String 함수 목록

참고

For even more powerful string handling and manipulating functions take a look at the POSIX regular expression functions and the Perl compatible regular expression functions.

Table of Contents

  • addcslashes — C 형식으로 문자열을 슬래시로 인용
  • addslashes — 문자열을 슬래시로 인용
  • bin2hex — 바이너리 데이터를 16진 표현으로 변환
  • chop — 별칭: rtrim
  • chr — 특정 문자를 반환
  • chunk_split — 문자열을 작은 조각으로 나눕니다
  • convert_cyr_string — 키릴 문자셋을 다른 키릴 문자셋으로 변환
  • convert_uudecode — Decode a uuencoded string
  • convert_uuencode — Uuencode a string
  • count_chars — 문자열 안에 사용한 문자에 대한 정보를 반환
  • crc32 — 문자열의 crc32 값을 계산
  • crypt — 단방향 문자열 암호화(해시)
  • echo — 하나 이상의 문자열을 출력
  • explode — 문자열을 문자열로 나눕니다
  • fprintf — 형식화한 문자열을 스트림에 기록
  • get_html_translation_table — htmlspecialchars와 htmlentities에서 사용하는 변환표를 반환합니다
  • hebrev — 논리 히브리 텍스트를 표시 텍스트로 변환
  • hebrevc — 줄바꿈을 포함하여 논리 히브리 텍스트를 표시 텍스트로 변환
  • hex2bin — Decodes a hexadecimally encoded binary string
  • html_entity_decode — 모든 HTML 엔티티를 해당하는 문자로 변환
  • htmlentities — 해당하는 모든 문자를 HTML 엔티티로 변환
  • htmlspecialchars_decode — Convert special HTML entities back to characters
  • htmlspecialchars — 특수 문자를 HTML 엔터티로 변환
  • implode — 문자열로 배열 원소를 결합
  • join — 별칭: implode
  • lcfirst — Make a string's first character lowercase
  • levenshtein — Calculate Levenshtein distance between two strings
  • localeconv — Get numeric formatting information
  • ltrim — 문자열 시작에서 공백(이나 다른 문자)를 제거
  • md5_file — 주어진 파일의 md5 해시를 계산
  • md5 — 문자열의 md5 해시를 계산
  • metaphone — 문자열의 메타폰 키를 계산
  • money_format — Formats a number as a currency string
  • nl_langinfo — Query language and locale information
  • nl2br — 문자열의 모든 줄바꿈 앞에 HTML 줄바꿈 태그를 삽입
  • number_format — Format a number with grouped thousands
  • ord — 문자의 아스키 값을 반환
  • parse_str — 문자열을 변수로 해석
  • print — 문자열을 출력
  • printf — 형식화한 문자열을 출력
  • quoted_printable_decode — 인용된 인쇄용 문자열을 8비트 문자열로 변환
  • quoted_printable_encode — Convert a 8 bit string to a quoted-printable string
  • quotemeta — 메타 문자를 인용
  • rtrim — 문자열 마지막의 공백(이나 다른 문자)을 제거
  • setlocale — Set locale information
  • sha1_file — 파일의 sha1 해시를 계산
  • sha1 — 문자열의 sha1 해시를 계산
  • similar_text — 두 문자열 사이의 유사성을 계산
  • soundex — 문자열의 soundex 키를 계산
  • sprintf — 형식화한 문자열을 반환
  • sscanf — 문자열을 형식에 따라 해석
  • str_getcsv — Parse a CSV string into an array
  • str_ireplace — 대소문자 구분 없는 str_replace
  • str_pad — 문자열을 지정한 길이가 되도록 다른 문자열로 채웁니다
  • str_repeat — 문자열을 반복
  • str_replace — 발견한 모든 검색 문자열을 치환 문자열로 교체
  • str_rot13 — 문자열에 rot13 변환을 수행
  • str_shuffle — 문자열을 무작위로 섞습니다
  • str_split — 문자열을 배열로 변환
  • str_word_count — 문자열에서 사용한 단어에 대한 정보를 반환
  • strcasecmp — 대소문자 구분 없는 바이너리 안전 문자열 비교
  • strchr — 별칭: strstr
  • strcmp — 바이너리 안전 문자열 비교
  • strcoll — 로케일 기반 문자열 비교
  • strcspn — 마스크에 매칭하지 않는 처음 세그먼트의 길이를 찾습니다
  • strip_tags — 문자열에서 HTML과 PHP 태그를 제거
  • stripcslashes — addcslashes로 인용한 문자열을 되돌립니다
  • stripos — 대소문자를 구분 없이 문자열이 처음 나타나는 위치를 탐색
  • stripslashes — 따옴표 처리한 문자열을 풉니다
  • stristr — 대소문자를 구분하지 않는 strstr
  • strlen — 문자열 길이를 얻습니다
  • strnatcasecmp — "자연순" 알고리즘을 이용한 대소문자 구분 없는 문자열 비교
  • strnatcmp — "자연순" 알고리즘을 이용한 문자열 비교
  • strncasecmp — 대소문자 구분 없는 처음 n 문자의 바이너리 안전 문자열 비교
  • strncmp — 처음 n 문자의 바이너리 안전 문자열 비교
  • strpbrk — Search a string for any of a set of characters
  • strpos — 문자열이 처음 나타나는 위치를 찾습니다
  • strrchr — 문자열에서 문자가 마지막으로 나오는 부분을 찾습니다
  • strrev — 문자열 뒤집기
  • strripos — 문자열에서 대소문자 구분 없이 문자열의 마지막 위치를 찾습니다
  • strrpos — 문자열에서 마지막 문자의 위치를 찾습니다
  • strspn — 마스크에 매치하는 초기 세그먼트의 길이를 찾는다
  • strstr — 문자열이 처음으로 나오는 부분을 찾습니다
  • strtok — 문자열을 토큰화
  • strtolower — 문자열을 소문자로 만듭니다
  • strtoupper — 문자열을 대문자로 만듭니다
  • strtr — Translate characters or replace substrings
  • substr_compare — Binary safe comparison of two strings from an offset, up to length characters
  • substr_count — Count the number of substring occurrences
  • substr_replace — Replace text within a portion of a string
  • substr — Return part of a string
  • trim — Strip whitespace (or other characters) from the beginning and end of a string
  • ucfirst — Make a string's first character uppercase
  • ucwords — Uppercase the first character of each word in a string
  • vfprintf — Write a formatted string to a stream
  • vprintf — Output a formatted string
  • vsprintf — Return a formatted string
  • wordwrap — Wraps a string to a given number of characters
add a note add a note

User Contributed Notes 24 notes

up
6
lrirwin at alum dot wustl dot edu
3 years ago
I'm converting 30 year old code and needed a string TAB function:

  //tab function similar to TAB used in old BASIC languages
  //though some of them did not truncate if the string were
  //longer than the requested position
  function tab($instring="",$topos=0){
    if(strlen($instring)<$topos){
      $result=str_pad($instring,$topos-1," ",STR_PAD_RIGHT);
    }else{
      $result=substr($instring,0,$topos-1);
    }
    return $result;
  }

  $pline="String with this tab to 50 and";
  $tline=tab($pline,50)."finish it.";
  echo $tline.PHP_EOL;
  $pline="101010101020202020203030303030404040404050505050506060606060";
  $tline=tab($pline,50)."finish it.";
  echo $tline.PHP_EOL;

  //Results in this output:
  //String with this tab to 50 and                   finish it.
  //1010101010202020202030303030304040404040505050505finish it.
up
-7
t0russ at gmail dot com
19 years ago
to kristin at greenaple dot on dot ca:
thanx for sharing.
your function in recursive form proved to be slightly faster and it returns false (as it should) when the character is not found instead of number 0:
<?php
function strnposr($haystack, $needle, $occurance, $pos = 0) {
    return (
$occurance<2)?strpos($haystack, $needle, $pos):strnposr($haystack,$needle,$occurance-1,strpos($haystack, $needle, $pos) + 1);
}
?>
up
-6
kristin at greenapple dot on dot ca
19 years ago
I really searched for a function that would do this as I've seen it in other languages but I couldn't find it here. This is particularily useful when combined with substr() to take the first part of a string up to a certain point.

strnpos() - Find the nth position of needle in haystack.

<?php

   
function strnpos($haystack, $needle, $occurance, $pos = 0) {
       
        for (
$i = 1; $i <= $occurance; $i++) {
           
$pos = strpos($haystack, $needle, $pos) + 1;
        }
        return
$pos - 1;
       
    }

?>

Example: Give me everything up to the fourth occurance of '/'.

<?php

    $haystack
= "/home/username/www/index.php";
   
$needle = "/";
   
   
$root_dir = substr($haystack, 0, strnpos($haystack, $needle, 4));
   
    echo
$root_dir;
   
?>

Returns: /home/username/www

Use this example with the server variable $_SERVER['SCRIPT_NAME'] as the haystack and you can self-discover a document's root directory for the purposes of locating global files automatically!
up
-9
admin at fivestarbuy dot com
19 years ago
This example lets you parse an unparsed strings variables. Warning: This could cause security leaks if you allow users to pass $variables through this engine. I recommend only using this for your Content Management System.

<?
$mytime
=time();
$mydog="My Dog Ate My PHP!";

# Your Parsing String:
$s1 = 'Hyphen Variable Preserving: $mytime, and $mydog';
echo
"Before: <br><br>$s1<br><br>";

# Remember, wherever you define this, it will not be defined GLOBAL into the function
# which is why we define it here. Defining it global could lead to security issues.
$vardata=get_defined_vars();

# Parse the string
$s1 = StrParse($s1,$vardata);

echo
"After: <br><br>$s1";

function
StrParse($str,$vardata) {
# Takes a string, or piece of data, that contains PHP Variables

# For example, unparsed variables like:  Test using time: $mytime
# This example shows $mytime, and not the actual variable value.
# The end result shows the actual variable value of $mytime.

# This is useful for building a content management system,
# and directing your variables into your content data,
# where content is stored in a file or database, unparsed.
# Of course this could slow down page loads, but it's a good way
# to parse data from current variables into your loaded new data
# making it compatible.

# Then the variables are replaced with the actual variable..
$getvarkeys=array_keys($vardata);
$ret=$str;
for (
$x=0; $x < count($getvarkeys); $x++) {
   
$myvar=$getvarkeys[$x];
   
#echo "Variable: " . $myvar . " [" . $vardata[$myvar] . "]<br>";
   
$ret=str_replace('$' . $myvar, $vardata[$myvar], $ret);
}
return
$ret;

}

?>
up
-8
[tab!]
20 years ago
//
// string strtrmvistl(  string str, [int maxlen = 64],
//                      [bool right_justify = false],
//                      [string delimter = "<br>\n"])
//
// splits a long string into two chunks (a start and an end chunk)
// of a given maximum length and seperates them by a given delimeter.
// a second chunk can be right-justified within maxlen.
// may be used to 'spread' a string over two lines.
//

function strtrmvistl($str, $maxlen = 64, $right_justify = false, $delimter = "<br>\n") {
    if(($len = strlen($str = chop($str))) > ($maxlen = max($maxlen, 12))) {
        $newstr = substr($str, 0, $maxlen - 3);

        if($len > ($maxlen - 3)) {
            $endlen = min(($len - strlen($newstr)), $maxlen - 3);
            $newstr .= "..." . $delimter;

            if($right_justify)
                $newstr .= str_pad('', $maxlen - $endlen - 3, ' ');

            $newstr .= "..." . substr($str, $len - $endlen);
        }

        return($newstr);
    }

    return($str);
}
up
-8
webmaster at cafe-clope dot net
19 years ago
A comprehensive concatenation function, that works with array and strings

<?php
function str_cat() {
 
$args = func_get_args() ;
   
 
// Asserts that every array given as argument is $dim-size.
  // Keys in arrays are stripped off.
  // If no array is found, $dim stays unset.
 
foreach($args as $key => $arg) {
    if(
is_array($arg)) {
      if(!isset(
$dim))
       
$dim = count($arg) ;
      elseif(
$dim != count($arg))
        return
FALSE ;
     
$args[$key] = array_values($arg) ;
    }
  }
       
 
// Concatenation
 
if(isset($dim)) {
   
$result = array() ;
    for(
$i=0;$i<$dim;$i++) {
     
$result[$i] = '' ;
      foreach(
$args as $arg)
       
$result[$i] .= ( is_array($arg) ? $arg[$i] : $arg ) ;
    }
    return
$result ;
  } else {
    return
implode($args) ;
  }
}
?>

A simple example :

<?php
str_cat
(array(1,2,3), '-', array('foo' => 'foo', 'bar' => 'bar', 'noop' => 'noop')) ;
?>

will return :
Array (
  [0] => 1-foo
  [1] => 2-bar
  [2] => 3-noop
)

More usefull :

<?php
$myget
= $_GET ; // retrieving previous $_GET values
$myget['foo'] = 'b a r' ; // changing one value
$myget = str_cat(array_keys($myget), '=', array_map('rawurlencode', array_values($myget))) ;
$querystring = implode(ini_get('arg_separator.output'), $myget)) ;
?>

will return a valid querystring with some values changed.

Note that <?php str_cat('foo', '&', 'bar') ; ?> will return 'foo&bar', while <?php str_cat(array('foo'), '&', 'bar') ; ?> will return array(0 => foo&bar)
up
-9
terry dot greenlaw at logicalshift dot com
20 years ago
Here's a simpler "simplest" way to toggle through a set of 1..n colors for web backgrounds:

<?php
$colours
= array('#000000', '#808080', '#A0A0A0', '#FFFFFF');

// Get a colour
$color = next($colors) or $color = reset($colors);
?>

The code doesn't need to know anything about the number of elements being cycled through. That way you won't have to tracking down all the code when changing the number of colors or the color values.
up
-10
Anonymous
19 years ago
In response to hackajar <matt> yahoo <trot> com,

No string-to-array function exists because it is not needed. If you reference a string with an offset like you do with an array, the character at that offset will be return. This is documented in section III.11's "Strings" article under the "String access and modification by character" heading.
up
-10
str at maphpia dot com
11 years ago
I was looking for a function to find the common substring in 2 different strings. I tried both the mb_string_intersect and string_intersect functions listed here but didn't work for me. I found the algorithm at http://en.wikibooks.org/wiki/Algorithm_implementation/Strings/Longest_common_substring#PHP so here I post you the function

<?php

/**
* Finds the matching string between 2 strings
*
* @param string $string1
* @param string $string2
* @param number $minChars
*
* @return NULL|string
*
* @link http://en.wikibooks.org/wiki/Algorithm_implementation/Strings/Longest_common_substring#PHP
*/
function string_intersect($string_1, $string_2)
{
   
$string_1_length = strlen($string_1);
   
$string_2_length = strlen($string_2);
   
$return          = "";

    if (
$string_1_length === 0 || $string_2_length === 0) {
       
// No similarities
       
return $return;
    }

   
$longest_common_subsequence = array();

   
// Initialize the CSL array to assume there are no similarities
   
for ($i = 0; $i < $string_1_length; $i++) {
       
$longest_common_subsequence[$i] = array();
        for (
$j = 0; $j < $string_2_length; $j++) {
           
$longest_common_subsequence[$i][$j] = 0;
        }
    }

   
$largest_size = 0;

    for (
$i = 0; $i < $string_1_length; $i++) {
        for (
$j = 0; $j < $string_2_length; $j++) {
           
// Check every combination of characters
           
if ($string_1[$i] === $string_2[$j]) {
               
// These are the same in both strings
               
if ($i === 0 || $j === 0) {
                   
// It's the first character, so it's clearly only 1 character long
                   
$longest_common_subsequence[$i][$j] = 1;
                } else {
                   
// It's one character longer than the string from the previous character
                   
$longest_common_subsequence[$i][$j] = $longest_common_subsequence[$i - 1][$j - 1] + 1;
                }

                if (
$longest_common_subsequence[$i][$j] > $largest_size) {
                   
// Remember this as the largest
                   
$largest_size = $longest_common_subsequence[$i][$j];
                   
// Wipe any previous results
                   
$return       = "";
                   
// And then fall through to remember this new value
               
}

                if (
$longest_common_subsequence[$i][$j] === $largest_size) {
                   
// Remember the largest string(s)
                   
$return = substr($string_1, $i - $largest_size + 1, $largest_size);
                }
            }
           
// Else, $CSL should be set to 0, which it was already initialized to
       
}
    }

   
// Return the list of matches
   
return $return;
}
up
-9
admin at rotarymulundeast dot org
18 years ago
Here's an easier way to find nth...

function nth($numbex){
  if ($numbex%10 == 1 && $numbex%100 != 11) $sth='st';
  elseif ($numbex%10 == 2 && $numbex%100 != 12) $sth='nd';
  elseif ($numbex%10 == 3 && $numbex%100 != 13) $sth='rd';
  else $sth = 'th';
  return $sth;
}

there is is no need to check if the user has entered a non-integer as we may be using this function for expressing variables as well eg. ith value of x , nth root of z ,etc...
up
-10
james dot d dot baker at gmail dot com
19 years ago
<?php
/*
Written By James Baker, May 27th 2005

sentenceCase($string);
    $string: The string to convert to sentence case.

Converts a string into proper sentence case (First letter of each sentance capital, all the others smaller)

Example Usage:
echo sentenceCase("HELLO WORLD!!! THIS IS A CAPITALISED SENTENCE. this isn't.");

Returns:
Hello world!!! This is a capitalised sentence. This isn't.
*/

function sentenceCase($s){
   
$str = strtolower($s);
   
$cap = true;
   
    for(
$x = 0; $x < strlen($str); $x++){
       
$letter = substr($str, $x, 1);
        if(
$letter == "." || $letter == "!" || $letter == "?"){
           
$cap = true;
        }elseif(
$letter != " " && $cap == true){
           
$letter = strtoupper($letter);
           
$cap = false;
        }
       
       
$ret .= $letter;
    }
   
    return
$ret;
}
?>
up
-9
navarr at gmail dot com
18 years ago
stripos for PHP4.x

<?php
 
function stripos($haystack,$needle) {
    return
strpos(strtoupper($haystack),strtoupper($needle));
  }
?>
up
-11
SteveRusin
18 years ago
The functions below:

function beginsWith( $str, $sub )
function endsWith( $str, $sub )

Are correct, but flawed.  You'd need to use the === operator instead:

function beginsWith( $str, $sub ) {
   return ( substr( $str, 0, strlen( $sub ) ) === $sub );
}
function endsWith( $str, $sub ) {
   return ( substr( $str, strlen( $str ) - strlen( $sub ) ) === $sub );
}

Otherwise, endsWith would return "foobar.0" ends with ".0" as well as "0" or "00" or any amount of zeros because numerically .0 does equal 0.
up
-11
Anonymous
19 years ago
to: james dot d dot baker at gmail dot com

PHP has a builtin function  for doing what your function does,

http://php.net/ucfirst
http://php.net/ucwords
up
-13
Verdauga
16 years ago
Just a note in regards to bloopletech a few posts down:

The word "and" should not be used when converting numbers to text.  "And" (at least in US English) should only be used to indicate the decimal place.

Example:
1,796,706 => one million, seven hundred ninety-six thousand, seven hundred six.
594,359.34 => five hundred ninety four thousand, three hundred fifty nine and thirty four hundredths
up
-13
andy a t onesandzeros d o t biz
19 years ago
I use these little doo-dads quite a bit. I just thought I'd share them and maybe save someone a little time. No biggy. :)

// returns true if $str begins with $sub
function beginsWith( $str, $sub ) {
    return ( substr( $str, 0, strlen( $sub ) ) == $sub );
}

// return tru if $str ends with $sub
function endsWith( $str, $sub ) {
    return ( substr( $str, strlen( $str ) - strlen( $sub ) ) == $sub );
}

// trims off x chars from the front of a string
// or the matching string in $off is trimmed off
function trimOffFront( $off, $str ) {
    if( is_numeric( $off ) )
        return substr( $str, $off );
    else
        return substr( $str, strlen( $off ) );
}

// trims off x chars from the end of a string
// or the matching string in $off is trimmed off
function trimOffEnd( $off, $str ) {
    if( is_numeric( $off ) )
        return substr( $str, 0, strlen( $str ) - $off );
    else
        return substr( $str, 0, strlen( $str ) - strlen( $off ) );
}
up
-11
mike &#34;eyes&#34; moe
17 years ago
Here is a truly random string generator it uses the most common string functions it will work on anywhere.

<?php
   
function random_string($max = 20){
       
$chars = explode(" ", "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 0 1 2 3 4 5 6 7 8 9");
        for(
$i = 0; $i < $max; $i++){
           
$rnd = array_rand($chars);
           
$rtn .= base64_encode(md5($chars[$rnd]));
        }
        return
substr(str_shuffle(strtolower($rtn)), 0, $max);
    }
?>
up
-12
m
17 years ago
Regarding the code for the function beginsWith($str, $sub), I found that it has problems when only one character is present after the string searched for. I found that this works better instead:

<?php
function beginsWith($str, $sub) {
    return (
strncmp($str, $sub, strlen($sub)) == 0);
}
?>
up
-13
Stephen Dewey
15 years ago
If you want a function to return all text in a string up to the Nth occurrence of a substring, try the below function.

Works in PHP >= 5.

(Pommef provided another sample function for this purpose below, but I believe it is incorrect.)

<?php

// Returns all of $haystack up to (but excluding) the $n_occurrence occurrence of $needle. Therefore:
//        If there are < $n_occurrence occurrences of $needle in $haystack, the entire string will be returned.
//        If there are >= $n_occurrence occurrences of $needle in $haystack, the returned string will end before the $n_occurrence'th needle.
// This function only makes sense for $n_occurrence >= 1
function nsubstr($needle, $haystack, $n_occurrence)
{
   
// After exploding by $needle, every entry in $arr except (possibly) part of the last entry should have its content returned.
   
$arr = explode($needle,$haystack,$n_occurrence);
   
// Examine last entry in $arr. If it contains $needle, cut out all text except for the text before $needle.
   
$last = count($arr) - 1;
   
$pos_in_last = strpos($arr[$last],$needle);
   
    if (
$pos_in_last !== false)
       
$arr[$last] = substr($arr[$last],0,$pos_in_last);
   
    return
implode($needle,$arr);
}

$string = 'd24jkdslgjldk2424jgklsjg24jskgldjk24';

print
'S:  ' . $string . '<br>';
print
'1: ' . nsubstr('24',$string,1) . '<br>';
print
'2: ' . nsubstr('24',$string,2) . '<br>';
print
'3: ' . nsubstr('24',$string,3) . '<br>';
print
'4: ' . nsubstr('24',$string,4) . '<br>';
print
'5: ' . nsubstr('24',$string,5) . '<br>';
print
'6: ' . nsubstr('24',$string,6) . '<br>';
print
'7: ' . nsubstr('24',$string,7) . '<br>';

/*
// prints:
S: d24jkdslgjldk2424jgklsjg24jskgldjk24
1: d
2: d24jkdslgjldk
3: d24jkdslgjldk24
4: d24jkdslgjldk2424jgklsjg
5: d24jkdslgjldk2424jgklsjg24jskgldjk
6: d24jkdslgjldk2424jgklsjg24jskgldjk24
7: d24jkdslgjldk2424jgklsjg24jskgldjk24
*/

?>

Note that this function can be combined with wordwrap() to accomplish a routine but fairly difficult web design goal, namely, limiting inline HTML text to a certain number of lines. wordwrap() can break your string using <br>, and then you can use this function to only return text up to the N'th <br>.

You will still have to make a conservative guess of the max number of characters per line with wordwrap(), but you can be more precise than if you were simply truncating a multiple-line string with substr().

See example:

<?php

$text
= 'Lorem ipsum dolor sit amet, consectetuer adipiscing elit. Pellentesque id massa. Duis sollicitudin ipsum vel diam. Aliquam pulvinar sagittis felis. Nullam hendrerit semper elit. Donec convallis mollis risus. Cras blandit mollis turpis. Vivamus facilisis, sapien at tincidunt accumsan, arcu dolor suscipit sem, tristique convallis ante ante id diam. Curabitur mollis, lacus vel gravida accumsan, enim quam condimentum est, vitae rutrum neque magna ac enim.';

$wrapped_text = wordwrap($text,100,'<br>',true);

$three_lines = nsubstr('<br>',$wrapped_text,3);

print
'<br><br>' . $three_lines;

$four_lines = nsubstr('<br>',$wrapped_text,4);

print
'<br><br>' . $four_lines;

/*
prints:

Lorem ipsum dolor sit amet, consectetuer adipiscing elit. Pellentesque id massa. Duis sollicitudin
ipsum vel diam. Aliquam pulvinar sagittis felis. Nullam hendrerit semper elit. Donec convallis
mollis risus. Cras blandit mollis turpis. Vivamus facilisis, sapien at tincidunt accumsan, arcu

Lorem ipsum dolor sit amet, consectetuer adipiscing elit. Pellentesque id massa. Duis sollicitudin
ipsum vel diam. Aliquam pulvinar sagittis felis. Nullam hendrerit semper elit. Donec convallis
mollis risus. Cras blandit mollis turpis. Vivamus facilisis, sapien at tincidunt accumsan, arcu
dolor suscipit sem, tristique convallis ante ante id diam. Curabitur mollis, lacus vel gravida

*/

?>
up
-16
administrador(ensaimada)sphoera(punt)com
18 years ago
I've prepared this simple function to obtain a string delimited between tags (not only XML tags!). Anybody needs something like this?.

<?php

function get_string_between($string, $start, $end){
   
$string = " ".$string;
    
$ini = strpos($string,$start);
     if (
$ini == 0) return "";
    
$ini += strlen($start);    
    
$len = strpos($string,$end,$ini) - $ini;
     return
substr($string,$ini,$len);
}

$string = "this [custom] function is useless!!";
echo
get_string_between($string,"[","]");
// must return "custom";
?>
more functions at http://www.sphoera.com
up
-13
rh at richardhoward dot net
19 years ago
<?php
/**
Utility class: static methods for cleaning & escaping untrusted (i.e.
user-supplied) strings.

Any string can (usually) be thought of as being in one of these 'modes':

pure = what the user actually typed / what you want to see on the page /
       what is actually stored in the DB
gpc  = incoming GET, POST or COOKIE data
sql  = escaped for passing safely to RDBMS via SQL (also, data from DB
       queries and file reads if you have magic_quotes_runtime on--which
       is rare)
html = safe for html display (htmlentities applied)

Always knowing what mode your string is in--using these methods to
convert between modes--will prevent SQL injection and cross-site scripting.

This class refers to its own namespace (so it can work in PHP 4--there is no
self keyword until PHP 5). Do not change the name of the class w/o changing
all the internal references.

Example usage: a POST value that you want to query with:
$username = Str::gpc2sql($_POST['username']);
*/

//This sets SQL escaping to use slashes; for Sybase(/MSSQL)-style escaping
// ( ' --> '' ), set to true.
define('STR_SYBASE', false);

class
Str {
    function
gpc2sql($gpc, $maxLength = false)
    {
        return
Str::pure2sql(Str::gpc2pure($gpc), $maxLength);
    }
    function
gpc2html($gpc, $maxLength = false)
    {
        return
Str::pure2html(Str::gpc2pure($gpc), $maxLength);
    }
    function
gpc2pure($gpc)
    {
        if (
ini_get('magic_quotes_sybase'))
           
$pure = str_replace("''", "'", $gpc);
        else
$pure = get_magic_quotes_gpc() ? stripslashes($gpc) : $gpc;
        return
$pure;
    }
    function
html2pure($html)
    {
        return
html_entity_decode($html);
    }
    function
html2sql($html, $maxLength = false)
    {
        return
Str::pure2sql(Str::html2pure($html), $maxLength);
    }
    function
pure2html($pure, $maxLength = false)
    {
        return
$maxLength ? htmlentities(substr($pure, 0, $maxLength))
                          :
htmlentities($pure);
    }
    function
pure2sql($pure, $maxLength = false)
    {
        if (
$maxLength) $pure = substr($pure, 0, $maxLength);
        return (
STR_SYBASE)
               ?
str_replace("'", "''", $pure)
               :
addslashes($pure);
    }
    function
sql2html($sql, $maxLength = false)
    {
       
$pure = Str::sql2pure($sql);
        if (
$maxLength) $pure = substr($pure, 0, $maxLength);
        return
Str::pure2html($pure);
    }
    function
sql2pure($sql)
    {
        return (
STR_SYBASE)
               ?
str_replace("''", "'", $sql)
               :
stripslashes($sql);
    }
}
?>
up
-18
php at moechofe dot com
19 years ago
<?php
 
/*
  * str_match
  *
  * return a string with only cacacteres defined in a expression return false if the expression is not valid
  *
  * @param $str string the string
  * @param $match the expression based on the class definition off a PCRE regular expression.
  *   the '[', ']', '\' and '^' at class start need to be escaped.
  *   like : -a-z0-9_@.
  */
 
function str_match( $str, $match )
  {
   
$return = '';
    if(
eregi( '(.*)', $match, $class ) )
    {
     
$match = '['.$regs[1].']';
      for(
$i=0; $i<strlen($str); $i++ )
      if(
ereg( '['.$class[1].']', $str[$i] ) )
     
$return .= $str{$i};
      return
$return;
    }
    else return
false;
  }

 
/*
  * example
  * accept only alphanum caracteres from the GET/POST parameters 'a'
  */

 
if( ! empty($_REQUEST['a']) )
   
$_REQUEST['a'] = str_match( $_REQUEST['a'], 'a-zA-Z0-9' );
  else
   
$_REQUEST['a'] = 'default';
?>
up
-14
Pommef
19 years ago
Example: Give me everything up to the fourth occurance of '/'.

<?php

   $haystack
= "/home/username/www/index.php";
  
$needle = "/";
 
   function
strnpos($haystack, $needle, $occurance, $pos = 0) {
       
       
$res = implode($needle,$haystack);
       
       
$res = array_slice($res, $pos$occurance);
       
        return
explode ($needle,$res);
    }
?>
up
-21
Tomek Rychtyk
12 years ago
Get the intersection of two strings using array_intersect

<?php

function string_intersect($string1, $string2)
{
   
$array1 = $array2 = array();

    for(
$i = 0, $j = 0, $s1_len = strlen($string1), $s2_len = strlen($string2);($i < $s1_len) || ($j < $s2_len); $i++, $j++) {
        if(
$i < $s1_len) {
           
$array1[] = $string1[$i];
        }
        if(
$j < $s2_len) {
           
$array2[] = $string2[$j];
        }
    }

    return
implode('', array_intersect($array1, $array2));
}

?>

For more advanced comparison you can use array_uintersect as well.
To Top