« Posts under Work

My CV and some other Updates

Just a quick post to say hello.

I’ve updated my CV and tidied it up a little.
So any feedback on it would be useful! (Drop Me a Line rather than a Comment please)
Link in the SideBar on the right!

I should really write on my blog some more, but not a lot to write about of late.
JetPack Extras is doing OK, but JetPack core decided to implement the main feature I added, that being Pinterest but, Extras has evolved, it still of course has the ability to control button placement and has the extra Twitter Via/Related options, and the ability to share the WP.me shortened URL, makes a nice Twitter Card, embedding a short Preview of the Post with the Tweet, (check this Meta Tweet for an example) works with any wp.me url you share on Twitter by the way!

Most useful, mainly updated to keep up with JetPack core and use the new shiny hooks they have.
You can check it out on extend. Feedback/request are always Welcome!

Made a couple of tweaks and updates to the Blog in the background, added a better 404 page and updated the .htaccess rules, I have all subdomains pointing to the here, (unless their is a separate to show), so its a good idea to redirect non sites to here, rather than duplicating the content.
Also tweaked the Root .htaccess to have a Error Document (404) since it was just standard Apache Error, bit nasty, but now better, LOLCATS TO THE RESCUE! 😀

Heres a extract of the rules, if you find it useful/handy:

RewriteEngine on

RewriteCond %{HTTP_HOST} !^barrycarlyon.co.uk  
RewriteRule ^(.*)$ http://barrycarlyon.co.uk/%{REQUEST_URI} [R=301,L]
ErrorDocument 404 http://barrycarlyon.co.uk/wordpress/404-2/
ErrorDocument 403 http://barrycarlyon.co.uk/wordpress/404-2/

Next thing to do, is probably update and replace the MineCraft Server Site probably using Twitter Bootstrap as a new basis. It’s now a whitelisted server for Armchair Heroes or on Facebook, but of late I’ve been getting my MineCraft fix playing over at PhantomCraft, like a lot of open Servers I tend to get robbed/cleaned out and griefed a lot, but then again I am a little close to spawn. It’s a fun server to play on and the regulars are quite nice.

Hopefully moving further away when I get a chance to play some more, thinks are pretty busy at the moment between Magento fun at Day Job and Freshers Fun at the Union.

Then I really should get the Portfolio page on the Blog fixed up to showcase what I’ve done before, since I am a Freelance Web Developer (you got any work going or want a quote for? (Drop me a line)

So a couple more weeks and we’ll be back to the usual grind.
Hopefully I will have more to write about soon!

Facebook Credits, the Order ID, and dealing with large integers in 32 bit PHP

So, currently at work I’ve been rewriting all our Payment Gateways for Your Members. Making them more better Class based an abstracting out the common functions to a base class, to save memory footprint and code etc.

And came across a problem when rewriting the Facebook Credits handler.

The order ID that gets passed around, as well as the Application ID is too long for a 32 Bit PHP installation to handle when treating it as an Integer.

So a number like “239724439419867” was ending up as “1.7592246582797E+14” when stored/processed, or converted to a string.
Now the obvious way to process this is to convert it to a string another way, which is somewhat difficult since before you even start it is in a format that you can’t handle.

The saving grace here is the fact that data is passed to you as a JSON packet (aside from the fact the Order ID is in the $_POST variable and thus a string, so I could of used it from there).

This led to me to look at the JSON packet, which starts as a string but when decoded the large integers are still a larger integer than can be processed (whether you json_decode to a Array or to an Object, the problem persists). So I thought about how to process the raw JSON packet and make sure the order ID, and other large integers are treated as strings.

If you look at a Raw JSON packet you can easily spot whats a string, integer, object or array.

Take this example Facebook Credits JSON Packet, (wrapped for readability):

{"order_id":239724439419867,"buyer":197803678,"app":148748711865470,
"receiver":197803678,"amount":5,"update_time":1320075413,
"time_placed":1320075408,"data":"","items":[{"item_id":"<item_id>","title":"Post",
"description":"A New Purchasable Post",
"image_url":"<some url>","product_url":"<some other url>",
"price":5,"data":"<data>"}],"status":"<status>"}
  • order_id is an integer as it has no ” around it
  • status is a string as its surrounded by “
  • Items is an array as its surrounded by [, in this case containing a single object, entries/items are comma separated.

So after thinking about this I decided the best way to sort this out was to convert the integers in the raw JSON packet to strings before decoding.

In pseudo code.

Lop off the { and } from the start and end
Explode around ,
array walk each item
split on :
check if there are no " in the second bit
if none wrap in "
glue back together

Something alone the lines of:

function largeint($rawjson) {
  $rawjson = substr($rawjson, 1, -1);
  $rawjson = explode(',' , $rawjson);
  array_walk($rawjson, 'strfun');
  $rawjson = implode(',', $rawjson);
  $rawjson = '{', . $rawjson . '}';
  $json = json_decode($rawjson);
  return $json;
}

function strfun(&$entry, $key) {
  $data = explode(':', $entry);
  if (FALSE === strpos($data[1], '"')) {
    $data[1] = '"' . $data[1] . '"';
    $entry = implode(':', $data);
  }
}

I’m not sure in terms of memory footprint if its cheaper to do a substr of $data[1] and check if its a ” or not.
I suppose it could test the string length of the detected integer to see if its invalid/unable to process, but then you would be performing a string function on an integer and again the problem would arise.

But as a block of code is does the job, its obviously not ideal for all situations, since in this case I want the integer as a string, I’m not using it for math, but if I did, we probably need to do some bizarre unpack-ing or something.

Any opinions or improvements give us a comment below 😀

Thoughts on working on a different computer when your normal one is not available

So currently my Mac Book Pro is in with Clockwork Leeds getting its Hard Drive replaced as it failed/has failed/is failing.

So currently at work, I am sans my normal Development machine and find myself on the Mac Mini server that sits and does our File Server and System Monitoring.

Its quite odd having to get used to a system that for starts is the Server build of Mac OSX as well as having none of my personal preferences.
Its bringing to light how slightly useless some of the things I have setup on my Personal machine are that I thought were useful.

For example, I’m stuck with a single screen, so no dual screen lovely-ness. So I’m very restricted to the screen real estate I have available.
So not having new Terminal Windows opening to fill half the screen like on my mac is quite useful. Whereas before having the secondary screen with two terminal windows full height, half width was the norm, as well as using a little app I’ve forgotten the name of which allows me to press ctrl + ~ to get a window appear from the top, is a bit of a pain, but resolved now with having three default sized terminals and chrome open.

So far its seemed to work just about fine like that.

I also find myself using Spaces/Expose a lot less, I’m doing a lot less jumping between spaces all the time. But I also don’t have the “App only opens in space x” setup so I get a lot less forced jumping.

Finally tho I miss my music collection, so I’ve switched to plugging my headphones into my iPad and listening to Spotify. Tho I could just plug in my external drive thats in the office and have most of my older music on hand….

I’m also on a different desk which is nice.
I’m not by the door and have the window behind me and I’m in the corner.
I quite like it over here. Seems a bit more separated from everyone else. Which is a good thing and a bad thing in its own right.

I love working with the team, *can’t think of right words to express feelings will just skip to post summary*.

Summary

My Mac is in the repair shop getting fixed.
The hard drive was failing and is pending spare parts from Apple.
Clockwork are amazing for Apple Repairs.
I’m in a corner at work
I have no home machine to really play or work on. (I made my home windoze laptop blue screen just by looking at it….)

I watched last nights Torchwood on my iPad using the BBC iPlayer Application and it worked perfectly.

Finally my local dev environment is shafted as that was my Mac Book Pro. Luckily everything is in offsite repo’s and I have a Server in a Server Room down the corridor to talk to 😀

-End-

My Leeds Hack 2 Project

Thought I would write a blog post about what I built in 24 hours for Leeds Hack 2 last weekend. I kinda got distracted last week and not had the time to write this post.

You can look at all the projects that people finished over on the Leeds Hack Website.

So what did I write

I decided to write something that I had been pondering about for a while. It’s called Spotify Roulette and revolves around the idea of crowd sourcing a new artist or area of music to listen to.

Originally it started off with me posting on Twitter quite a while back asking for something new to listen to, to which @Stanton responded with Hybrid. (I can’t find the original tweet on twitter but it’s here on Spotwitfy) Which are actually quite nice to listen to and I have a nice Spotify playlist to listen to.

I was wondering if there was a way to automate this.

So thru the combination of the Twitter and Spotify Meta Data API’s, means I can post out to Twitter, await a response, parse out the Artist and then pop open Spotify with a random track by that Artist and if the requester wants a playlist can be generated, and thru a limitation in Spotify, drag and dropped into Spotify to listen further to the Artist.

In short

In short its pretty straight forward.
Just a handful of calls to a couple of different API end points to get an Artist ID from the name, then their albums and the tracks on these albums. Chuck in a little GeoIP to hopefully check the tracks are available in the requesters’ region.
Grab the first track, pop open Spotify and grab another 10 tracks to make a playlist.

Finally I used Nerf Guns to help demo! Russian Roulette stylee.

Responses

People on Twitter can either response with text, an artist which we parse out the @user and the hashtag word. Then look that up on Spotify.
Or a response can be either an HTTP open Spotify link or a Spotify protocol link. Either to the artist page, track or album.
if it’s the Artist, I can parse out the artist ID and look for albums and tracks.
Currently not playlist urls, but that’s easy to implement.

Technology

So we used,

  • Twitter API, using @Abraham Twitter oAuth Library
  • Spotify Meta Data API
  • MaxMind GeoIP
  • jQuery and jQuery UI
  • Rick Astley
  • Nerf Guns

Give it a Go

Give it a go and offer me some feedback.
Watch out for Rick Astley tho. He likes to crop up every now and again….

It’s still a little rough round the edges in terms of theme/layout.

Spotify Roulette (http://spotifyroulette.com/)

Future

Hopefully if people like it and use it I can expand further.

Perhaps use Facebook to share playlists, or another way to crowd source a new artist. And if Spotify release a HTTP API for generating and saving playlists then incorporate that too.

WordPress and the Admin Bar

So WordPress 3.1 introduced the new admin bar. Essentially its another way to navigate around your WordPress Blog’s inner workings, the bar itself can be enabled and disabled both inside and outside the admin system, for I like to have it enabled on both the admin and non admin sides of WordPress.

The WordPress.com Stats Plugin makes quite good use, on the non admin side, by showing a graph of site visits over the last 48hours:

This led me to think about how to put my own links on, so I turned to adding the navigation from YourMembers to the admin bar (but in this case on Both Sides, admin and non admin).

Won’t show you a admin side one as it looks exactly the same 😛

Actually writing the code to make the YourMembers navigation be added to the nav bar, is actually relatively simple, since the admin bar itself has been written rather nicely in its own class.

Opening up the admin-bar.php file in wp-includes is a bit messy, but looking at the first function (_wp_admin_bar_init) shows the class function add_menus() and further down add_menu, so off I went to find them…. A quick grep later finds that is class-wp-admin-bar.php and from there just a matter of reading what add_menu wants….

<?php

	function add_menu( $args = array() ) {
		$defaults = array(
			'title' => false,
			'href' => false,
			'parent' => false, // false for a root menu, pass the ID value for a submenu of that menu.
			'id' => false, // defaults to a sanitized title value.
			'meta' => false // array of any of the following options: array( 'html' => '', 'class' => '', 'onclick' => '', target => '', title => '' );
		);

I went thru and checked to see where the other “normal admin” links were being added, the function add_menus in the same class file shows that. The key thing to be aware of is the use of priority on those do_actions (line 182 thru 194) in order to order the links, so needing/wanting the YourMembers links to be at the end I went for a sensible priority of 90 since the last normal admin was set to 80.

Then it was just a matter of setting up my own add_action on the same hook (‘admin_bar_menu’) and then globalising the pre existing class method and adding my menus to it.

Since it is a drop down menu, needed to add the top menu as a parent and then adding the child nodes.

Its quite a clever class since the class function of add_menu can handle both parents and childs at the same time with little fuss, as long as you pass it the right ID variables.

Heres a copy of my function:

<?php

function ym_admin_bar() {
	global $wp_admin_bar;

	if (ym_admin_user_has_access(TRUE)) {
		$wp_admin_bar->add_menu(
			array(
				'id'		=> 'yourmembers_adb',
				'title'		=> 'Your Members',
				'href'		=> YM_ADMIN_INDEX_URL
			)
		);
		global $ym_nav;
		foreach ($ym_nav as $page => $subpages) {
			if (is_array($subpages)) {
				$first = array_shift($subpages);
			} else {
				$first = '';
			}
			$id = 'ym_adb_' . strtolower($first);
			$url = strtolower($first);
			if (substr($first, 0, 5) == 'other') {
				$url = 'ym-other&action=' . substr($first, 6);
			}
			$wp_admin_bar->add_menu(
				array(
					'parent'	=> 'yourmembers_adb',
					'id'		=> $id,
					'title'		=> $page,
					'href'		=> YM_ADMIN_INDEX_URL . '&ym_page=' . $url
				)
			);
		}
	}
	return;
}
add_action('admin_bar_menu', 'ym_admin_bar', 90);
?>

There are a couple of funky bits in there due to the method we currently generate the nav items.
The essential bit to consider is the array passed to $wp_admin_bar->add_menu at the start (the parent element note, no parent key) and the $wp_admin_bar->add_menu at the end, specifying a child element.

The function ym_admin_user_has_access just checks to see if the user is of the correct level or not.

You might want to throw in a switch on is_admin() to return different navigational items depending on if you are in the admin or not.

I’m sure there is a lot more you can do with the class function but when I installed WordPress 3.1 and saw the admin bar and WordPress.com Site stats this was the first thing I thought of and implemented, tho why its taken me this long to blog about it I don’t know…..