string *what = NULL;
ZEND_PARSE_PARAMETERS_START(1, 2)
Z_PARAM_STR(str)
Z_PARAM_OPTIONAL
Z_PARAM_STR(what)
ZEND_PARSE_PARAMETERS_END();
ZVAL_STR(return_value, php_trim(str, (what ? ZSTR_VAL(what) : NULL), (what ? ZSTR_LEN(what) : 0), mode));
}
PHPAPI zend_string *php_trim(zend_string *str, char *what, size_t what_len, int mode)
{
const char *c = ZSTR_VAL(str);
size_t len = ZSTR_LEN(str);
register size_t i;
size_t trimmed = 0;
char mask[256];
if (what) {
if (what_len == 1) {
char p = *what;
if (mode & 1) {
for (i = 0; i < len; i++) {
if (c[i] == p) {
trimmed++;
} else {
break;
}
}
len -= trimmed;
c += trimmed;
}
if (mode & 2) {
if (len > 0) {
i = len - 1;
do {
if (c[i] == p) {
len--;
} else {
break;
}
} while (i-- != 0);
}
}
} else {
php_charmask((unsigned char*)what, what_len, mask);
if (mode & 1) {
for (i = 0; i < len; i++) {
if (mask[(unsigned char)c[i]]) {
trimmed++;
} else {
break;
}
}
len -= trimmed;
c += trimmed;
}
if (mode & 2) {
if (len > 0) {
i = len - 1;
do {
if (mask[(unsigned char)c[i]]) {
len--;
} else {
break;
}
} while (i-- != 0);
}
}
}
} else {
if (mode & 1) {
for (i = 0; i < len; i++) {
if ((unsigned char)c[i] <= ' ' &&
(c[i] == ' ' || c[i] == '\n' || c[i] == '\r' || c[i] == '\t' || c[i] == '\v' || c[i] == '\0')) {
trimmed++;
} else {
break;
}
}
len -= trimmed;
c += trimmed;
}
if (mode & 2) {
if (len > 0) {
i = len - 1;
do {
if ((unsigned char)c[i] <= ' ' &&
(c[i] == ' ' || c[i] == '\n' || c[i] == '\r' || c[i] == '\t' || c[i] == '\v' || c[i] == '\0')) {
len--;
} else {
break;
}
} while (i-- != 0);
}
}
}
if (ZSTR_LEN(str) == len) {
return zend_string_copy(str);
} else {
return zend_string_init(c, len, 0);
}
}
/* {{{ php_charmask
* Fills a 256-byte bytemask with input. You can specify a range like 'a..z',
* it needs to be incrementing.
* Returns: FAILURE/SUCCESS whether the input was correct (i.e. no range errors)
*/
static inline int php_charmask(unsigned char *input, size_t len, char *mask)
{
unsigned char *end;
unsigned char c;
int result = SUCCESS;
memset(mask, 0, 256);
for (end = input+len; input < end; input++) {
c=*input;
if ((input+3 < end) && input[1] == '.' && input[2] == '.'
&& input[3] >= c) {
memset(mask+c, 1, input[3] - c + 1);
input+=3;
} else if ((input+1 < end) && input[0] == '.' && input[1] == '.') {
/* Error, try to be as helpful as possible:
(a range ending/starting with '.' won't be captured here) */
if (end-len >= input) { /* there was no 'left' char */
php_error_docref(NULL, E_WARNING, "Invalid '..'-range, no character to the left of '..'");
result = FAILURE;
continue;
}
if (input+2 >= end) { /* there is no 'right' char */
php_error_docref(NULL, E_WARNING, "Invalid '..'-range, no character to the right of '..'");
result = FAILURE;
continue;
}
if (input[-1] > input[2]) { /* wrong order */
php_error_docref(NULL, E_WARNING, "Invalid '..'-range, '..'-range needs to be incrementing");
result = FAILURE;
continue;
}
/* FIXME: better error (a..b..c is the only left possibility?) */
php_error_docref(NULL, E_WARNING, "Invalid '..'-range");
result = FAILURE;
continue;
} else {
mask[c]=1;
}
}
return result;
}
/* }}} */
責任編輯：張燕妮
來源：
周夢康的博客
