array_unique($array);

[utahcon]

So let me make sure I am clear here -

when I do this

$some_arr = array_unique($array);

$some_arr should be an array of the keys with unique values (I assume the first of each value?)

Now if I do

$some_arr = array_count_values($array);

is there anyway to echo out the Key?

[paulnaj]

Originally posted by utahcon
I assume the first of each value?

Just in case it's important, that's not true ... here's a note from the manual -Note that keys are preserved. array_unique() sorts the values treated as string at first, then will keep the first key encountered for every value, and ignore all following keys. It does not mean that the key of the first related value from the unsorted array will be kept.

is there anyway to echo out the Key?

Yes, see array_keys() .

Paul 🙂

[utahcon]

Ok - maybe this will help here is my code::

$site = $_SERVER['HTTP_REFERER'];

//This opens the page we are scraping for content
$handle = fopen($site, 'r') or die("Could not open $site");

//loop through page until we have the whole thing
while(!feof($handle)){
     $siteCode .= fread($handle, 1000000);
}
//trim the code of whitespace at the beginning and end of file.
$siteCode = trim($siteCode);

//close the page
fclose($handle);

//strip out the HTML code
$siteCode = strip_tags($siteCode);

//remove some periods at the end of words
$siteCode = str_replace(". ", ".", $siteCode);

//build an array out of the final string
$words = explode(" ",$siteCode);
echo count($words) ." words<br>";

So what I need to get is a complete list of the unique words on the page, and how many times they appear.

I need these in a way that I can sort them and display them through a loop.

[utahcon]

well here is the work around I came up with:

		for($i=0;$i<count($keysKey);$i++){
			$key = $keysKey[$i];
			$word = $keys[$key];
			$count = $counts[$word];
			$final[$i]['word'] = $word;
			$final[$i]['count'] = $count;
		}

[kburger]

Try this:

$site = $_SERVER['HTTP_REFERER'];
$string = strip_tags(file_get_contents($site));
$stringarray = preg_split('/\s+/', $string);
$wordarray = array_count_values($stringarray);
arsort($wordarray);
foreach ($wordarray as $word => $times) {
	echo "$word appeared $times times<br>\n";
}

That regexp is actually / \s+/
The board removed the backslash.

[Weedpacket]

Funnily enough, today I found this wee bit of code that has been sitting on my drive for no earthly reason that I can still discern:

<?php

function sort_words($a,$b)
{
	// Sorting in descending order by frequency (0th element);
	// tiebreaks decided lexicographically (1st element)
	if($t = $b[0]-$a[0]) return $t;
	return strcmp($b[1],$a[1]);
}

// Read in a file from stdin, convert it to lower case, break it into
// "words" of consecutive letters, and count the words. Don't include
// the two null words that "appear" at the very ends of the file.
$words = array_count_values(
	preg_split('/[^a-z]+/',
		strtolower(file_get_contents('php://stdin')),
		-1, PREG_SPLIT_NO_EMPTY));

// Get the key and value for each element of the $words array,
// make an array of the pair, and store it as an element.
// Annoyingly necessary, since the inbuilt sort functions neither
// accept key-value pairs directly, nor - using quicksort as they do -
// are they stable. (If they were stable, one could have sorted the
// list as per spec in place by writing
// arsort($words);
// krsort($words);
// instead of the custom comparison function and the following two lines.)
$words = array_map(null, array_values($words), array_keys($words));
usort($words,'sort_words');

foreach($words as $pair)
{
	printf("%7d %s\n", $pair[0], $pair[1]);
}
?>