substr

For getting a substring of UTF-8 characters, I highly recommend mb_substr

<?php
        $utf8string = "cakeæøå";

        echo substr($utf8string,0,5);
        // output cake#
        echo mb_substr($utf8string,0,5,'UTF-8');
        //output cakeæ
?>

may be by following functions will be easier to extract the needed sub parts from a string:

<?php
after ('@', 'biohazard@online.ge');
//returns 'online.ge'
//from the first occurrence of '@'

before ('@', 'biohazard@online.ge');
//returns 'biohazard'
//from the first occurrence of '@'

between ('@', '.', 'biohazard@online.ge');
//returns 'online'
//from the first occurrence of '@'

after_last ('[', 'sin[90]*cos[180]');
//returns '180]'
//from the last occurrence of '['

before_last ('[', 'sin[90]*cos[180]');
//returns 'sin[90]*cos['
//from the last occurrence of '['

between_last ('[', ']', 'sin[90]*cos[180]');
//returns '180'
//from the last occurrence of '['
?>

here comes the source:

<?php

    function after ($this, $inthat)
    {
        if (!is_bool(strpos($inthat, $this)))
        return substr($inthat, strpos($inthat,$this)+strlen($this));
    };

    function after_last ($this, $inthat)
    {
        if (!is_bool(strrevpos($inthat, $this)))
        return substr($inthat, strrevpos($inthat, $this)+strlen($this));
    };

    function before ($this, $inthat)
    {
        return substr($inthat, 0, strpos($inthat, $this));
    };

    function before_last ($this, $inthat)
    {
        return substr($inthat, 0, strrevpos($inthat, $this));
    };

    function between ($this, $that, $inthat)
    {
        return before ($that, after($this, $inthat));
    };

    function between_last ($this, $that, $inthat)
    {
     return after_last($this, before_last($that, $inthat));
    };

// use strrevpos function in case your php version does not include it
function strrevpos($instr, $needle)
{
    $rev_pos = strpos (strrev($instr), strrev($needle));
    if ($rev_pos===false) return false;
    else return strlen($instr) - $rev_pos - strlen($needle);
};
?>

[English]
I created python similar accesing list or string with php substr & strrev functions.

Use: str($string,$pattern)

About the python pattern,
http://docs.python.org/release/1.5.1p1/tut/strings.html
http://effbot.org/zone/python-list.htm

About of pattern structures
[start:stop:step]

Example,
<?php
$s = "fatihmertdogancan";
echo str($s,"1:9:-2");
echo "<br/>";
echo str($s,"1:-3:-2");
echo "<br/>";
echo str($s,"1:-11:-5");
echo "<br/>";
echo str($s,"1:9:4");
?>

Output,
thetoacn
eht
aom
htan

This is function phpfiddle link: http://phpfiddle.org/main/code/e82-y5d

or source;

<?php
    function str($str,$pattern){
        //[start:stop:step]
        //pattern ->            ([-]?[0-9]*|\s):([-]?[0-9]*|\s):([-]?[0-9]*|\s)
        preg_match("/([-]?[0-9]*|\s?):([-]?[0-9]*|\s?):?([-]?[0-9]*|\s?)/", $pattern, $yakala);
        $start = $yakala[1];
        $stop = $yakala[2];
        $step = $yakala[3];
        
        if(empty($start) && empty($stop) && $step == "-1"){//istisna durum
            return strrev($str);
        }else if(empty($start) && empty($stop) && isset($step)){//istisna durum
            $rev = "";
            $yeni = "";
            if($step[0] == "-" && $stop != "-1"){$rev = "VAR";}
            $atla = abs($step);
            for($i = 0; $i <= strlen($str); $i++){
                $offset = $i*$atla;
                if(isset($str[$offset])){
                    $yeni = $yeni.$str[$offset];
                }
            }
            if($rev != "VAR"){
                return substr($yeni,0,strlen($str)-1);
                //"hepsi boş, step dolu o da +";
            }else{
                return strrev(substr($yeni,0,strlen($str)-1));
                //"hepsi boş, step dolu o da -";
            }
        }
        
        if(empty($start) && empty($stop) && empty($step)){
            return $str;
            //"hepsi boş";
        }else if(empty($start)){
            if(isset($stop) && empty($step)){
                $rev = "";
                if($stop[0] == "-"){$rev = "VAR";}
                if($rev != "VAR"){
                    return substr($str,0,$stop);
                    //"start ve step boş, stop dolu"
                }else{
                    return strrev(substr($str,0,$stop));
                    //"start ve step boş, stop -1";
                }
            }else if(isset($stop) && isset($step)){
                $rev = "";
                if($stop[0] == "-"){$rev = "VAR";}
                $yeni = "";
                if($step == 1){
                    if($rev != "VAR"){
                        return $str;
                        //"start boş, stop ve step dolu, step 1";
                    }else{
                        return strrev(substr($str,0,abs($stop))); //abs -> mutlak değer (-5 = 5)
                        //"start boş, stop -, step dolu, step 1";
                    }
                }else{
                    $atla = abs($step);
                    for($i = 0; $i <= strlen($str); $i++){
                        $offset = $i*$atla;
                        if(isset($str[$offset])){
                            $yeni = $yeni.$str[$offset];
                        }
                    }
                    if($rev != "VAR"){
                        return substr($yeni,0,$stop);
                        //"start boş, step ve stop dolu";
                    }else{
                        return strrev(substr($yeni,0,abs($stop)));
                        //"start boş, step ve stop -";
                    }
                }
            }
        //start boş değilse
        }else if(!empty($start)){
            if(isset($stop) && empty($step)){
                $rev = "";
                if($stop[0] == "-"){$rev = "VAR";}
                if($rev != "VAR"){
                    return substr($str,$start,$stop);
                    //return "step boş, start ve stop dolu";
                }else{
                    return strrev(substr($str,0,abs($stop)));
                    //"step boş, start ve stop dolu, stop -";
                }
            }else if(isset($stop) && isset($step)){
                
                //hepsi dolu
                $rev = "";
                if($stop[0] == "-"){$rev = "VAR";}
                $yeni = "";
                if($step == 1){
                    if($rev != "VAR"){
                        return substr($str,$start,$stop);
                        //"hepsi dolu, step 1";
                    }else{
                        return substr($str,$start,abs($stop));
                        //"hepsi dolu, step 1, stop -";
                    }
                }else{
                    if($stop[0] == "-"){$rev = "VAR";}
                    $atla = abs($step);
                    for($i = 0; $i <= strlen($str); $i++){
                        $offset = $i*$atla;
                        if(isset($str[$offset])){
                            $yeni = $yeni.$str[$offset];
                        }
                    }
                    if($rev != "VAR"){
                        return substr($yeni,$start,$stop);
                        //"hepsi dolu";
                    }else{
                        return strrev(substr($yeni,$start,abs($stop)));
                        //"hepsi dolu, stop -";
                    }
                }
            }
        }
    }
?>

Good works..

Coming to PHP from classic ASP I am used to the Left() and Right() functions built into ASP so I did a quick PHPversion. hope these help someone else making the switch

function left($str, $length) {
    return substr($str, 0, $length);
}

function right($str, $length) {
    return substr($str, -$length);
}

<?Php 

### SUB STRING  BY WORD USING substr() and strpos()  #####

### THIS SCRIPT WILL RETURN PART OF STRING  WITHOUT WORD BREAK ###

$description = ‘your description here your description here your description here your description here your description here your description here your description hereyour description here your description here’  // your description here .

$no_letter = 30 ;

if(strlen($desctiption) > 30 )
{
     echo substr($description,0,strpos($description,’ ‘,30));             //strpos to find ‘ ‘ after 30 characters.
}
else {
     echo $description;
}

?>

Be aware of a slight inconsistency between substr and mb_substr

mb_substr("", 4);      returns empty string

substr("", 4);              returns boolean false

tested in PHP 7.1.11 (Fedora 26) and PHP 5.4.16 (CentOS 7.4)

I wanted to work out the fastest way to get the first few characters from a string, so I ran the following experiment to compare substr, direct string access and strstr:

<?php
/* substr access */
beginTimer();
for ($i = 0; $i < 1500000; $i++){
    $opening = substr($string,0,11);
    if ($opening == 'Lorem ipsum'){
        true;
    }else{
        false;
    }
}
$endtime1 = endTimer();

/* direct access */
beginTimer();
for ($i = 0; $i < 1500000; $i++){
    if ($string[0] == 'L' && $string[1] == 'o' && $string[2] == 'r' && $string[3] == 'e' && $string[4] == 'm' && $string[5] == ' ' && $string[6] == 'i' && $string[7] == 'p' && $string[8] == 's' && $string[9] == 'u' && $string[10] == 'm'){
        true;
    }else{
        false;
    }
}
$endtime2 = endTimer();

/* strstr access */
beginTimer();
for ($i = 0; $i < 1500000; $i++){
    $opening = strstr($string,'Lorem ipsum');
    if ($opening == true){
        true;
    }else{
        false;
    }
}
$endtime3 = endTimer();

echo $endtime1."\r\n".$endtime2."\r\n".$endtime3;
?>

The string was 6 paragraphs of Lorem Ipsum, and I was trying match the first two words. The experiment was run 3 times and averaged. The results were:

(substr) 3.24
(direct access) 11.49
(strstr) 4.96

(With standard deviations 0.01, 0.02 and 0.04)

THEREFORE substr is the fastest of the three methods for getting the first few letters of a string.

This returns the portion of str specified by the start and length parameters..
It can performs multi-byte safe on number of characters. like mb_strcut() ...

Note:
1.Use it like this bite_str(string str, int start, int length [,byte of on string]);
2.First character's position is 0. Second character position is 1, and so on...
3.$byte is one character length of your encoding, For example: utf-8 is "3", gb2312 and big5 is "2"...you can use the function strlen() get it... 
Enjoy it :) ...

--- Bleakwind
QQ:940641
http://www.weaverdream.com

PS:I'm sorry my english is too poor... :( 

<?php
// String intercept By Bleakwind
// utf-8:$byte=3 | gb2312:$byte=2 | big5:$byte=2
function bite_str($string, $start, $len, $byte=3)
{
    $str     = "";
    $count   = 0;
    $str_len = strlen($string);
    for ($i=0; $i<$str_len; $i++) {
        if (($count+1-$start)>$len) {
            $str  .= "...";
            break;
        } elseif ((ord(substr($string,$i,1)) <= 128) && ($count < $start)) {
            $count++;
        } elseif ((ord(substr($string,$i,1)) > 128) && ($count < $start)) {
            $count = $count+2;
            $i     = $i+$byte-1;
        } elseif ((ord(substr($string,$i,1)) <= 128) && ($count >= $start)) {
            $str  .= substr($string,$i,1);
            $count++;
        } elseif ((ord(substr($string,$i,1)) > 128) && ($count >= $start)) {
            $str  .= substr($string,$i,$byte);
            $count = $count+2;
            $i     = $i+$byte-1;
        }
    }
    return $str;
}

// Test
$str = "123456???ֽ?123456?ַ???123456??ȡ????";
for($i=0;$i<30;$i++){
    echo "<br>".bite_str($str,$i,20);    
}
?>

Hmm ... this is a script I wrote, whitch is very similar to substr, but it isn't takes html and bbcode for counting and it takes portion of string and show avoided (html & bbcode) tags too ;] 
Specially usefull for show part of serach result included html and bbcode tags

<?php

/**
 * string csubstr ( string string, int start [, int length] )
 *
 * @author FanFataL
 * @param string string
 * @param int start
 * @param [int length]
 * @return string
 */
function csubstr($string, $start, $length=false) {
    $pattern = '/(\[\w+[^\]]*?\]|\[\/\w+\]|<\w+[^>]*?>|<\/\w+>)/i';
    $clean = preg_replace($pattern, chr(1), $string);
    if(!$length)
        $str = substr($clean, $start);
    else {
        $str = substr($clean, $start, $length);
        $str = substr($clean, $start, $length + substr_count($str, chr(1)));
    }
    $pattern = str_replace(chr(1),'(.*?)',preg_quote($str));
    if(preg_match('/'.$pattern.'/is', $string, $matched))
        return $matched[0];
    return $string;
}

?>

Using this is similar to simple substr.

Greatings ;]
...

To quickly trim an optional trailing slash off the end of a path name:

if (substr( $path, -1 ) == '/') $path = substr( $path, 0, -1 );

If you want to have a string BETWEEN two strings, just use this function:

<?php
function get_between($input, $start, $end)
{
  $substr = substr($input, strlen($start)+strpos($input, $start), (strlen($input) - strpos($input, $end))*(-1));
  return $substr;
}

//Example:

$string = "123456789";
$a = "12";
$b = "9";

echo get_between($string, $a, $b);

//Output:
//345678
?>

I created some functions for entity-safe splitting+lengthcounting:

<?php
function strlen_entities($text)
{
    preg_match_all(
        '/((?:&(?:#[0-9]{2,}|[a-z]{2,});)|(?:[^&])|'.         
        '(?:&(?!\w;)))s',$text,$textarray);
    return count($textarray[0]);
}  
function substr_entities($text,$start,$limit=0)
{
    $return = '';
    preg_match_all(
        '/((?:&(?:#[0-9]{2,}|[a-z]{2,});)|(?:[^&])|'.         
        '(?:&(?!\w;)))s',$text,$textarray);
    $textarray = $textarray[0];
    $numchars = count($textarray)-1;
    if ($start>=$numchars)
        return false;
    if ($start<0)
    {
        $start = ($numchars)+$start+1;
    }
    if ($start>=0)
    {
        if ($limit==0)
        {
            $end=$numchars;
        }
        elseif ($limit>0)
        {
            $end = $start+($limit-1);
        }
        else
        {
            $end = ($numchars)+$limit;
        }

        for ($i=$start;$i<=$end;$i++)
        {
            $return .= $textarray[$i];
        }
        return $return;
    }
}
?>

Drop extensions of a file (even from a file location string)

<?php

$filename = "c:/some dir/abc defg. hi.jklmn";

echo substr($filename, 0, (strlen ($filename)) - (strlen (strrchr($filename,'.'))));

?>

output: c:/some dir/abc defg. hi

Hope it may help somebody like me.. (^_^)

Anyone coming from the Python world will be accustomed to making substrings by using a "slice index" on a string.  The following function emulates basic Python string slice behavior. (A more elaborate version could be made to support array input as well as string, and the optional third "step" argument.)

<?php

function py_slice($input, $slice) {
    $arg = explode(':', $slice);
    $start = intval($arg[0]);
    if ($start < 0) {
        $start += strlen($input);
    }
    if (count($arg) === 1) {
        return substr($input, $start, 1);
    }
    if (trim($arg[1]) === '') {
        return substr($input, $start);
    }
    $end = intval($arg[1]);
    if ($end < 0) {
        $end += strlen($input);
    }
    return substr($input, $start, $end - $start);
}

print py_slice('abcdefg', '2') . "\n";
print py_slice('abcdefg', '2:4') . "\n";
print py_slice('abcdefg', '2:') . "\n";
print py_slice('abcdefg', ':4') . "\n";
print py_slice('abcdefg', ':-3') . "\n";
print py_slice('abcdefg', '-3:') . "\n";

?>

The $slice parameter can be a single character index, or a range separated by a colon. The start of the range is inclusive and the end is exclusive, which may be  counterintuitive. (Eg, py_slice('abcdefg', '2:4') yields 'cd' not 'cde'). A negative range value means to count from the end of the string instead of the beginning. Both the start and end of the range may be omitted; the start defaults to 0 and the end defaults to the total length of the input.

The output from the examples:
c
cd
cdefg
abcd
abcd
efg

You might expect substr('123456', 6) to return an empty string. Instead it returns boolean FALSE.

This behavior should be mentioned in the Return Values section of the manual. Instead it is only mentioned in the Parameters section.

If you need an empty string instead of a boolean FALSE you should typecast the result to a string.

<?php
$a = substr('123456', 6);              // equivalent to $a = FALSE
$a = (string) substr('123456', 6);   // equivalent to $a = '';
?>

Add on to (a function originally written by) "Matias from Argentina": str_format_number function.

Just added handling of $String shorter then $Format by adding a side to start the fill and a string length to the while loop.

<?php
function str_format_number($String, $Format, $Start = 'left'){
    //If we want to fill from right to left incase string is shorter then format
    if ($Start == 'right') {
        $String = strrev($String);
        $Format = strrev($Format); 
    }
    if($Format == '') return $String;
    if($String == '') return $String;    
    $Result = '';
    $FormatPos = 0;
    $StringPos = 0;
    while ((strlen($Format) - 1) >= $FormatPos && strlen($String) > $StringPos) {
        //If its a number => stores it
        if (is_numeric(substr($Format, $FormatPos, 1))) {
            $Result .= substr($String, $StringPos, 1);
            $StringPos++;
            //If it is not a number => stores the caracter
        } else {
            $Result .= substr($Format, $FormatPos, 1);
        }
        //Next caracter at the mask.
        $FormatPos++;
    }
    if ($Start == 'right') $Result = strrev($Result);
    return $Result;
}
?>

Using a 0 as the last parameter for substr().

As per examples
<?php $var = substr($var, 4); ?>

works no problem. However 
<?php $var = substr($var, 4, 0); ?>

will get you nothing. Just a quick heads up

I needed to cut a string after x chars at a  html converted utf-8 text (for example Japanese text like 嬰謰弰脰欰罏). 
The problem was, the different length of the signs, so I wrote the following function to handle that.
Perhaps it helps.

<?php

function html_cutstr ($str, $len)
{
    if (!preg_match('/\&#[0-9]*;.*/i', $str))
    {
        $rVal = strlen($str, $len);
        break;
    }

    $chars = 0;
    $start = 0;
    for($i=0; $i < strlen($str); $i++)
    {
        if ($chars >= $len)
        break;

        $str_tmp = substr($str, $start, $i-$start);
        if (preg_match('/\&#[0-9]*;.*/i', $str_tmp))
        {
            $chars++;
            $start = $i;
        }
    }
    $rVal = substr($str, 0, $start);
    if (strlen($str) > $start)
    $rVal .= " ...";
    return $rVal;
}
?>

I needed a function like lpad from oracle, or right from SQL
 then I use this code :

<?php
function right($string,$chars)
{
    $vright = substr($string, strlen($string)-$chars,$chars);
    return $vright;
    
}

    echo right('0r0j4152',4);
?>

Result: 
 4152
------------------------------------------------
This function is really simple, I just wanted to share, maybe helps someone out there.  

regards,

Shortens the filename and its expansion has seen.

<?php
$file = "Hellothisfilehasmorethan30charactersandthisfayl.exe";

function funclongwords($file)
{
if (strlen($file) > 30)
{
$vartypesf = strrchr($file,".");
$vartypesf_len = strlen($vartypesf);
$word_l_w = substr($file,0,15);
$word_r_w = substr($file,-15);
$word_r_a = substr($word_r_w,0,-$vartypesf_len);

return $word_l_w."...".$word_r_a.$vartypesf;
}
else
return $file;
}
// RETURN: Hellothisfileha...andthisfayl.exe
?>

Regarding the utf8_substr function from lmak: The pattern '/./u' doesn't match newline characters. This means that the substring from 0 to the total length of the string will miss the number of characters in the end matching the number of newlines in the string. To fix this one can add the s modifier (PCRE_DOTALL) in the pattern:

<?php
function utf8_substr($str,$start)
{
   preg_match_all("/./su", $str, $ar);

   if(func_num_args() >= 3) {
       $end = func_get_arg(2);
       return join("",array_slice($ar[0],$start,$end));
   } else {
       return join("",array_slice($ar[0],$start));
   }
}
?>

Just a little function to cut a string by the wanted amount. Works in both directions.

<?php
function cutString($str, $amount = 1, $dir = "right")
{
  if(($n = strlen($str)) > 0)
  {
    if($dir == "right")
    {
      $start = 0;
      $end = $n-$amount;
    } elseif( $dir == "left") {
      $start = $amount;
      $end = $n;
    }
    
    return substr($str, $start, $end);
  } else return false;
}
?>

Enjoy ;)

<?php

//removes string from the end of other

function removeFromEnd($string, $stringToRemove) {
    $stringToRemoveLen = strlen($stringToRemove);
    $stringLen = strlen($string);
    
    $pos = $stringLen - $stringToRemoveLen;

    $out = substr($string, 0, $pos);

    return $out;
}

$string = 'picture.jpg.jpg';
$string = removeFromEnd($string, '.jpg');
?>

Here we have gr8 function which simply convert ip address to a number using substr with negative offset.
You can need it if you want to compare some IP addresses converted to a numbers. 
For example when using ip2country, or eliminating same range of ip addresses from your website :D

<?php

function ip2no($val)
{    
    list($A,$B,$C,$D)    =    explode(".",$val);
    return 
        substr("000".$A,-3).
        substr("000".$B,-3).
        substr("000".$C,-3).
        substr("000".$D,-3);
}

$min        =    ip2no("10.11.1.0");
$max        =    ip2no("111.11.1.0");
$visitor    =    ip2no("105.1.20.200");

if($min<$visitor && $visitor<$max)    
    {    echo 'Welcome !';    }
else    
    {    echo 'Get out of here !';    }

?>

<?php

/**
 * string substrpos(string $str, mixed $start [[, mixed $end], boolean $ignore_case])
 *
 * If $start is a string, substrpos will return the string from the position of the first occuring $start to $end
 *
 * If $end is a string, substrpos will return the string from $start to the position of the first occuring $end
 *
 * If the first character in (string) $start or (string) $end is '-', the last occuring string will be used.
 *
 * If $ignore_case is true, substrpos will not care about the case.
 * If $ignore_case is false (or anything that is not (boolean) true, the function will be case sensitive.
 *        Both of the above: only applies if either $start or $end are strings.
 *
 * echo substrpos('This is a string with 0123456789 numbers in it.', 5, '5');
 *        // Prints 'is a string with 01234';
 *
 * echo substrpos('This is a string with 0123456789 numbers in it.', '5', 5);
 *        // Prints '56789'
 *
 * echo substrpos('This is a string with 0123456789 numbers in it and two strings.', -60, '-string')
 *        // Prints 's is a string with 0123456789 numbers in it and two '
 *
 * echo substrpos('This is a string with 0123456789 numbers in it and two strings.', -60, '-STRING', true)
 *        // Prints 's is a string with 0123456789 numbers in it and two '
 *
 * echo substrpos('This is a string with 0123456789 numbers in it and two strings.', -60, '-STRING', false)
 *        // Prints 's is a string with 0123456789 numbers in it and two strings.'
 *
 * Warnings:
 *        Since $start and $end both take either a string or an integer:
 *            If the character or string you are searching $str for is a number, pass it as a quoted string.
 *        If $end is (integer) 0, an empty string will be returned.
 *        Since this function takes negative strings ('-search_string'):
 *            If the string your using in $start or $end is a '-' or begins with a '-' escape it with a '\'.
 *            This only applies to the *first* character of $start or $end.
 */

// Define stripos() if not defined (PHP < 5).
if (!is_callable("stripos")) {
    function stripos($str, $needle, $offset = 0) {
        return strpos(strtolower($str), strtolower($needle), $offset);
    }
}

function substrpos($str, $start, $end = false, $ignore_case = false) {
    // Use variable functions
    if ($ignore_case === true) {
        $strpos = 'stripos'; // stripos() is included above in case it's not defined (PHP < 5).
    } else {
        $strpos = 'strpos';
    }

    // If end is false, set it to the length of $str
    if ($end === false) {
        $end = strlen($str);
    }

    // If $start is a string do what's needed to make it an integer position for substr().
    if (is_string($start)) {
        // If $start begins with '-' start processing until there's no more matches and use the last one found.
        if ($start{0} == '-') {
            // Strip off the '-'
            $start = substr($start, 1);
            $found = false;
            $pos = 0;
            while(($curr_pos = $strpos($str, $start, $pos)) !== false) {
                $found = true;
                $pos = $curr_pos + 1;
            }
            if ($found === false) {
                $pos = false;
            } else {
                $pos -= 1;
            }
        } else {
            // If $start begins with '\-', strip off the '\'.
            if ($start{0} . $start{1} == '\-') {
                $start = substr($start, 1);
            }
            $pos = $strpos($str, $start);
        }
        $start = $pos !== false ? $pos : 0;
    }

    // Chop the string from $start to strlen($str).
    $str = substr($str, $start);

    // If $end is a string, do exactly what was done to $start, above.
    if (is_string($end)) {
        if ($end{0} == '-') {
            $end = substr($end, 1);
            $found = false;
            $pos = 0;
            while(($curr_pos = strpos($str, $end, $pos)) !== false) {
                $found = true;
                $pos = $curr_pos + 1;
            }
            if ($found === false) {
                $pos = false;
            } else {
                $pos -= 1;
            }
        } else {
            if ($end{0} . $end{1} == '\-') {
                $end = substr($end, 1);
            }
            $pos = $strpos($str, $end);
        }
        $end = $pos !== false ? $pos : strlen($str);
    }

    // Since $str has already been chopped at $start, we can pass 0 as the new $start for substr()
    return substr($str, 0, $end);
}

?>

When using a value of a wrong type as second parameter , substr() does not return FALSE but NULL although the docs say, it should return FALSE on error.

Prior to PHP 5.3, substr() tries to cast the second parameter to int and doesn't throw any errors. Since PHP 5.3 a warning is thrown.

I have developed a function with a similar outcome to jay's

Checks if the last character is or isnt a space. (does it the normal way if it is)
It explodes the string into an array of seperate works, the effect is... it chops off anything after and including the last space.

<?php
function limit_string($string, $charlimit)
{
    if(substr($string,$charlimit-1,1) != ' ')
    {
        $string = substr($string,'0',$charlimit);
        $array = explode(' ',$string);
        array_pop($array);
        $new_string = implode(' ',$array);

        return $new_string.'...';
    }
    else
    {    
        return substr($string,'0',$charlimit-1).'...';
    }
}
?>

Well this is a script I wrote, what it does is chop up long words with malicious meaning into several parts. This way, a chat in a table will not get stretched anymore.

<?php

function text($string,$limit=20,$chop=10){

$text = explode(" ",$string);
while(list($key, $value) = each($text)){
    $length = strlen($value);
    if($length >=20){
        for($i=0;$i<=$length;$i+=10){
            $new .= substr($value, $i, 10);
            $new .= " ";
        }
         $post .= $new;
    }
    elseif($length <=15){
        $post .= $value;
    }
    $post .= " ";
}
return($post);
}

// for example, this would return:
$output = text("Well this text doesn't get cut up, yet thisssssssssssssssssssssssss one does.", 10, 5);

echo($output); // "Well this text doesn't get cup up, yet thiss sssss sssss sssss sssss sss one does."
?>

I hope it was useful.. :)

If you need to parse utf-8 strings char by char, try this one:

<?php
     $utf8marker=chr(128);
     $count=0;
     while(isset($string{$count})){
       if($string{$count}>=$utf8marker) {
         $parsechar=substr($string,$count,2);
         $count+=2;
       } else {
         $parsechar=$string{$count};
         $count++;
       }
       /* do what you like with parsechar ... , eg.:*/  echo $parsechar."<BR>\r\n";
     }
?>

- it works without mb_substr
- it is fast, because it grabs characters based on indexes  when possible and avoids any count and split functions

If you need just a single character from the string you don't need to use substr(), just use curly braces notation:

<?php
    // both lines will output the 3rd character
    echo substr($my_string, 2, 1);
    echo $my_string{2};  
?>

curly braces syntax is faster and more readable IMHO..

And as always there is bound to be a bug:

<?php
function strlen_entities($text)
{
    preg_match_all(
        '/((?:&(?:#[0-9]{2,}|[a-z]{2,});)|(?:[^&])|'.        
        '(?:&(?!\w;)))s',$text,$textarray);
    return count($textarray[0]);
} 
function substr_entities($text,$start,$limit=0)
{
    $return = '';
    preg_match_all(
        '/((?:&(?:#[0-9]{2,}|[a-z]{2,});)|(?:[^&])|'.        
        '(?:&(?!\w;)))s',$text,$textarray);
    $textarray = $textarray[0];
    $numchars = count($textarray)-1;
    if ($start>=$numchars)
        return false;
    if ($start<0)
    {
        $start = ($numchars)+$start+1;
    }
    if ($start>=0)
    {
        if ($limit==0)
        {
            $end=$numchars;
        }
        elseif ($limit>0)
        {
            $end = $start+($limit-1);
        }
        else
        {
            $end = ($numchars)+$limit;
        }

        for ($i=$start;($i<=$end && isset($textarray[$i]));$i++)
        {
            $return .= $textarray[$i];
        }
        return $return;
    }
}
?>

The javascript charAt equivalent in php of felipe has a little bug. It's necessary to compare the type (implicit) aswell or the function returns a wrong result:
<?php
function charAt($str,$pos) {
    return (substr($str,$pos,1) !== false) ? substr($str,$pos,1) : -1;
}
?>

Truncate a float number. Similar to the Excel trunc function.

<?php
function truncate_number($val,$decimals=2){
    
        $number=array();
        $number=explode(".",$val);            
        $result=0;

        if (count($number)>1){

            $result = $number[0] . "." . substr($number[1],0,$decimals);  
            
        } else {
        
            $result = $val;
            
        }                

        unset($number);

        return $result;    
}

echo truncate_number(99.123456,2); //result = 99.12
echo truncate_number(99.123456,5); //result = 99.12345
echo truncate_number(99.123456,1); //result = 99.1
?>

Prior to PHP 8, specifying length with zero-length strings or non-string values as input can produce potentially unexpected results.

<?php
foreach (['normal', '', true, false, NULL] as $value) {
    echo gettype(substr($value, 0, 10)) . ' ' . substr($value, 0, 10);
}

/*
string normal
boolean
string 1
boolean
boolean
*/
?>