Need a few pointers

[newbtophp]

Hi, I'm in a bit of trouble.

 

If I have a list of words each on a separate line:

 

Example Words:

something
say
phpfreaks us all
something
help
php
php arithmetic
php functions

 

How can i place each word within a variable; (like with foreach)?

 

So at the end in this foreach if i echo that variable it should echo 5 words.

 

Also how would i retrieve the top recurring word from the words? (like for e.g. in this case its 'php' since that word is mentioned the most)

 

:-\

 

[wildteen88]

Have a look in to the file function.

[newbtophp]

Have a look in to the file function.

 

I've already tried but the values need to be a filename (unless i've missed anything).

 

In my case the words are submitted via $_POST['words']

[wildteen88]

Oh, I thought you where reading the lines from a file.

 

You'd want to use explode instead

$words = str_replace(array("\r\n", "\r"), "\n", $_POST['words']);

// convert the list of words into an array
$wordsArray = explode("\n", $words);

echo '<pre>'.print_r($wordsArray, true).'</pre>';

[teamatomic]

//$array=$_POST['words'];

// would give you the array I make below.

// the code following it gives up the word count

//file_put_contents("$file","$str")//will write your string to $file

//$str = file_get_contents("$file")//will put the contents into a string

//$array = file("$file")//will put the contents into $array-with each line as an element

 

$array=array(
something,
say,
'phpfreaks us all',
something,
help,
php,
'php arithmetic',
'php functions'
);
$str = implode(" ", $array);

$word_array_check=array();

foreach($array as $word)
{
$count = substr_count($str,$word);
$check= "\"$word\" occurred $count times";
array_push($word_array_check,$check);
echo "$check<br>";
}

echo '<p>';
$word_array = array_unique($word_array_check);

foreach($word_array as $key=>$value)
{
  echo "$value<br>";
}

 

 

HTH

Teamatomic

[Alex]

I could've sworn there was a built in function that would return an associative array containing each string and how many times it occurred. Maybe I'm just going crazy though. You could use something like this:

 

function count_word_occurrences($str, $case_sensitive = true)
{
$str = $case_sensitive ? $str : strtolower($str);
$occurrences = array();
foreach(str_word_count($str, 1) as $word)
{
	$word = $case_sensitive ? $word : strtolower($word);
	if(!in_array($word, array_keys($occurrences)))
	{
		$occurrences[$word] = preg_match_all("~\b$word\b~", $str, $matches);
	}
}
return $occurrences;
}

$text = <<<TEXT
The Law, as quoted, lays down a fair conduct of life, and one not easy to follow.
I have been fellow to a beggar again and again under circumstances which prevented 
either of us finding out whether the other was worthy. I have still to be brother 
to a Prince, though I once came near to kinship with what might have been a veritable 
King and was promised the reversion of a Kingdom army, law-courts, revenue and policy 
all complete. But, to-day, I greatly fear that my King is dead, and if I want a crown 
I must go and hunt it for myself.
TEXT;

print_r(count_word_occurrences($text));

 

Output:

 

Array
(
    [The] => 1
    [Law] => 1
    [as] => 1
    [quoted] => 1
    [lays] => 1
    [down] => 1
    [a] => 6
    [fair] => 1
    [conduct] => 1
    [of] => 3
    [life] => 1
    [and] => 6
    [one] => 1
    [not] => 1
    [easy] => 1
    [to] => 6
    [follow] => 1
    [i] => 6
    [have] => 3
    [been] => 2
    [fellow] => 1
    [beggar] => 1
    [again] => 2
    [under] => 1
    [circumstances] => 1
    [which] => 1
    [prevented] => 1
    [either] => 1
    [us] => 1
    [finding] => 1
    [out] => 1
    [whether] => 1
    [the] => 2
    [other] => 1
    [was] => 2
    [worthy] => 1
    [still] => 1
    [be] => 1
    [brother] => 1
    [Prince] => 1
    [though] => 1
    [once] => 1
    [came] => 1
    [near] => 1
    [kinship] => 1
    [with] => 1
    [what] => 1
    [might] => 1
    [veritable] => 1
    [King] => 2
    [promised] => 1
    [reversion] => 1
    [Kingdom] => 1
    [army] => 1
    [law-courts] => 1
    [revenue] => 1
    [policy] => 1
    [all] => 1
    [complete] => 1
    [but] => 1
    [to-day] => 1
    [greatly] => 1
    [fear] => 1
    [that] => 1
    [my] => 1
    [is] => 1
    [dead] => 1
    [if] => 1
    [want] => 1
    [crown] => 1
    [must] => 1
    [go] => 1
    [hunt] => 1
    [it] => 1
    [for] => 1
    [myself] => 1
)

 

If you're wanting to only get the top occurring words in the string you can use arsort and only take the first few elements of the array.

 

@teamatomic: There's a few problems with that method. Most notably that substr_count won't count only full words. So if you were searching for the word "a", it would also count all the a's that are parts of other words. Even if you padded spaces onto the string you're searching for that doesn't take into account other word barriers (like ",", ";", etc). However, the \b modifier in regex does.

[Alex]

Or alternatively..

 

$words = str_word_count($text, 1);
print_r(array_count_values($words));

 

And to limit to only the top 3:

 

$words = array_count_values(str_word_count($text, 1));
arsort($words);
$words = array_slice($words, 0, 3);