字符串函数
在线手册:中文 英文
PHP手册

str_replace

(PHP 4, PHP 5)

str_replace子字符串替换

说明

mixed str_replace ( mixed $search , mixed $replace , mixed $subject [, int &$count ] )

该函数返回一个字符串或者数组。该字符串或数组是将 subject 中全部的 search 都被 replace 替换之后的结果。

如果没有一些特殊的替换需求(比如正则表达式),你应该使用该函数替换 ereg_replace()preg_replace()

参数

如果 searchreplace 为数组,那么 str_replace() 将对 subject 做二者的映射替换。如果 replace 的值的个数少于 search 的个数,多余的替换将使用空字符串来进行。如果 search 是一个数组而 replace 是一个字符串,那么 search 中每个元素的替换将始终使用这个字符串。该转换不会改变大小写。

如果 searchreplace 都是数组,它们的值将会被依次处理。

search

查找的目标值,也就是 needle。一个数组可以指定多个目标。

replace

search 的替换值。一个数组可以被用来指定多重替换。

subject

执行替换的数组或者字符串。也就是 haystack

如果 subject 是一个数组,替换操作将遍历整个 subject,返回值也将是一个数组。

count

Note: 如果被指定,它的值将被设置为替换发生的次数。

返回值

该函数返回替换后的数组或者字符串。

更新日志

版本 说明
5.0.0 新增 count 参数。
4.3.3 函数行为改变。旧的版本中存在一个 BUG —— 当 searchreplace 两个参数都是数组的时候,将导致空的 search 索引被跳过,但是却没有同时前移 replace 内部指针。该错误发生在 PHP 4.3.3,任何依赖于此 BUG 的脚本应该先除去空的查找值,从而模拟原始的行为。
4.0.5 大多数参数都可以为数组

范例

Example #1 str_replace() 基本范例

<?php
// 赋值: <body text='black'>
$bodytag str_replace("%body%""black""<body text='%body%'>");

// 赋值: Hll Wrld f PHP
$vowels = array("a""e""i""o""u""A""E""I""O""U");
$onlyconsonants str_replace($vowels"""Hello World of PHP");

// 赋值: You should eat pizza, beer, and ice cream every day
$phrase  "You should eat fruits, vegetables, and fiber every day.";
$healthy = array("fruits""vegetables""fiber");
$yummy   = array("pizza""beer""ice cream");

$newphrase str_replace($healthy$yummy$phrase);

// 赋值: 2
$str str_replace("ll""""good golly miss molly!"$count);
echo 
$count;
?>

Example #2 可能的 str_replace() 替换范例

<?php
// 替换顺序
$str     "Line 1\nLine 2\rLine 3\r\nLine 4\n";
$order   = array("\r\n""\n""\r");
$replace '<br />';

// 首先替换 \r\n 字符,因此它们不会被两次转换
$newstr str_replace($order$replace$str);

// 输出 F ,因为 A 被 B 替换,B 又被 C 替换,以此类推...
// 由于从左到右依次替换,最终 E 被 F 替换
$search  = array('A''B''C''D''E');
$replace = array('B''C''D''E''F');
$subject 'A';
echo 
str_replace($search$replace$subject);

// 输出: apearpearle pear
// 由于上面提到的原因
$letters = array('a''p');
$fruit   = array('apple''pear');
$text    'a p';
$output  str_replace($letters$fruit$text);
echo 
$output;
?>

注释

Note: 此函数可安全用于二进制对象。

Caution

了解替换顺序

由于 str_replace() 的替换时从左到右依次进行的,进行多重替换的时候可能会替换掉之前插入的值。参见该文档的范例。

Note:

该函数区分大小写。使用 str_ireplace() 可以进行不区分大小写的替换。

参见


字符串函数
在线手册:中文 英文
PHP手册
PHP手册 - N: 子字符串替换

用户评论:

Denzel Morris (06-Apr-2012 10:43)

Maybe obvious to veteran PHP programmers but less so to novice PHP programmers is the fact that this is invalid:
<?php
str_replace
($search, $replace, $subject, 1);
?>
At a glance it appears to be a reasonable request, until you realize that the fourth parameter must be a variable in order to be passed as a reference. A replacement:
<?php
str_replace
($search, $replace, $subject, $temp = 1);
// or
$temp = 1;
str_replace($search, $replace, $subject, $temp);
?>

cc at cc dot com (31-Mar-2012 01:26)

I found a pretty low tech solution to avoid the "gotcha" without worrying about the array order of how things are replaced. I could not "order" the replacement array easily because it was being read from a database table.

Anyway if you add an identifiable token to each replaced word, then just filter this out at the very end, no nested search terms are found. I just dynamically add the %% after the first char of each word before pumping it into the str_ireplace function.

$find = array("as1", "as2", "as3", "flex");
$replace = array("<a href = \"#as1\">A%%uto S%%entry R%%ev. A%%</a>", "<a href = \"#as2\">A%%uto S%%entry E%%xp</a>", "<a href = \"#as3\">A%%uto S%%entry f%%lex</a>", "<a style = \"color: red;\" href = \"#flex\">f%%lex</a>");
$text = str_ireplace($find, $replace, $text);
echo str_ireplace("%%", "", $text);

In this case I am using %% as my token as this is an unlikely char combo for me.

Kai Hoerner (30-Mar-2012 06:34)

If you want to replace only a specific occurance of a needle, you can use this handy function:

<?php
function str_replace_occurance($search, $replace, $subject, $occurance) {
  
$pos = 0;
   for (
$i = 0; $i <= $occurance; $i++) {
      
$pos = strpos($subject, $search, $pos);
   }
   return
substr_replace($subject, $replace, $pos, strlen($search));
}

$str = "hello world! hello again! say hello!";
$str = str_replace_occurance('hello', 'hi', $str, 1);
$str = str_replace_occurance('hello', 'lets do this', $str, 2);
echo
$str, PHP_EOL;

?>

Example output: "hi world! lets do this again! say hello!"

dmartineztrilla at hotmail dot com (14-Feb-2012 10:17)

windowshater try to change/switch the script file encoding to ANSI/UTF-8. I had some problems with that and I solved them this way.

windowshater (24-Nov-2011 08:15)

I hope this helps someone who has gone insane for 4 hours...

If you have someone posting weird windows curly quotes into your forms and you cannot replace them using str_replace.

I tried using mb_convert_encoding to convert the string to UTF-8 and then to do a str_replace but still no joy.

I tried to str_replace using the \xe2\x80\x98 codes and also using the chr(145), chr(146), chr(147) codes.

Nothing worked... It would replace the quotes but then ADD a question mark in a diamond !!!! no idea why..

in the end i just url encoded the string and replaced teh html codes:

$search_weird_chars = array(    '%E2%80%99', 
                        '%E2%80%99',
                        '%E2%80%98',
                        '%E2%80%9C',
                        '%E2%80%9D'
                        );

$this->post_array[$key] = urlencode($this->post_array[$key]);
               
                $this->post_array[$key] = str_replace($search_weird_chars, "'", $this->post_array[$key]);
               
                $this->post_array[$key] = urldecode($this->post_array[$key]);

this works although it is a totally retarded thing to have to do .. oh well. :-)

moostende at gmail dot com (27-Sep-2011 07:20)

Note that this does not replace strings that become part of replacement strings. This may be a problem when you want to remove multiple instances of the same repetative pattern, several times in a row.

If you want to remove all dashes but one from the string '-aaa----b-c-----d--e---f' resulting in '-aaa-b-c-d-e-f', you cannot use str_replace. Instead, use preg_replace:

<?php
$challenge
= '-aaa----b-c-----d--e---f';
echo
str_replace('--', '-', $challenge).'<br>';
echo
preg_replace('/--+/', '-', $challenge).'<br>';
?>

This outputs the following:
-aaa--b-c---d-e--f
-aaa-b-c-d-e-f

matthew at euphoriccode dot com (09-Sep-2011 06:52)

If you are looking to implement str with password generation the following function will help;

<?php
function gen_pass_process($girth = 17) {
    
$str = '';
     for($
1 = 0; $i<$girth; $i++) {
         
$str . = chr(rand(48,122));
     }
     return
$str;
}
?>

didiergm (29-Jul-2011 11:25)

php 5.3.2 (may exist before, but it never saw it until today)

the replace is always evaluated before the search takes place.

So if like me you use str_replace to replace some macros in the html code, and these replacement call a function, you must be aware that the functions always runs, not only when  the search is found.

for example:

str_replace (
   array ('%macro1%', '%macro2%'),
   array (myfunction1(), myfunction2(),
   thesource);

myfunction1() and myfunction2() are called regardless of the fact that the source contains %macro1%' or '%macro2%' or not. I genuinely believed that myfunction1 would only be called if %macro1% was found (resp myfunction2 and %macro2%)

I noticed it because myfunction1 was running a database query to increment a value and I noticed that this value was incremented w/o need. It took me a wee while to get to the bottom of it

mbullard at accuvista dot co dot uk (14-Jul-2011 10:42)

Insert space after comma.

If you have a form that stores results in a database field as comma separated values, when you display this data you can use the following to insert a space after each comma:

<?php
$find
[] = ',';
$replace[] = '&#44;&nbsp;';
$text = str_replace($find, $replace, $row_rsRecordset['Field']);
print_r($text);
?>

Notes:
1) To get round the Replacement Order Gotcha, the comma is also replaced with its code equivalent: &#44;
2) You can adapt the $replace section to suit your needs: swap out the &nbsp; code with <br/> or replace comma and space with &nbsp;&middot;&nbsp; etc.

hakre (06-Jul-2011 04:08)

I was looking for a str_replace function supporting callbacks. As I didn't found one I wrote one my own. Works exactly like str_replace, but the replace parameter is a callback or an array of callbacks (instead of string/strings in str_replace). The callback function accepts two arguments, the string that is being replaced and the count of the replacement being done.

<?php
/**
 * str_ureplace
 *
 * str_replace like function with callback
 *
 * @param string|array search
 * @param callback|array $replace
 * @param string|array $subject
 * @param int $replace_count
 * @return string|array subject with replaces, FALSE on error.
 */
function str_ureplace($search, $replace, $subject, &$replace_count = null) {
   
$replace_count = 0;
   
   
// validate input
   
$search = array_values((array) $search);
   
$searchCount = count($search);
    if (!
$searchCount) {
        return
$subject;
    }
    foreach(
$search as &$v) {
       
$v = (string) $v;
    }
    unset(
$v);
   
$replaceSingle = is_callable($replace);   
   
$replace = $replaceSingle ? array($replace) : array_values((array) $replace);
    foreach(
$replace as $index=>$callback) {
        if (!
is_callable($callback)) {
            throw new
Exception(sprintf('Unable to use %s (#%d) as a callback', gettype($callback), $index));
        }
    }
   
   
// search and replace
   
$subjectIsString = is_string($subject);
   
$subject = (array) $subject;
    foreach(
$subject as &$haystack) {
        if (!
is_string($haystack)) continue;
        foreach(
$search as $key => $needle) {
            if (!
$len = strlen($needle))
                continue;           
           
$replaceSingle && $key = 0;           
           
$pos = 0;
            while(
false !== $pos = strpos($haystack, $needle, $pos)) {
               
$replaceWith = isset($replace[$key]) ? call_user_func($replace[$key], $needle, ++$replace_count) : '';
               
$haystack = substr_replace($haystack, $replaceWith, $pos, $len);
            }
        }
    }
    unset(
$haystack);
   
    return
$subjectIsString ? reset($subject) : $subject;
}
?>

ahmed dot 3abdolah at gmail dot com (24-Jun-2011 07:05)

Hello , this is a simple function programed by str_replace for crypting password and ....etc .
You can do verry much think with str_replace :
look at this:
<?PHP
/*
 *This Function Is programed By MR-O
 *The first arabic Function for crypting
*/
function abo($variable){
$variable =str_replace("1",1 ^ 52 ,$variable);
$variable =str_replace("2",2 ^ 5 ,$variable);
$variable =str_replace("3",3 ^ 8 ,$variable);
$variable =str_replace("4",4 ^ 18 ,$variable);
$variable =str_replace("5",5 ^ 30 ,$variable);
$variable =str_replace("6",6 ^ 42 ,$variable);
$variable =str_replace("7",7 ^ 26 ,$variable);
$variable =str_replace("8",8 ^ 62 ,$variable);
$variable =str_replace("9",9 ^ 1 ,$variable);
$variable =str_replace("0",0 ^ 3 ,$variable);
$variable =str_replace("a","m8" ,$variable);
$variable =str_replace("b","2s" ,$variable);
$variable =str_replace("c","9e" ,$variable);
$variable =str_replace("d","6s" ,$variable);
$variable =str_replace("e","a3" ,$variable);
$variable =str_replace("f","m3" ,$variable);
$variable =str_replace("g","6f" ,$variable);
$variable =str_replace("h","9c" ,$variable);
$variable =str_replace("i","85" ,$variable);
$variable =str_replace("j","32" ,$variable);
$variable =str_replace("k","1a" ,$variable);
$variable =str_replace("l","6e" ,$variable);
$variable =str_replace("m","5f" ,$variable);
$variable =str_replace("n","9r" ,$variable);
$variable =str_replace("o","z0" ,$variable);
$variable =str_replace("p","y6" ,$variable);
$variable =str_replace("q","9i" ,$variable);
$variable =str_replace("r","y9" ,$variable);
$variable =str_replace("s","5g" ,$variable);
$variable =str_replace("t","p3" ,$variable);
$variable =str_replace("u","d5" ,$variable);
$variable =str_replace("v","t8" ,$variable);
$variable =str_replace("w","2x" ,$variable);
$variable =str_replace("x","6w" ,$variable);
$variable =str_replace("y","v9" ,$variable);
$variable =str_replace("z","zv" ,$variable);
$variable =str_replace("A","m8" ,$variable);
$variable =str_replace("B","24d" ,$variable);
$variable =str_replace("C","9e" ,$variable);
$variable =str_replace("D","6s2" ,$variable);
$variable =str_replace("E","qw" ,$variable);
$variable =str_replace("F","oe" ,$variable);
$variable =str_replace("G","0t" ,$variable);
$variable =str_replace("H","o9" ,$variable);
$variable =str_replace("I","74" ,$variable);
$variable =str_replace("J","dd" ,$variable);
$variable =str_replace("K","2b" ,$variable);
$variable =str_replace("L","rt" ,$variable);
$variable =str_replace("M","0c" ,$variable);
$variable =str_replace("N","2w" ,$variable);
$variable =str_replace("O","te" ,$variable);
$variable =str_replace("P","sc" ,$variable);
$variable =str_replace("Q","uy" ,$variable);
$variable =str_replace("R","hp" ,$variable);
$variable =str_replace("S","h1" ,$variable);
$variable =str_replace("T","s5" ,$variable);
$variable =str_replace("U","fs" ,$variable);
$variable =str_replace("V","e1s" ,$variable);
$variable =str_replace("W","1y" ,$variable);
$variable =str_replace("X","dv" ,$variable);
$variable =str_replace("Y","w2" ,$variable);
$variable =str_replace("Z","w5" ,$variable);
$variable = str_replace('1','5b',$variable);
$variable = str_replace('2','b1',$variable);
$variable = str_replace('3','0j',$variable);
$variable = str_replace('4','1e',$variable);
$variable = str_replace('5','z2',$variable);
$variable = str_replace('6','15e',$variable);
$variable = str_replace('7','d2',$variable);
$variable = str_replace('8','e5b',$variable);
$variable = str_replace('9','y8',$variable);
$variable = str_replace('0','g1y7',$variable);
echo
$variable ;
}

?>

thanks , MR-O .

biohazard dot ge at gmail dot com (20-May-2011 10:33)

Replace and match strings between two needles in large haystack quicker (?):

<?php
//Assume you have input:

$input = "<p> Images
 <img src='some/image_1.jpg'> This is a cat
and <img src=some/image_2.jpg> That is a bat
and <img src="
some/image_3.jpg"> What is this
and <img src=some/image_4.jpg> What is that
</p>
"
;

between_replace ('<img src="','">', $input, "{*}");
between_replace ('<img src=\'','\'>', $input, "{*}");
between_replace ('<img src=','>', $input, "'{*}'");

//As result we have

//$input == "<p> Images
// <img src='some/image_1.jpg'> This is a cat
//and <img src='some/image_2.jpg'> That is a bat
//and <img src='some/image_3.jpg'> What is this
//and <img src='some/image_4.jpg'> What is that
//</p>
//";

$result = between_fetch ("<img src='","'>", $input);

// $result == array ('some/image_1.jpg', 'some/image_2.jpg', 'some/image_3.jpg', 'some/image_4.jpg');
?>

Not a good example but only with them and str_replace I parsed word converted html for epub. This is what people invent not willing to get familiar with regular expressions...

<?php

   
function between_replace ($open, $close, &$in, $with, $limit=false, $from=0)
    {
        if (
$limit!==false && $limit==0)
        {
            return
$in;
        }       
       
$open_position = strpos ($in, $open, $from);
        if (
$open_position===false)
        {
            return
false;
        };
       
$close_position = strpos ($in, $close, $open_position+strlen($open));
        if (
$close_position===false)
        {
            return
false;
        };
       
$current = false;
        if (
strpos($with,'{*}')!==false)
        {
           
$current = substr ($in, $open_position+strlen($open), $close_position-$open_position-strlen($open));
           
$current = str_replace ('{*}',$current,$with);
           
//debug_echo ($current);
       
}
        else
        {
           
$current = $with;
        }
       
$in = substr_replace ($in, $current, $open_position+strlen($open), $close_position-$open_position-strlen($open));
       
$next_position = $open_position + strlen($current) + 1;
        if (
$next_position>=strlen($in))
        {
            return
false;
        }
        if (
$limit!==false)
        {
           
$limit--;
        }       
       
between_replace ($open, $close, $in, $with, $limit, $next_position);
        return
$in;
    }
   
    function
between_fetch ($open, $close, &$in, &$result=null, $with='', $limit=false, $from=0)
    {
        if (
$limit!==false && $limit==0)
        {
            return
$in;
        };
        if (!
is_array($result))
        {
           
$result = array ();
        }
       
$open_position = strpos ($in, $open, $from);
        if (
$open_position===false)
        {
            return
false;
        };
       
$close_position = strpos ($in, $close, $open_position+strlen($open));
        if (
$close_position===false)
        {
            return
false;
        };
       
$current = substr ($in, $open_position+strlen($open), $close_position-$open_position-strlen($open));
       
$next_position = $open_position + strlen($current) + 1;
        if (
$with!='')
        {
           
$current = str_replace('{*}', $current, $with);
        }
       
$result[] = $current;
        if (
$next_position>=strlen($in))
        {
            return
false;
        }
        if (
$limit!==false)
        {
           
$limit--;
        }
       
between_fetch ($open, $close, $in, $result, $with, $limit, $next_position);
        return
$in;
    }

?>

jay_knows_(all)uk at hotmail dot com (16-Feb-2011 03:16)

This strips out horrible MS word characters.

Just keep fine tuning it until you get what you need, you'll see ive commented some out which caused problems for me.

There could be some that need adding in, but its a start to anyone who wishes to make their own custom function.

<?php

function msword_conversion($str)
{
$str = str_replace(chr(130), ',', $str);    // baseline single quote
$str = str_replace(chr(131), 'NLG', $str);  // florin
$str = str_replace(chr(132), '"', $str);    // baseline double quote
$str = str_replace(chr(133), '...', $str);  // ellipsis
$str = str_replace(chr(134), '**', $str);   // dagger (a second footnote)
$str = str_replace(chr(135), '***', $str);  // double dagger (a third footnote)
$str = str_replace(chr(136), '^', $str);    // circumflex accent
$str = str_replace(chr(137), 'o/oo', $str); // permile
$str = str_replace(chr(138), 'Sh', $str);   // S Hacek
$str = str_replace(chr(139), '<', $str);    // left single guillemet
// $str = str_replace(chr(140), 'OE', $str);   // OE ligature
$str = str_replace(chr(145), "'", $str);    // left single quote
$str = str_replace(chr(146), "'", $str);    // right single quote
// $str = str_replace(chr(147), '"', $str);    // left double quote
// $str = str_replace(chr(148), '"', $str);    // right double quote
$str = str_replace(chr(149), '-', $str);    // bullet
$str = str_replace(chr(150), '-–', $str);    // endash
$str = str_replace(chr(151), '--', $str);   // emdash
// $str = str_replace(chr(152), '~', $str);    // tilde accent
// $str = str_replace(chr(153), '(TM)', $str); // trademark ligature
$str = str_replace(chr(154), 'sh', $str);   // s Hacek
$str = str_replace(chr(155), '>', $str);    // right single guillemet
// $str = str_replace(chr(156), 'oe', $str);   // oe ligature
$str = str_replace(chr(159), 'Y', $str);    // Y Dieresis
$str = str_replace('°C', '&deg;C', $str);    // Celcius is used quite a lot so it makes sense to add this in
$str = str_replace('?', '&pound;', $str);
$str = str_replace("'", "'", $str);
$str = str_replace('"', '"', $str);
$str = str_replace('–', '&ndash;', $str);

return
$str;
}

?>

Oleg Butuzov butuzov at made dot com dot ua (01-Feb-2011 03:26)

Sometimes we need replace just first occurrence , a small function that can help you.

<?php
    $string
= 'alpharomero alpha arlekino';
   
   
   
var_dump(str_replace_once('alpha','[replaced]', $string));
   
var_dump($string);
    echo
'<hr>';
   
   
var_dump(str_replace_once('a','[replaced]', $string));
   
var_dump($string);
    echo
'<hr>';
   
   
var_dump(str_replace_once('abnormal','[replaced]', $string));
   
var_dump($string);
    echo
'<hr>';
   
   
var_dump(str_replace_once('n','[replaced]', $string));
   
var_dump($string);
    echo
'<hr>';   
   

    function
str_replace_once($str_pattern, $str_replacement, $string){
       
        if (
strpos($string, $str_pattern) !== false){
           
$occurrence = strpos($string, $str_pattern);
            return
substr_replace($string, $str_replacement, strpos($string, $str_pattern), strlen($str_pattern));
        }
       
        return
$string;
    }
?>

mevdschee (29-Jan-2011 11:19)

say you want every "a" replaced by "abba" and every "b" replaced by "baab" in the string "ab" you do:

<?php
$search
= array("a","b");
$replace = array("abba","baab");
echo
str_replace($search,$replace,"ab");
?>

that results in "abaabbaababaab" and not the expected "abbabaab"

I wrote this little snippet to solve the replacements-in-replacement problem:

<?php
function search_replace($s,$r,$sql)
{
$e = '/('.implode('|',array_map('preg_quote', $s)).')/';
 
$r = array_combine($s,$r);
  return
preg_replace_callback($e, function($v) use ($s,$r) { return $r[$v[1]]; },$sql);
}

echo
search_replace($search,$replace,"ab");
?>

that results in the expected "abbabaab"

Viam (17-Jan-2011 05:57)

This way it is possible to make a encryption of some sort
A small script i wrote.
Proof of concept:

<?php
//Data to be encrypted.Take note that it has to be text.
$Input_data  = "i want a apple";//Place the text here
$Encrypter_variables= array("a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n", "o", "p", "q", "r", "s", "t", "u", "v", "w", "x", "y", "z");
$Variables_replace = array("1", "2", "3", "4", "5", "6", "7", "8", "9", "%", "!", "@", "#", "$", "^", "&", "*","~", "+", ":", "-", "=", "?", "'", ".", "<");
$New_data = str_replace($Encrypter_variables, $Variables_replace, $Input_data);

echo
$New_data;//Displays the encrypted data,In this case it is 9 ?1$: 1 1&&@5

//Data to be decrypted
$Output_data  = "9 ?1$: 1 1&&@5";////Place the encrypted data here
$Decrypter_variables =  array("1", "2", "3", "4", "5", "6", "7", "8", "9", "%", "!", "@", "#", "$", "^", "&", "*","~", "+", ":", "-", "=", "?", "'", ".", "<");
$Variables_assign  = array("a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n", "o", "p", "q", "r", "s", "t", "u", "v", "w", "x", "y", "z");
$Newer = str_replace($Decrypter_variables, $Variables_assign  , $Output_data);

echo
$Newer;//Displays the decrypted data Which is i want a apple
?>

Steve (14-Jan-2011 02:22)

Just a note about the json_decode solution from nikolaz dot tang at hotmail dot com
It may return an instance of stdClass, which may not quite be what you want.  (using multi-dimensional arrays)

IF you want to be sure, stick to the proper recursive solution.

nikolaz dot tang at hotmail dot com (12-Nov-2010 04:57)

A faster way to replace the strings in multidimensional array is to json_encode() it, do the str_replace() and then json_decode() it, like this:

<?php
function str_replace_json($search, $replace, $subject){
     return
json_decode(str_replace($search, $replacejson_encode($subject)));

}
?>

This method is almost 3x faster (in 10000 runs.) than using recursive calling and looping method, and 10x simpler in coding.

Compared to:

<?php
function str_replace_deep($search, $replace, $subject)
{
    if (
is_array($subject))
    {
        foreach(
$subject as &$oneSubject)
           
$oneSubject = str_replace_deep($search, $replace, $oneSubject);
        unset(
$oneSubject);
        return
$subject;
    } else {
        return
str_replace($search, $replace, $subject);
    }
}
?>

jpoirier at hotmail.fr (25-Jun-2010 01:40)

I've found something interesting about the $count param:

As of 5.0.0 the last parameter (count) is passed by reference (it is optional). So 2 choices :
1)passed not by reference - it's limit
2)passed by reference : the var is is_null() - all occurences are replaced the number is returned in "count", otoh if the var !is_null() -> use as limit.

On version 5.3.0 the 4th parameter, count, does not work when not passed
by reference.

http://bugs.php.net/bug.php?id=11457

Hope this help

apmuthu at usa dot net (19-Jun-2010 08:31)

If we have a html template that contains placeholders in curly braces that need to be replaced in runtime, the following function will do it using str_replace:

<?php

function parse_template($filename, $data) {
// example template variables {a} and {bc}
// example $data array
// $data = Array("a" => 'one', "bc" => 'two');
   
$q = file_get_contents($filename);
    foreach (
$data as $key => $value) {
       
$q = str_replace('{'.$key.'}', $value, $q);
    }
    return
$q;
}

?>

cableray (11-Jun-2010 08:03)

If you wish to get around the 'gotcha', you could do something like this:

<?php

$find
=array('a', 'p', '^', '*');
$replace = array('^', '*', 'apple', 'pear');
str_replace($find, $replace, 'a p');

?>

The idea here is that you first replace the items you want with unique identifiers (that you are unlikely to find in the subject) and then search for those identifiers and then replace them.

christian dot reinecke at web dot de (14-May-2010 08:06)

If you need to replace a string in another, but only once but still in all possible combinations (f.e. to replace "a" with "x" in "aba" to get array("xba", "abx")) you can use this function:
<?php
function getSingleReplaceCombinations($replace, $with, $inHaystack)
{
   
$splits = explode($replace, $inHaystack);
   
$result = array();
    for (
$i = 1, $ix = count($splits); $i < $ix; ++$i) {
       
$previous = array_slice($splits, 0, $i);
       
$next     = array_slice($splits, $i);
       
       
$combine  = array_pop($previous) . $with . array_shift($next);
       
$result[] = implode($replace, array_merge($previous, array($combine), $next));
    }
    return
$result;
}
var_dump(getSingleReplaceCombinations("a", "x", "aba")); // result as mentioned above
?>
It may not be the best in performance, but it works.

jbarnett at jmbelite dot com (26-Apr-2010 09:23)

Might be worth mentioning that a SIMPLE way to accomplish Example 2 (potential gotchas) is to simply start your "replacements" in reverse.

So instead of starting from "A" and ending with "E":

<?php
$search 
= array('A', 'B', 'C', 'D', 'E');
$replace = array('B', 'C', 'D', 'E', 'F');
// replaces A to B, B to C, C to D, D to E, E to F (makes them all F)
// start from "E" and end with "A":

$search  = array('E', 'D', 'C', 'B', 'A');
$replace = array('F', 'E', 'D', 'C', 'B');
// replaces E to F, D to E, C to D, B to C, A to B (prevents from
// multiple replacements of already replaced values)
?>

So basically start from the "end" and put the replacements in an order where the "replaced value" won't equal a value that exists later in the "search array".

pjcdawkins at googlemail dot com (24-Apr-2010 01:23)

Here's a deep replace function allowing multi-dimensional arrays in $search, $replace and $subject. The keys and other structure of $subject are preserved.

<?php
// Auxiliary function:
function _replaceWithAnything($search,$replace,$subject){
  if(!
is_array($search) || !is_array($replace)){
   
$search=array($search);
   
$replace=array($replace);
  }
 
$match=array_search($subject,$search,true);
  if(
$match!==false && array_key_exists($match,$replace))
   
$subject=$replace[$match];
  return
$subject;
}

// Main function:
function deepReplace($search,$replace,$subject){
  if(!
is_array($subject))
    return
_replaceWithAnything($search,$replace,$subject);
  foreach(
$subject as &$val){
    if(
is_array($val)){
     
$val=deepReplace($search,$replace,$val);
      continue;
    }
   
$val=_replaceWithAnything($search,$replace,$val);
  }
  return
$subject;
}
?>

fleshgrinder at gmx dot at (16-Apr-2010 01:50)

Fast function to replace new lines from a given string. This is interesting to replace all new lines from e. g. a text formatted in HTML retrieved from database and printing it without the unnecessary new lines. This results in slightly faster rendering in the Web browser.

<?php

/**
 * Replace all linebreaks with one whitespace.
 *
 * @access public
 * @param string $string
 *   The text to be processed.
 * @return string
 *   The given text without any linebreaks.
 */
function replace_newline($string) {
  return (string)
str_replace(array("\r", "\r\n", "\n"), '', $string);
}

?>

Wes Foster (17-Dec-2009 09:32)

Feel free to optimize this using the while/for or anything else, but this is a bit of code that allows you to replace strings found in an associative array.

For example:
<?php
$replace
= array(
'dog' => 'cat',
'apple' => 'orange'
'chevy'
=> 'ford'
);

$string = 'I like to eat an apple with my dog in my chevy';

echo
str_replace_assoc($replace,$string);

// Echo: I like to eat an orange with my cat in my ford
?>

Here is the function:

<?php
function strReplaceAssoc(array $replace, $subject) {
   return
str_replace(array_keys($replace), array_values($replace), $subject);   
}
?>

[Jun 1st, 2010 - EDIT BY thiago AT php DOT net: Function has been replaced with an updated version sent by ljelinek AT gmail DOT com]

jonathon at compwright dot com (12-Nov-2009 05:23)

If you get a blank page when passing an object to str_replace() (relying on __toString() to convert the object to a string) then you may need to force string context using (string), like so:

<?php
private function _load_vars($vars, &$source = false) {
    if(!
$source) $source =& $this->code;
        foreach((array)
$vars as $key => $var) {
           
$source = str_replace("{".$key."}", (string) $var, $source);
        }
    }
}
?>

I was running Ubuntu Server with PHP 5.2.6 and getting apache segfaults and mysterious blank pages when $var happened to be certain objects.

Chris (01-Nov-2009 04:22)

Compress a string's internal spaces:
<?php
$str
= ' This is    a    test   ';
$count = 1;
while(
$count)
   
$str = str_replace('  ', ' ', $str, $count);
?>

Decko (15-Oct-2009 09:41)

As mentioned earlier you should take the order into account when substituting multiple values.

However it is worth noticing that str_replace doesn't seem to re-read the string when doing single replacements. Take the following example.

<?php
  $s
= '/a/a/';
 
$s = str_replace('/a/', '/', $s);
?>

You would expect the following.

First replacement '/a/a/' -> '/a/'
Second replacement '/a/'->'/'

This is not the case, the actual result will be '/a/'.

To fix this, you will have to put str_replace in a while-loop.

<?php
  $s
= '/a/a/';
  while(
strpos($s, '/a/') !== false)
   
$s = str_replace('/a/', '/', $s); //eventually $s will == '/'
?>

Alberto Lepe (16-Jun-2009 03:44)

Be careful when replacing characters (or repeated patterns in the FROM and TO arrays):

For example:

<?php
$arrFrom
= array("1","2","3","B");
$arrTo = array("A","B","C","D");
$word = "ZBB2";
echo
str_replace($arrFrom, $arrTo, $word);
?>

I would expect as result: "ZDDB"
However, this return: "ZDDD"
(Because B = D according to our array)

To make this work, use "strtr" instead:

<?php
$arr
= array("1" => "A","2" => "B","3" => "C","B" => "D");
$word = "ZBB2";
echo
strtr($word,$arr);
?>

This returns: "ZDDB"

moz667 at gmail dot com (21-May-2009 05:49)

<?php
/*
This is a function for made recursive str_replaces in an array
*/
function recursive_array_replace($find, $replace, &$data) {
    if (
is_array($data)) {
        foreach (
$data as $key => $value) {
            if (
is_array($value)) {
               
recursive_array_replace($find, $replace, $data[$key]);
            } else {
               
$data[$key] = str_replace($find, $replace, $value);
            }
        }
    } else {
       
$data = str_replace($find, $replace, $data);
    }
}

$a = array();
$a['a'] = "a";
$a['b']['a'] = "ba";
$a['b']['b'] = "bb";
$a['c'] = "c";
$a['d']['a'] = "da";
$a['d']['b'] = "db";
$a['d']['c'] = "dc";
$a['d']['d'] = "dd";

echo
"Before Replaces";
print_r($a);

recursive_array_replace("a", "XXXX", $a);

echo
"After Replaces";
print_r($a);
?>

michael dot moussa at gmail dot com (29-Jan-2009 02:38)

As previous commentators mentioned, when $search contains values that occur earlier in $replace, str_replace will factor those previous replacements into the process rather than operating solely on the original string.  This may produce unexpected output.

Example:

<?php
$search
= array('A', 'B', 'C', 'D', 'E');
$replace = array('B', 'C', 'D', 'E', 'F');
$subject = 'ABCDE';

echo
str_replace($search, $replace, $subject); // output: 'FFFFFF'
?>

In the above code, the $search and $replace should replace each occurrence in the $subject with the next letter in the alphabet.  The expected output for this sample is 'BCDEF'; however, the actual output is 'FFFFF'.

To more clearly illustrate this, consider the following example:

<?php
$search
= array('A', 'B', 'C', 'D', 'E');
$replace = array('B', 'C', 'D', 'E', 'F');
$subject = 'A';

echo
str_replace($search, $replace, $subject); // output: 'F'
?>

Since 'A' is the only letter in the $search array that appears in $subject, one would expect the result to be 'B'; however, replacement number $n does *not* operate on $subject, it operates on $subject after the previous $n-1 replacements have been completed.

The following function utilizes array_combine and strtr to produce the expected output, and I believe it is the most efficient way to perform the desired string replacement without prior replacements affecting the final result.

<?php
/**
* When using str_replace(...), values that did not exist in the original string (but were put there by previous
* replacements) will be replaced continuously.  This string replacement function is designed replace the values
* in $search with those in $replace while not factoring in prior replacements.  Note that this function will
* always look for the longest possible match first and then work its way down to individual characters.
*
* The "o" in "stro_replace" represents "original", indicating that the function operates only on the original string.
*
* @param array $search list of strings or characters that need to be replaced
* @param array $replace list of strings or characters that will replace the corresponding values in $search
* @param string $subject the string on which this operation is being performed
*
* @return string $subject with all substrings in the $search array replaced by the values in the $replace array
*/
function stro_replace($search, $replace, $subject)
{
    return
strtr( $subject, array_combine($search, $replace) );
}

$search = array('A', 'B', 'C', 'D', 'E');
$replace = array('B', 'C', 'D', 'E', 'F');
$subject = 'ABCDE';

echo
stro_replace($search, $replace, $subject); // output: 'BCDEF'
?>

Some other examples:

<?php
$search
= array(' ', '&');
$replace = array('&nbsp;', '&amp;');
$subject = 'Hello & goodbye!';

// We want to replace the spaces with &nbsp; and the ampersand with &amp;
echo str_replace($search, $replace, $subject); // output: "Hello&amp;nbsp&amp;&amp;nbspgoodbye!" - wrong!

echo stro_replace($search, $replace, $subject); // output: "Hello&nbsp;&amp;&nbsp;goodbye!" - correct!

/*
    Note:  Run the above code in the CLI or view source on your web browser - the replacement strings for stro_replace are HTML entities which the browser interprets.
*/
?>

<?php
$search
= array('ERICA', 'AMERICA');
$replace = array('JON', 'PHP');
$subject = 'MIKE AND ERICA LIKE AMERICA';

// We want to replace the name "ERICA" with "JON" and the word "AMERICA" with "PHP"
echo str_replace($search, $replace, $subject); // output: "MIKE AND JON LIKE AMJON", which is not correct

echo stro_replace($search, $replace, $subject); // output: "MIKE AND JON LIKE PHP", which is correct
?>

nospam at nospam dot com (02-Dec-2008 10:55)

Replacement for str_replace in which a multiarray of numerically keyed data can be properly evaluated with the given template without having a search for 11 be mistaken for two 1's next to each other

<?php

function data_template($input, $template) {
  if (
$template) { // template string
   
if ($split = str_split($template)) { // each char as array member
     
foreach ($split as $char) { // each character
       
if (is_numeric($char)) { // test for digit
         
if ($s != 1) { // new digit sequence
           
$i++;
           
$s = 1;
          }
         
$digits[$i] .= $char; // store digit
       
} else { // not a digit
         
if ($s != 2) { // new non-digit sequence
           
$i++;
           
$s = 2;
          }
         
$strings[$i] .= $char; // store string
       
}
      }
      if (
$i && $input && is_array($input)) { // input data
       
foreach ($input as $sub) { // each subarray
         
if (is_array($sub)) {
           
$out = ''; // reset output
           
for ($j = 0; $j <= $i; $j++) { // each number/string member
             
if ($number = $digits[$j]) { // number
               
$out .= $sub[$number]; // add value from subarray to output
             
} else { // string
               
$out .= $strings[$j]; // add to output
             
}
            }
           
$a[] = $out;
          }
        }
        return
$a;
      }
// input
   
} // split
 
} // template
}

$input = array(array(1=>'yellow', 2=>'banana', 11=>'fruit'), array(1=>'green', 2=>'spinach', 11=>'vegetable'), array(1=>'pink', 2=>'salmon', 11=>'fish'));

print_r (data_template($input, '2: a 1, healthy 11'));

/*
Array
(
    [0] => banana: a yellow, healthy fruit
    [1] => spinach: a green, healthy vegetable
    [2] => salmon: a pink, healthy fish
)
*/

// str_replace would have wanted to output 'banana: a yellow, healthy yellowyellow

?>

Not sure if this will help anyone but I wrote it for my application and thought I would share just in case

nick at NOSPAM dot pitchinteractive dot com (07-Oct-2008 12:12)

I tried max at efoxdesigns dot com solution for str_replace_once but it didn't work quite right so I came up with this solution (all params must be strings):

<?php
function str_replace_once($search, $replace, $subject) {
   
$firstChar = strpos($subject, $search);
    if(
$firstChar !== false) {
       
$beforeStr = substr($subject,0,$firstChar);
       
$afterStr = substr($subject, $firstChar + strlen($search));
        return
$beforeStr.$replace.$afterStr;
    } else {
        return
$subject;
    }
}
?>

paolo A T doppioclick D O T com (05-Sep-2008 12:15)

For PHP 4 < 4.4.5 and PHP 5 < 5.2.1 you may occur (like me) in this bug:

http://www.php-security.org/MOPB/MOPB-39-2007.html

troy at troyonline dot com (23-Jun-2008 06:18)

Yet another deep replace function:

<?php
   
function str_replace_deep( $search, $replace, $subject)
    {
       
$subject = str_replace( $search, $replace, $subject);

        foreach (
$subject as &$value)
           
is_array( $value) and $value =str_replace_deep( $search, $replace, $value);
           
        return
$subject;
    }
?>

David Gimeno i Ayuso (info at sima dot cat) (09-Aug-2007 08:22)

With PHP 4.3.1, at least, str_replace works fine when working with single arrays but mess it all with two or more dimension arrays.

<?php
$subject
= array("You should eat this","this","and this every day.");
$search  = "this";
$replace = "that";
$new     = str_replace($search, $replace, $subject);

print_r($new); // Array ( [0] => You should eat that [1] => that [2] => and that every day. )

echo "<hr />";

$subject = array(array("first", "You should eat this")
                ,array(
"second","this")
                ,array(
"third", "and this every day."));
$search  = "this";
$replace = "that";
$new     = str_replace($search, $replace, $subject);

print_r($new); // Array ( [0] => Array [1] => Array [2] => Array )

?>

tim at hysniu dot com (05-Jun-2007 07:27)

I found that having UTF-8 strings in as argument didnt
work for me using heavyraptors function.
Adding UTF-8 as argument on htmlentities
fixed the problem.

cheers, tim at hysniu.com

<?php
function replace_accents($str) {
 
$str = htmlentities($str, ENT_COMPAT, "UTF-8");
 
$str = preg_replace(
'/&([a-zA-Z])(uml|acute|grave|circ|tilde);/',
'$1',$str);
  return
html_entity_decode($str);
}

?>

kole (26-Feb-2007 01:48)

My input is MS Excel file but I want to save ‘,’,“,” as ',',",".

    $badchr        = array(
        "\xc2", // prefix 1
        "\x80", // prefix 2
        "\x98", // single quote opening
        "\x99", // single quote closing
        "\x8c", // double quote opening
        "\x9d"  // double quote closing
    );
       
    $goodchr    = array('', '', '\'', '\'', '"', '"');
       
    str_replace($badchr, $goodchr, $strFromExcelFile);

Works for me.

rlee0001 at sbcglobal dot net (16-Feb-2007 08:30)

This is a more rigid alternative to spectrereturns at creaturestoke dot com's replace_different function:

<?php

       
function str_replace_many ($search, $replacements, $subject) {
           
$index = strlen($subject);
           
$replacements = array_reverse($replacements);

            if (
count($replacements) != substr_count($subject, $search)) {
                return
FALSE;
            }

            foreach (
$replacements as $replacement) {
               
$index = strrpos(substr($subject, 0, $index), $search);
               
$prefix = substr($subject, 0, $index);
               
$suffix = substr($subject, $index + 1);
               
$subject = $prefix . $replacement . $suffix;
            }

            return
$subject;
        }
?>

This will return false if there are a different number of $replacements versus number of occurrences of $search in $subject. Additionally, $search much be exactly one character (if a string is provided, only the first character in the string will be used). Examples:

<?php
       
echo str_replace_many('?',array('Jane','banana'),'? is eating a ?.');
?>

prints: "Jane is eating a banana."

(15-Jan-2007 09:42)

Before spending hours searching your application why it makes UTF-8 encoding into some malformed something with str_replace, make sure you save your PHP file in UTF-8 (NO BOM).

This was at least one of my problems.

matt wheaton (30-Mar-2006 03:40)

As an effort to remove those Word copy and paste smart quotes, I've found that this works with UTF8 encoded strings (where $text in the following example is UTF8). Also the elipsis and em and en dashes are replaced.

There is an "invisible" character after the ?€ for the right side double smart quote that doesn't seem to display here. It is chr(157).

<?php
  $find
[] = '?€?'// left side double smart quote
 
$find[] = '?€?'// right side double smart quote
 
$find[] = '?€?'// left side single smart quote
 
$find[] = '?€?'// right side single smart quote
 
$find[] = '?€?'// elipsis
 
$find[] = '?€”'// em dash
 
$find[] = '?€“'// en dash

 
$replace[] = '"';
 
$replace[] = '"';
 
$replace[] = "'";
 
$replace[] = "'";
 
$replace[] = "...";
 
$replace[] = "-";
 
$replace[] = "-";

 
$text = str_replace($find, $replace, $text);
?>

David Gimeno i Ayuso (info at sima-pc dot com) (25-Aug-2003 02:12)

Take care with order when using arrays in replacement.

<?php
$match
=array("ONE","TWO","THREE");
$replace=array("TWO WORDS","MANY LETTERS","OTHER IDEAS");
$sample="ONE SAMPLE";
echo
str_replace($match,$replace,$sample);
?>

It will show: "MANY LETTERS WORDS SAMPLE"

That is, after replacing "ONE" with "TWO WORDS", process follows with next array item and it changes "TWO" with "MANY LETTERS".