// create the image resource from a file
$i = imagecreatefromjpeg('otter.jpg');

// define the sharpen matrix
$sharpen = array(
	array(0.0, -1.0, 0.0),
	array(-1.0, 5.0, -1.0),
	array(0.0, -1.0, 0.0)
);

// calculate the sharpen divisor
$divisor = array_sum(array_map('array_sum', $sharpen));

// apply the matrix
imageconvolution($i, $sharpen, $divisor, 0);

// output the image
header('Content-Type: image/jpeg');
imagejpeg($i);

Otter, before and after sharpening

Image courtesy Kevin Law

Since the news of an API last week, I’ve been working on a little class wrapper – partly to hone my OOP skills and partly to support some great work by Bastian – and my class is now ready. I’ve named it ZooPHP.

$zoo = new ZooPHP($key, $secret);
$zoo->getPopularItems();

To start using ZooPHP, you’ll need a Zootool API key. The script is released under a BSD license and is documented, but should be fairly self-explanatory:

$zoo->setAuth('username', 'password');
$zoo->setFormat('array');
$zoo->getUser('bastian');
$zoo->getUserFriends('bastian');
$zoo->getUserFollowers('bastian');
$zoo->getUserItems('bastian');
$zoo->getLatestItems();
$zoo->getItem('iw6og3');
$zoo->getPopularItems();
$zoo->addItem('adamhopkinson.co.uk', 'adamhopkinson.co.uk', 'zoophp, php, wrapper');

You’ll only need to call setAuth if you want to add items or return private items with getItems.

Click here to download your copy of ZooPHP »

With all the reports over the approval process for apps being submitted to App Store, can Orbicule guarantee that their app will get through with these features? Apple previously refused to approve Podcaster on the grounds that it duplicated the functionality of the podcast section of iTunes.

To use, put your file path/url in the shortcode [external].

Click here to download External Files

The plugin will be available very soon.

Get a PCA account

First, you’ll need to sign up to Postcode Anywhere and add some credits. A beginner pack of 300 credits will set you back £25+VAT and will get you 300 address lookups (if you only want the basic details returned for each address).

Use PHP to build the webservice client

An AJAX lookup works by passing some data to a script on the webserver, which performs some translation (in our case, connects to the webservice) and returns the result to the page where javascript makes the updates. Create a new php file on your webserver call getpostcode.php

Set up the defaults for the lookup:

<?php

$account_code = 'PCA_ACCOUNT';
$license_code = 'PCA_LICENCE';

$postcode = '';
$building = '';

?>

Then, get the postcode and building number or name from the query string:

if ( array_key_exists('postcode', $_QUERY) ) {
	$postcode = $_QUERY['postcode'];
}

if ( array_key_exists('building', $_QUERY) ) {
	$building = $_QUERY['building'];
}

Then we have to build the REST url that will ask the webservice for the address:

$url = 'http://services.postcodeanywhere.co.uk/csv.aspx?';
$url .= 'account_code=' . $account_code;
$url .= '&license_code=' . $license_code;
$url .= '&action=fetch';
$url .= '&postcode=' . $postcode;
$url .= '&building=' . $building;
$url .= '&style=simple';

Use the php file function to get the data from the webservice into a variable called $data

$data = file($url);

We’ve asked the webservice to return the data in csv format, so we’re expecting two lines. The first will contain the field names and the second will contain the values. First, we want to return an error if the webservice call failed:

if(!$data) {
	die('{"status":"error"}');
}

The status: error message is in JSON format so that it can be understood by the javascript which will handle the address lookup.

The php file function returns an array, with each line in a different array element – so the first line will be in $data[0] and the second in $data[1] (arrays always start at 0 not 1!).

Next we want to trim any whitespace such as spaces or newline characters from the end of each line:

$data[0] = trim($data[0]);
$data[1] = trim($data[1]);

then we trim all extra speech marks from the start and end to help us split the lines up

$data[0] = trim($data[0], '"');
$data[1] = trim($data[1], '"');

Now we can split the lines into two arrays, one for field names and one for values

$field_names = explode('","', $data[0]);
$field_values = explode('","', $data[1]);

and throw up an error if the number of fields isn’t the same as the number of values

if(count($field_names) != count($field_values)) {
	die('{"status":"error"}');
}

Empty the $data array and loop through the field names and values, building up an array of name => value pairs

$data = array();

for($counter = 0; $counter < count($field_names); $counter++) {
	$data[$field_names[$counter]] = $field_values[$counter];
}

PostcodeAnywhere returns a field called error_number if there’s a problem in the request. If this field exists, return an error to the form:

if(array_key_exists('error_number', $data)) {
	die('{"status":"error"}');
}

Finally, output the data in JSON format:

$output = '';
foreach($data AS $key=>$value) {
	$output .= '"' . $key . '":"' . $value . '",';
}
//$output = trim($output, ',');
$output = '{"status":"ok",' . $output . '}';
echo $output;

In part 2, we’ll look at how to use the Yahoo Connection Manager to put this data on your form. Check back in a few days…

I use Newsgator online and therefore have a paid FeedDemon install on my Dell laptop, as these synchronise. This means I can read stuff in FeedDemon when i’m at my desk and Newsgator when i’m not without having to read the same items twice over.

This presents an obvious choice for a Mac feedreader – NetNewsWire (Lite). This also synchronises with Newsgator and so will save me a lot of time and effort during the daily catch-up.

I feel though that as Newsgator brought in each component, they have done the bare minimum to integrate it into the brand. Neither of the application-based readers under the Newsgator umbrella do any more than synch with Newsgator. Surely it would make more sense to have them support the same features and synchronise settings (ie flagged items, watch lists) with Newsgator instead of simply synchronising subscriptions. One step further would be to bring the same look and feel to both feedreaders – if iTunes can look the same on both Windows and Mac, then FeedDemon and NetNewsWire can.

RSS is a technology that is begging to be jumped on by the uneducated masses – as shown by the inclusion of feed readers in IE7 – but Newsgator don’t seem to have spotted this potential yet.

One thing that occurred to me is that even when (if?) Boot Camp is to be included in the upcoming OSX Leopard, users still need to run the program within OSX to install Windows on a Mac.

If I were Steve Jobs, I’d plan to integrate Boot Camp into the wizard that OSX launches (to configure the machine) when a new Mac is first used by the end user. If the options available were…

1. Customise OSX
2. Customise OSX and install Windows
3. Just install Windows, and set it as the startup

…Apple would be able to catch the market of users who will never want to leave Windows but do want a reliable and stylish machine. This wouldn’t necessarily speed up the downfall of Windows, but would bring Apple some of the market share currently owned by Dell, Lenovo and other large scale hardware manufacturers.