need advice on re-associating an array value with original key - php

Question

I'm trying to write a program that will grab a bunch of images from a webpage and figure out which of the images is the largest.
So far I've taken the images, put them into an array, used the getimagesize() function to determine the heights. I then placed all of the heights into another array and sorted them in reverse order to get the largest one. So far so good.
My problem now is that I must find a way to re-associate the largest image with its initial image link. I've thought of potentially running the initial code to fetch the images again from the web page. Then comparing the 1st value in the array I've used to determine the biggest image with the images a second time, but this seems like a waste of bandwidth and I get a feeling there is an easier way to re-associate the height value with its initial image. Am I right?
<?php
$url = 'http://lockerz.com/s/104049300';
// Fetch page
$string = FetchPage($url);
// Regex that extracts the images (full tag)
$image_regex_src_url = '/<img[^>]*'.
'src=[\"|\'](.*)[\"|\']/Ui';
preg_match_all($image_regex_src_url, $string, $out, PREG_PATTERN_ORDER);
$img_tag_array = $out[0];
echo "<pre>"; print_r($img_tag_array); echo "</pre>";
// Regex for SRC Value
$image_regex_src_url = '/<img[^>]*'.
'src=[\"|\'](.*)[\"|\']/Ui';
preg_match_all($image_regex_src_url, $string, $out, PREG_PATTERN_ORDER);
$images_url_array = $out[1];
$image_heights_array = array();
foreach ($images_url_array as $imagelink)
{
if (substr($imagelink,0,7)=="http://")
{
$getheight = getimagesize($imagelink);
array_push($image_heights_array,"$getheight[1]");
}
}
rsort($image_heights_array);
echo "<pre>"; print_r($image_heights_array); echo "</pre>";
// Fetch Page Function
function FetchPage($path)
{
$file = fopen($path, "r");
if (!$file)
{
exit("The was a connection error!");
}
$data = '';
while (!feof($file))
{
// Extract the data from the file / url
$data .= fgets($file, 1024);
}
return $data;
}
?>

Answer 1

First off:
// Regex that extracts the images (full tag)
$image_regex_src_url = '/<img[^>]*'.
'src=[\"|\'](.*)[\"|\']/Ui';
preg_match_all($image_regex_src_url, $string, $out, PREG_PATTERN_ORDER);
$img_tag_array = $out[0];
echo "<pre>"; print_r($img_tag_array); echo "</pre>";
// Regex for SRC Value
$image_regex_src_url = '/<img[^>]*'.
'src=[\"|\'](.*)[\"|\']/Ui';
preg_match_all($image_regex_src_url, $string, $out, PREG_PATTERN_ORDER);
HTML parsing regex ewww. Let's simplify this... with an HTML parser.
<?php
// EDIT: Use a custom function to do the
// reverse of SORT_NUMERIC with asort
function height_compare($a, $b)
{
if ($a == $b) {
return 0;
}
return ($a > $b) ? -1 : 1;
}
$url = 'http://lockerz.com/s/104049300';
$doc = new DOMDocument();
// Okay this is HTML is kind of screwy
// So we're going to supress errors
#$doc->loadHTMLFile($url);
// Get all images
$images_list = $doc->getElementsByTagName('img');
$images = array();
foreach($images_list as $image) {
// Get the src attribute
$image_source = $image->getAttribute('src');
$image_size_info = getimagesize($image_source);
$images[$image_source] = $image_size_info[1];
}
// Do a numeric sort on the height
uasort($images, "height_compare");
print_r($images);
?>
Much shorter and more readable. The result:
$ php test.php
Array
(
[http://c0013784.cdn1.cloudfiles.rackspacecloud.com/x2_633aa94] => 328
[http://ad.doubleclick.net/ad/buz.plixi/photos;pos=300a;celeb=;kw=;tile=2;sz=300x250,300x600;ord=123456789?] => 250
[http://static.lockerz.com/pegasus/images/video_thumb.jpg?1.0.0] => 207
[http://static.lockerz.com/pegasus/images/plixi-banner.png?1.0.0] => 107
[http://ad.doubleclick.net/ad/buz.plixi/photos;pos=728a;celeb=;kw=;tile=1;sz=728x90;ord=123456789?] => 90
[http://c0013787.cdn1.cloudfiles.rackspacecloud.com/x2_636f30c] => 79
[http://c0013787.cdn1.cloudfiles.rackspacecloud.com/x2_637676e] => 79
[http://c0013787.cdn1.cloudfiles.rackspacecloud.com/x2_63735a0] => 79
[http://c0013787.cdn1.cloudfiles.rackspacecloud.com/x2_636e73c] => 79
[http://c0013787.cdn1.cloudfiles.rackspacecloud.com/x2_63795d0] => 79
[http://c0013787.cdn1.cloudfiles.rackspacecloud.com/x2_636a2c7] => 79
[http://c0013787.cdn1.cloudfiles.rackspacecloud.com/x2_636bf79] => 79
[http://c0013787.cdn1.cloudfiles.rackspacecloud.com/x2_636ca08] => 79
[http://c0013787.cdn1.cloudfiles.rackspacecloud.com/x2_636e419] => 79
[http://c0013787.cdn1.cloudfiles.rackspacecloud.com/x2_636deca] => 79
[http://c0013787.cdn1.cloudfiles.rackspacecloud.com/x2_6384277] => 79

Answer 2

Keep sizes and URLs in one array and use usort to sort multidimensional array.
$images_url_array = $out[1];
$images = array();
foreach ($images_url_array as $imagelink)
{
if (substr($imagelink, 0, 7)=="http://")
{
$getheight = getimagesize($imagelink);
$images[] = array('height' => $getheight[1], 'url' => $imagelink);
}
}
usort($images, function ($a, $b)
{
if ($a['height'] > $b['height']) return 1;
elseif ($a['height'] < $b['height']) return -1;
else return 0;
});

Answer 3

I don't know PHP so I can't help you with the code, but maybe you could try this idea. Have the first array store an ImageObject. This class to have three properties, imagesize, id, and src.
You can go through the first array, find the imagesize and then sort it by biggest size. Use the id to get individual objects.

Answer 4

if you use arsort() the array keys will be preserved, then you can reset() the array pointer to start of the array, and get the first key().
But you will also need to preserve the same keys across the images_url_array and image_height_array
foreach ($images_url_array as $key => $imagelink)
{
if (substr($imagelink,0,7)=="http://")
{
$getheight = getimagesize($imagelink);
$image_heights_array[$key] = $getheight[1];
}
}
//...
rsort($image_heights_array);
reset($image_heights_array);
$largest_image_key = key($image_heights_array);
$largest_image_url = $images_url_array[$key];