strcasecmp

str_split

Last updated: Fri, 05 Sep 2008

str_word_count

(PHP 4 >= 4.3.0, PHP 5)

str_word_count — Return information about words used in a string

Description

mixed str_word_count ( string $string [, int $format [, string $charlist ]] )

Counts the number of words inside string . If the optional format is not specified, then the return value will be an integer representing the number of words found. In the event the format is specified, the return value will be an array, content of which is dependent on the format . The possible value for the format and the resultant outputs are listed below.

For the purpose of this function, 'word' is defined as a locale dependent string containing alphabetic characters, which also may contain, but not start with "'" and "-" characters.

Parameters

string

The string

format

Specify the return value of this function. The current supported values are:

0 - returns the number of words found
1 - returns an array containing all the words found inside the string
2 - returns an associative array, where the key is the numeric position of the word inside the string and the value is the actual word itself

charlist

A list of additional characters which will be considered as 'word'

Return Values

Returns an array or an integer, depending on the format chosen.

ChangeLog

Version	Description
5.1.0	Added the `charlist` parameter

Examples

Example #1 A str_word_count() example


<?php

$str = "Hello fri3nd, you're
       looking          good today!";

print_r(str_word_count($str, 1));
print_r(str_word_count($str, 2));
print_r(str_word_count($str, 1, 'àáãç3'));

echo str_word_count($str);

?>

The above example will output:

Array
(
    [0] => Hello
    [1] => fri
    [2] => nd
    [3] => you're
    [4] => looking
    [5] => good
    [6] => today
)

Array
(
    [0] => Hello
    [6] => fri
    [10] => nd
    [14] => you're
    [29] => looking
    [46] => good
    [51] => today
)

Array
(
    [0] => Hello
    [1] => fri3nd
    [2] => you're
    [3] => looking
    [4] => good
    [5] => today
)

7


Here is a code for a function str_word_count() compatible with UTF-8. I'm sorry that the comments are in French because I am not very good in English: anyway, these comments only try to explain things that are in PCRE or Unicode documentations.



<?php

    /*

     * Explications du masque pour preg_match_all.

     *

     * La fonction str_word_count standard considère qu'un mot est

     * une séquence de caractères qui contient tous les caractères

     * alphabétiques, et qui peut contenir, mais pas commencer

     * par "'" et "-".

     *

     * Avec Unicode et UTF-8, une lettre peut être un caractères

     * ASCII non accentué tel que "e" ou "E", mais aussi un "é" ou

     * un "É", lequel peut se représenter sous la forme de deux

     * caractères : d'abord le "E" non accentué, puis l'accent tout

     * seul. Une lettre "E" ou "É" fait partie de la classe « L »,

     * un accent de la classe « Mn ».

     *

     * Par ailleurs, "-" n'est plus le seul trait d'union possible.

     * Plutôt que de les lister individuellement, j'ai choisi de

     * tester les caractères de la classe « Pd ». Un inconvénient

     * est que cela inclut aussi le tiret cadratin et d'autres,

     * mais cet inconvénient existait déjà avec str_word_count et

     * le tiret ascii, et en outre il ne concerne pas le français

     * (contrairement à l'anglais, il y a toujours des espaces

     * autour de ces tirets).

     *

     * Enfin, "'" n'est pas non plus la seule apostrophe possible.

     * Mais contrairement aux tirets je teste juste l'apostrophe

     * typographique U+2019 à part au lieu de tester la classe « Pf »

     * car cette dernière contient trop de signes de ponctuation

     * à exclure de la définition d'un mot.

     *

     * Un mot commence donc par une lettre \p{L}, éventuellement

     * accentuée (suivie par un nombre quelconque de \p{Mn}), et

     * ensuite on peut rencontrer un nombre quelconques d'autres

     * lettres (\p{L} et \p{Mn}), de tirets (\p{Pd}) ou d'apostrophes

     * (' et \x{2019}). Tout ceci, bien sûr, dans un masque compatible

     * avec UTF-8 (/u à la fin).

     *

     * Pour les références, voir :

     * http://fr2.php.net/manual/fr/regexp.reference.php #regexp.reference.unicode

     * http://fr2.php.net/manual/fr/reference.pcre.pattern.modifiers.php

     */

    define("WORD_COUNT_MASK", "/\p{L}[\p{L}\p{Mn}\p{Pd}'\x{2019}]*/u");



    function str_word_count_utf8($str)

    {

        return preg_match_all(WORD_COUNT_MASK, $str, $matches);

    }

luce
06-Sep-2008 10:52


Fix Cathy function bug.



Original Cathy function :

[code]

<?php

    function limit_text($text, $limit) {

      if (strlen($text) > $limit) {

          $words = str_word_count($text, 2);

          $pos = array_keys($words);

          $text = substr($text, 0, $pos[$limit]) . '...';

      }

      return $text;

    }

?>

[/code]



This function return undefined index if $limit < $text.



For fix it :

[code]

<?php

    function limit_text($text, $limitstr, $limitwrd) {

      if (strlen($text) > $limitstr) {

          $words = str_word_count($text, 2);

          if ($words > $limitwrd) {

              $pos = array_keys($words);

              $text = substr($text, 0, $pos[$limitwrd]) . '...';

          }

      }

      return $text;

    }

?>

[/code]

aspu.ru
17-Jun-2008 12:21


str_word_count: mixed (string string, [int format], [string charlist])



It can help you to solve problem with digest and some locales. Best regards.

robocop at robotix dot fr
30-Mar-2008 03:21


function count_words($texte)

{

$texte=trim($texte);

$motsinutiles = array(' * ', ' - ', ' : ', '\n');

$texte = str_replace($motsinutiles, '', $texte);

$texte = preg_replace("/\s\s+/", " ", $texte);

$decoupeapostrophes = count(explode('\'', $texte)); //On découpe la chaine en apostrophes

   if($decoupeapostrophes==0) $nombreapostrophes = 0;

   if ($decoupeapostrophes%2==0) {$nombreapostrophes = $decoupeapostrophes/2;}

   else  $nombreapostrophes = ($decoupeapostrophes/2)-0.5;

$nombreespace = count(explode(' ', $texte));



return $nombreespace+$nombreapostrophes;    

}

security_man
24-Dec-2007 09:13


there was a glitch in the code cathy put a post or two ago... should be:



    function limit_text($text, $limit) {

    $text = strip_tags($text);

      $words = str_word_count($text, 2);

      $pos = array_keys($words);

      if (count($words) > $limit) {

          $text = substr($text, 0, $pos[$limit]) . ' ...';

      }

    return $text;

    }



I also added the strip tags in case there is html in there to gum up the works

Adeel Khan
09-Dec-2007 04:01


<?php



/**

 * Returns the number of words in a string.

 * As far as I have tested, it is very accurate.

 * The string can have HTML in it,

 * but you should do something like this first:

 *

 *    $search = array(

 *      '@<script[^>]*?>.*?</script>@si',

 *      '@<style[^>]*?>.*?</style>@siU',

 *      '@<![\s\S]*?--[ \t\n\r]*>@'

 *    );

 *    $html = preg_replace($search, '', $html);

 *

 */



function word_count($html) {



  # strip all html tags

  $wc = strip_tags($html);



  # remove 'words' that don't consist of alphanumerical characters or punctuation

  $pattern = "#[^(\w|\d|\'|\"|\.|\!|\?|;|,|\\|\/|\-|:|\&|@)]+#";

  $wc = trim(preg_replace($pattern, " ", $wc));



  # remove one-letter 'words' that consist only of punctuation

  $wc = trim(preg_replace("#\s*[(\'|\"|\.|\!|\?|;|,|\\|\/|\-|:|\&|@)]\s*#", " ", $wc));



  # remove superfluous whitespace

  $wc = preg_replace("/\s\s+/", " ", $wc);



  # split string into an array of words

  $wc = explode(" ", $wc);



  # remove empty elements

  $wc = array_filter($wc);



  # return the number of words

  return count($wc);



}



?>

Cathy
19-Jul-2007 09:16


A cute little function for truncating text to a given word limit:

<?php

    function limit_text($text, $limit) {

      if (strlen($text) > $limit) {

          $words = str_word_count($text, 2);

          $pos = array_keys($words);

          $text = substr($text, 0, $pos[$limit]) . '...';

      }

      return $text;

    }

?>

geertdd at gmail dot com
13-Jun-2007 05:04


This is an update to my previously posted word_limiter() function. The regex is even more optimized now. Just replace the preg_match line. Change to:



<?php

preg_match('/^\s*(?:\S+\s*){1,'. (int) $limit .'}/', $str, $matches);

geertdd at gmail dot com
28-May-2007 12:52


Here's a very fast word limiter function that preserves the original whitespace.



<?php



function word_limiter($str, $limit = 100, $end_char = '&#8230;') {

    

    if (trim($str) == '')

        return $str;

    

    preg_match('/\s*(?:\S*\s*){'. (int) $limit .'}/', $str, $matches);



    if (strlen($matches[0]) == strlen($str))

        $end_char = '';



    return rtrim($matches[0]) . $end_char;

}



?>



For the thought process behind this function, please read: http://codeigniter.com/forums/viewthread/51788/



Geert De Deckere

joshua dot blake at gmail dot com
03-Mar-2007 01:02


I needed a function which would extract the first hundred words out of a given input while retaining all markup such as line breaks, double spaces and the like. Most of the regexp based functions posted above were accurate in that they counted out a hundred words, but recombined the paragraph by imploding an array down to a string. This did away with any such hopes of line breaks, and thus I devised a crude but very accurate function which does all that I ask it to:



function Truncate($input, $numWords) 

{

  if(str_word_count($input,0)>$numWords)

  {

    $WordKey = str_word_count($input,1);

    $PosKey = str_word_count($input,2);

    reset($PosKey);

    foreach($WordKey as $key => &$value)

    {

        $value=key($PosKey);

        next($PosKey);

    }

    return substr($input,0,$WordKey[$numWords]);

  }

  else {return $input;}

}



The idea behind it? Go through the keys of the arrays returned by str_word_count and associate the number of each word with its character position in the phrase. Then use substr to return everything up until the nth character. I have tested this function on rather large entries and it seems to be efficient enough that it does not bog down at all.



Cheers!



Josh

josh at joshblake.net
01-Mar-2007 11:57


I was interested in a function which returned the first few words out of a larger string.



In reality, I wanted a preview of the first hundred words of a blog entry which was well over that.



I found all of the other functions which explode and implode strings to arrays lost key markups such as line breaks etc.



So, this is what I came up with:



function WordTruncate($input, $numWords) {

if(str_word_count($input,0)>$numWords)

{

    $WordKey = str_word_count($input,1);

    $WordIndex = array_flip(str_word_count($input,2));

    return substr($input,0,$WordIndex[$WordKey[$numWords]]);

}

else {return $input;}

}



While I haven't counted per se, it's accurate enough for my needs. It will also return the entire string if it's less than the specified number of words.



The idea behind it? Use str_word_count to identify the nth word, then use str_word_count to identify the position of that word within the string, then use substr to extract up to that position.



Josh.

30-Jan-2007 04:15


Here is a php work counting function together with a javascript version which will print the same result.



<?php

      //Php word counting function

      function word_count($theString)

      {

        $char_count = strlen($theString);

        $fullStr = $theString." ";

        $initial_whitespace_rExp = "^[[:alnum:]]$";

        

        $left_trimmedStr = ereg_replace($initial_whitespace_rExp,"",$fullStr);

        $non_alphanumerics_rExp = "^[[:alnum:]]$";

        $cleanedStr = ereg_replace($non_alphanumerics_rExp," ",$left_trimmedStr);

        $splitString = explode(" ",$cleanedStr);

        

        $word_count = count($splitString)-1;

        

        if(strlen($fullStr)<2)

        {

          $word_count=0;

        }      

        return $word_count;

      }

?>



<?php

      //Function to count words in a phrase

      function wordCount(theString)

      {

        var char_count = theString.length;

        var fullStr = theString + " ";

        var initial_whitespace_rExp = /^[^A-Za-z0-9]+/gi;

        var left_trimmedStr = fullStr.replace(initial_whitespace_rExp, "");

        var non_alphanumerics_rExp = rExp = /[^A-Za-z0-9]+/gi;

        var cleanedStr = left_trimmedStr.replace(non_alphanumerics_rExp, " ");

        var splitString = cleanedStr.split(" ");

        

        var word_count = splitString.length -1;

        

        if (fullStr.length <2) 

        {

          word_count = 0;

        }      

        return word_count;

      }

?>

Aurelien Marchand
06-Oct-2006 05:06


I found a more reliable way to print, say the first 100 words and then print elipses. My code goes this way;



$threshold_length = 80; // 80 words max

$phrase = "...."; // populate this with the text you want to display

$abody = str_word_count($phrase,2);

if(count($abody) >= $threshold_length){ // gotta cut

  $tbody = array_keys($abody);

  echo "<p>" . substr($phrase,0,$tbody[$threshold_length]) . "... <span class=\"more\"><a href=\"?\">read more</a></span> </p>\n";

} else { // put the whole thing

  echo "<p>" . $phrase . "</p>\n";

}



For any questions, com.iname@artaxerxes2

lwright at psu dot edu
17-Aug-2006 07:51


If you are looking to count the frequency of words, try:



<?php



$wordfrequency = array_count_values( str_word_count( $string, 1) );



?>

rabin at rab dot in
05-Apr-2006 07:03


There is a small bug in the "trim_text" function by "webmaster at joshstmarie dot com" below. If the string's word count is lesser than or equal to $truncation, that function will cut off the last word in the string.





[EDITOR'S NOTE: above referenced note has been removed]





This fixes the problem:





<?php


function trim_text_fixed($string, $truncation = 250) {


    $matches = preg_split("/\s+/", $string, $truncation + 1);


    $sz = count($matches);


    if ( $sz > $truncation ) {


        unset($matches[$sz-1]);


        return implode(' ',$matches);


    }


    return $string;


}


?>

webmaster at joshstmarie dot com
26-Sep-2005 12:58


Trying to make an effiecient word splitter, and "paragraph limiter", eg, limit item text to 100, or 200 words and so-forth.



I don't know how well this compares, but it works nicely.



function trim_text($string, $word_count=100)

{

    $trimmed = "";

    $string = preg_replace("/\040+/"," ", trim($string));

    $stringc = explode(" ",$string);

    echo sizeof($stringc);

    if($word_count >= sizeof($stringc))

    {

        // nothing to do, our string is smaller than the limit.

      return $string;

    }

    elseif($word_count < sizeof($stringc))

    {

        // trim the string to the word count

        for($i=0;$i<$word_count;$i++)

        {

            $trimmed .= $stringc[$i]." ";

        }

        

        if(substr($trimmed, strlen(trim($trimmed))-1, 1) == '.')

          return trim($trimmed).'..';

        else

          return trim($trimmed).'...';

    }

}



$text = "some  test          text goes in here, I'm not sure, but ok.";

echo trim_text($text,5);

MadCoder
16-Aug-2005 05:12


Here's a function that will trim a $string down to a certian number of words, and add a...   on the end of it.

(explansion of muz1's 1st 100 words code)



----------------------------------------------

function trim_text($text, $count){

$text = str_replace("  ", " ", $text);

$string = explode(" ", $text);

for ( $wordCounter = 0; $wordCounter <= $count;wordCounter++ ){ 

$trimed .= $string[$wordCounter];

if ( $wordCounter < $count ){ $trimed .= " "; }

else { $trimed .= "..."; }

}

$trimed = trim($trimed);

return $trimed;

}



Usage

------------------------------------------------

$string = "one two three four";

echo trim_text($string, 3);



returns:

one two three...

jtey at uoguelph dot ca
15-Aug-2005 12:21


In the previous note, the example will only extract from the string, words separated by exactly one space.  To properly extract words from all strings, use regular expressions.



Example (extracting the first 4 words):

<?php

$string = "One    two three       four  five six";

echo implode(" ", array_slice(preg_split("/\s+/", $string), 0, 4));

?>



The above $string would not have otherwise worked when using the explode() method below.

jtey at uoguelph dot ca
14-Aug-2005 03:59


In reply to muz1's post below:



You can also take advantage of using other built in PHP functions to get to your final result.  Consider the following:

<?php

$string = "One two three four five six seven eight nine ten.";

// the first n words to extract

$n = 3;

// extract the words

$words = explode(" ", $string);

// chop the words array down to the first n elements

$firstN = array_slice($words, 0, $n);

// glue the 3 elements back into a spaced sentence

$firstNAsAString = implode(" ", $firstN);

// display it

echo $firstNAsAString;

?>



Or to do it all in one line:

<?php

echo implode(" ", array_slice(explode(" ", $string), 0, $n));

?>

muz1 at muzcore dot com
12-Aug-2005 08:56


This function is awesome however I needed to display the first 100 words of a string. I am submitting this as a possible solution but also to get feedback as to whether it is the most efficient way of doing it.



<?

                                    $currString = explode(" ", $string);

for ($wordCounter=0; $wordCounter<100; $wordCounter++) { echo $currString[$wordCounter]." "; }

?>

16-Jan-2005 02:38


This function seems to view numbers as whitespace. I.e. a word consisting of numbers only won't be counted.

aix at lux dot ee
14-Nov-2004 10:53


One function.

<?php

if (!function_exists('word_count')) {

function word_count($str,$n = "0"){

    $m=strlen($str)/2;

    $a=1;

    while ($a<$m) {

        $str=str_replace("  "," ",$str);

        $a++;

        }

    $b = explode(" ", $str);

    $i = 0; 

    foreach ($b as $v) {

        $i++;

        }

    if ($n==1) return $b;

    else  return $i;



    }

}

$str="Tere Tartu linn";

$c  = word_count($str,1); // it return an array

$d  = word_count($str); // it return int - how many words was in text

print_r($c);

echo $d;

?>

aidan at php dot net
26-Jun-2004 11:02


This functionality is now implemented in the PEAR package PHP_Compat.



More information about using this function without upgrading your version of PHP can be found on the below link:



http://pear.php.net/package/PHP_Compat

Kirils Solovjovs
22-Feb-2004 05:06


Nothing of this worked for me. I think countwords() is very encoding dependent. This is the code for win1257. For other layots you just need to redefine the ranges of letters...



<?php

function countwords($text){

        $ls=0;//was it a whitespace?

        $cc33=0;//counter

        for($i=0;$i<strlen($text);$i++){

                $spstat=false; //is it a number or a letter?

                $ot=ord($text[$i]);

                if( (($ot>=48) && ($ot<=57)) ||  (($ot>=97) && ($ot<=122)) || (($ot>=65) && ($ot<=90)) || ($ot==170) ||

                (($ot>=192) && ($ot<=214)) || (($ot>=216) && ($ot<=246)) || (($ot>=248) && ($ot<=254))  )$spstat=true;

                if(($ls==0)&&($spstat)){

                        $ls=1;

                        $cc33++;

                }

                if(!$spstat)$ls=0;

        }

        return $cc33;

}



?>

Artimis
15-Oct-2003 10:32


Never use this function to count/separate alphanumeric words, it will just split them up words to words, numbers to numbers.  You could refer to another function "preg_split" when splitting alphanumeric words.  It works with Chinese characters as well.

andrea at 3site dot it
19-May-2003 12:55


if string doesn't contain the space " ", the explode method doesn't do anything, so i've wrote this and it seems works better ... i don't know about time and resource





<?php


function str_incounter($match,$string) {


$count_match = 0;


for($i=0;$i<strlen($string);$i++) {


if(strtolower(substr($string,$i,strlen($match)))==strtolower($match)) {


$count_match++;


}


}


return $count_match;


}


?>





example 





<?php


$string = "something:something!!something";


$count_some = str_incounter("something",$string);


// will return 3


?>

megat at megat dot co dot uk
19-Apr-2003 02:29


[Ed: You'd probably want to use regular expressions if this was the case --alindeman @ php.net]





Consider what will happen in some of the above suggestions when a person puts more than one space between words. That's why it's not sufficient just to explode the string.

olivier at ultragreen dot net
11-Apr-2003 02:10


I will not discuss the accuracy of this function but one of the source codes above does this.





<?php


function wrdcnt($haystack) {


 $cnt = explode(" ", $haystack);


 return count($cnt) - 1;


}


?>





That could be replace by 





<?php


function wrdcnt($haystack) {


 return substr_count($haystack,' ') + 1;


}


?>





I doubt this does need to be a function :)

philip at cornado dot com
07-Apr-2003 03:30


Some ask not just split on ' ', well, it's because simply exploding on a ' ' isn't fully accurate.  Words can be separated by tabs, newlines, double spaces, etc.  This is why people tend to seperate on all whitespace with regular expressions.

rcATinterfacesDOTfr
16-Jan-2003 03:58


Here is another way to count words :

$word_count = count(preg_split('/\W+/', $text, -1, PREG_SPLIT_NO_EMPTY));

brettNOSPAM at olwm dot NO_SPAM dot com
09-Nov-2002 08:06


This example may not be pretty, but It proves accurate:





<?php


//count words


$words_to_count = strip_tags($body);


$pattern = "/[^(\w|\d|\'|\"|\.|\!|\?|;|,|\\|\/|\-\-|:|\&|@)]+/";


$words_to_count = preg_replace ($pattern, " ", $words_to_count);


$words_to_count = trim($words_to_count);


$total_words = count(explode(" ",$words_to_count));


?>





Hope I didn't miss any punctuation. ;-)

gorgonzola at nospam dot org
31-Oct-2002 10:48


i tried to write a wordcounter and ended up with this:





<?php


//strip html-codes or entities


$text = strip_tags(strtr($text, array_flip(get_html_translation_table(HTML_ENTITIES))));


//count the words


$wordcount = preg_match_all("#(\w+)#", $text, $match_dummy );


?>

add a note