The messages National Rail send allow you to plot trains to stations, or even along route, but the message only contains station IDs or train IDs – which is kinda boring. I like to visualise the data, on a map for example.

To get the station name and location for a particular train message one has to access an entirely different database, from another provider. Along comes ATOC with their CIF files, which look very scary compared to ActiveMQ.

To cut a long story as short as I can, the CIF files contain a lot of information (400MB+ files), using string length and new lines to split up the data. You can sign up to download the files on the ATOC website, and the specification for the files is available here.

Why the blog post? Well, I needed a way to parse these files to populate a Mongo database and wanted to promote the PHP CIF parser I’ve started work on: https://github.com/rb-cohen/php-cif-parser

It’s been a long time coming (the idea started in 2010), but my business partner and I have finally got around to organising a local PHP user group in Kent (UK).

I attend PHP London frequently, but am hoping to meet local developers and share ideas outside of that. Not able to get enough of PHP user groups, I plan to attend both on a monthly basis!

For more information on the meetup, like when the next event is scheduled, and to RSVP – please head over to the meetup page.

How does this work in practice?

In a controller you might want to do something like this:

1
2
3
4
5
6
7
8
9

<?php
namespace Application\Controller;</code>
 
use Zend\Mvc\Controller\AbstractActionController;
use Zend\View\Model\ViewModel;
 
class IndexController extends AbstractActionController {
    public function indexAction() {
        $data = $this->getRequest()->getContent();

But if you’re always expecting a certain format, say JSON, a controller plugin could give you back the array or object rather than a lot of json_decode calls in your controllers.

The plugin:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21

<?php
 
namespace My\Mvc\Controller\Plugin;
 
use Zend\Mvc\Controller\Plugin\Params as ZendParams;
 
class Params extends ZendParams {
 
    public function fromJson() {
        $body = $this->getController()->getRequest()->getContent();
        if (!empty($body)) {
            $json = json_decode($body, true);
            if (!empty($json)) {
                return $json;
            }
        }
 
        return false;
    }
 
}

And then its as simple as:

1
2

public function indexAction() {
    $data = $this->params()->fromJson();

Help! Sometimes getContent() doesn’t return my data.

The method uses file_get_contents(‘php://input’) to get the raw data from the HTTP request. Unfortuantely, the php://input stream can only be read once per process. I did have a plug-in that called file_get_contents(‘php://input’) too, and if $request->getContent() gets called after this, it will be empty!

The next version will have more information available (total points, most recent workout, etc). As soon as its possible (via some sort of official Fitocracy API) I’ll be removing the need to enter your own account details too!

The sad thing is, this is a massive problem for VM across the country. A quick look through posts on their forum, a large percentage of them are about poor service at peak times.

I’ve been monitoring the situation for the last week, this is what I’ve found so far. 50% pack loss on most evenings and for the majority of Saturday (today).

Virgin Media MEx Service status

If it continues, I’ll have to find an ADSL provider with a decent contention ratio.. does anyone know of one?

I was hoping to get my Fitocracy stats up on the blog as a widget, but turns out its (next to) impossible at the moment. I get that an API isn’t always first on the agenda or something that users should expect to have – but it would be nice!

As soon as I can get a feed of my profile data I’ll be adding the widget and then releasing the plugin to the world. I’ve put in a request on the Fito forums, lets see if anyone has any ideas.

There are some other unofficial Fitocracy APIs out there, but none seem to work at the moment!

Setting up the variables

Getting started is simple, we just have to tell phing to find itself in the phar file. On unix, you can run the following commands, or download the script I’ve made to automate it all:

$ export PHP_COMMAND=/usr/bin/php
$ export PHING_HOME=phar:///path/to/phing.phar
$ export PHP_CLASSPATH=${PHING_HOME}/classes
$ alias phing="${PHP_COMMAND} ${PHING_HOME}"

Now you can use phing as you would had you installed it from PEAR, like so:

$ phing -h

Automating this process

Nobody wants to type that out everytime, so just use my setup.sh script, stick it in the same directory as phing.phar and just run:

$ . ./setup.sh

Note the double dot, this is important (or it won’t export the variables to the current bash process).

The file setup.sh consists of:

#!/bin/bash
 
SOURCE="${BASH_SOURCE[0]}"
while [ -h "$SOURCE" ] ; do SOURCE="$(readlink "$SOURCE")"; done
DIR="$( cd -P "$( dirname "$SOURCE" )" && pwd )"
 
export PHP_COMMAND="/usr/bin/php"
export PHING_HOME="phar://${DIR}/phing.phar"
export PHP_CLASSPATH="${PHING_HOME}/classes"
 
alias phing="${PHP_COMMAND} ${PHING_HOME}"

But just download it to wherever phing.phar is, rename it to setup.sh, chmod it and run it – it’s quicker (kinda)!

Unfortunately the wordpress shortcode API can’t successfully process the following:

[intlink id="1" /]
Normal content
[intlink id="2"]Link content[/intlink]

OR

[intlink id="1"][/intlink]
Normal content
[intlink id="2"]Link content[/intlink]

To circumvent this, where users require no link text (to display the post title), the plugin will now accept the string “{{empty}}” as the link content like so:

[intlink id="1"]{{empty}}[/intlink]
Normal content
[intlink id="2"]Link content[/intlink]

I submitted a bug 2 years ago [http://core.trac.wordpress.org/ticket/9264], unfortunately it remains unfixed at the date of this post.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18

(function( $ ){
    $.fn.nextElementInDom = function(selector, options) {
        var defaults = { stopAt : 'body' };
        options = $.extend(defaults, options);
 
        var parent = $(this).parent();
        var found = parent.find(selector);
 
        switch(true){
            case (found.length > 0):
                return found;
            case (parent.length === 0 || parent.is(options.stopAt)):
                return $([]);
            default:
                return parent.nextElementInDom(selector);
        }
    };
})( jQuery );

Download it here: jquery.nextElementInDom.js

And here is how you would use it:

1

var elementNeeded = $('.firstElement').nextElementInDom('.secondElement');

Where your HTML might look like this:

<html>
  <body>
    <div>
      <span class="firstElement">Our first element</span>
    </div>
 
    <ul>
      <li>
        <span class="secondElement">Second element is in a different place</span>
      </li>
    </ul>
  </body>
</html>

Grab a copy of it from the Google Code project page here: php-ssrs

The QuickStart guide should be all you need to get going, there are also more detailed samples available in the download. Getting a rendered report is made real easy, just take a look at the sample below:

1
2
3
4
5
6
7
8
9
10
11
12
13

$options = array(
    'username' => 'testing',
    'password' => 'password'
);
 
$ssrs = new SSRS_Report('http://localhost/reportserver/', $options);
$result = $ssrs->loadReport('/Reports/Reference_Report');
 
$ssrs->setSessionId($result->executionInfo->ExecutionID);
$ssrs->setExecutionParameters(new SSRS_Object_ExecutionParameters($result->executionInfo->Parameters));
 
$output = $ssrs->render('HTML4.0'); // PDF | XML | CSV
echo $output;