Posted (Updated ) in Linux

This is an issue that plagues me constantly. For each PPA you add to Ubuntu, you also need to import a GPG key for it. This is all fine and well – until don’t have the key and are unsure how to get it (often happens after a reformat). Try doing an update without a valid key and you’ll get the following:

BSOD - Ubuntu Updates style
BSOD – Ubuntu Updates style
W: GPG error: The following signatures couldn’t be verified because the public key is not available: NO_PUBKEY

Well we finally have a solution: launchpad-getkeys! This handy little tool will automatically determine which PPA’s require GPG keys and import them for you.

Finally. Sweet relief!
Finally. Sweet relief!

Install with:

sudo apt-add-repository ppa:nilarimogard/webupd8
sudo apt-get update
sudo apt-get install launchpad-getkeys

Run with:

sudo launchpad-getkeys

Images taken from this howtogeek article. For more information see here and here.

Read More »

Posted (Updated ) in Linux

This weekend I was trying to run a server on port 9000 but something was already on there. I had no idea what it was and needed to find and stop it. Here’s a super quick and easy way to do so:

You’ll need lsof:

sudo apt-get install lsof
lsof -iTCP:9000

The result I got was as follows:

komodo-bi 31713 myuser   51u  IPv4 1234567      0t0  TCP *:9000 (LISTEN)

From there it was as simple as going into komodo and turning turning off its listener 🙂

Thanks to Laran Evans for his useful tutorial.

Read More »

Posted (Updated ) in Javascript, PHP

If you’re reading this page (or my blog in general) it’s a pretty safe bet you already have Google+. If you’ve uploaded any photos to Google+ from Chrome or FF, you would also have noticed its snazzy HTML5 file uploader at work. This weekend I took it upon myself to whip up a quick and dirty version of that uploader and share its inner workings with the world.

Google+ Image Uploader
Google+ Image Uploader

Here’s a short video of my uploader at work:


  • PHP4+
  • HTML5-enabled browser (File API – including drag and drop, XHR2)
  • Some images to upload

Disclaimer: Currently only Firefox and Webkit based browsers meet the requirements above. Opera supports the File API and probably XHR2 (I haven’t tested), however it doesn’t have drag and drop so this tutorial won’t work with it. If you’re curious about whether or not IE will work with this tutorial, I’m already laughing at you.

Download the finished script here.

Read More »

Posted (Updated ) in Linux

Update 2011-10-26: Added support for Ubuntu 11.04+ which uses slightly different FTP folder names

Lack of FTP keepalive functionality in Nautilus has been one of my biggest gripes in Ubuntu for a long time now. It’s infuriating attempting to open a folder only to discover the session has timed out and I need to reconnect. Well that problem is now solved!

Run the following bash script in the background each time you boot and you’re good to go:

while true
	#10.10 and earlier
	ls ~/.gvfs/ftp* &> /dev/null
	ls ~/.gvfs/FTP* &> /dev/null
	sleep 15

Many thanks go to the user who originally posted this script in a mailing list here.

PS. If anyone knows how to integrate this into a nautilus script I’d much appreciate it!

Read More »

Posted (Updated ) in Linux, Uncategorized

I’m currently working on a mobile version of an existing website utilizing the Responsive Web Design paradigm. One problem I instantly came across was a perplexing page width issue on Android. Even with a blank, HTML5 webpage, the page width was appearing at almost twice the width of my phones native resolution (and 2.5x that of my browsers width). For the record, I’m using Android 2.3.3 vanilla with the default browser on a Nexus One.

Firstly an example:

<!DOCTYPE html>
<html xmlns="">
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<title>Untitled Document</title>
<script type='text/javascript' src='' />
<script type='text/javascript'>
	$(document).ready(function() {
		document.body.innerHTML = $(window).width();

The above code gives a blank, HTML5 webpage that will display the pages width on load. When loaded, it would print ‘800’ on the screen – indicating the page was set to an 800px width. This was clearly wrong.

I quickly noticed that using the following doctype:

<!DOCTYPE html PUBLIC "-//WAPFORUM//DTD XHTML Mobile 1.0//EN" "">

gave me the width I was expecting – 320px. Obviously this isn’t an acceptable solution for responsive web design and so another was needed. Meet viewports. With the addition of a simple META tag I was able to fix the issue (albeit losing the ability to zoom in the process). Simply add the following to your HEAD tag and you should be good to go:

<meta name="viewport" content="width=device-width; initial-scale=1.0; maximum-scale=1.0; minimum-scale=1.0; user-scalable=0;" />

With this line added, my page now prints 320 to the screen as it should.

For more information, please see this StackOverflow post on the issue.

Read More »

Posted (Updated ) in Javascript

It appears JQuery Form Validation supports returning a string when using the remote validator but it’s not immediately clear how to do this and the examples tab doesn’t have an example for doing so. Getting this working is actually very easy. All you need to do is return a JavaScript string in your response. Here’s an example:

Firstly, the javascript:

$(document).ready(function() {
		rules: {
			username: {
				required: true,
				remote: "/your/url.php"
		messages: {
			username: "Username is required."

Pretty standard, nothing out of the ordinary here. Now for the server side response. As I’m sure you’re aware, you can return ‘true’ or ‘false’ however if you want a string, it needs to be formatted as a javascript string:

// /your/url.php
echo '"This username is already taken. Please enter a different username and try again."';

Read More »

Posted (Updated ) in Linux

Tonight I needed to transfer files directly between two servers via the terminal and figured this would be useful information for others, so here’s how to do it:

Upload to a remove server:

$ ssh username@remote_address cat < localfile ">" remote_file

Download to your server:

$ ssh username@remote_address cat remote_file > local_file

Source: the incomplete news project

Read More »

Posted (Updated ) in Uncategorized

I’ve been frustrated for a while now at Chromes lack of a decent Amazon S3 extension. Firefox has S3Fox but the closest Chrome users come is S3 which only lets you browse and not modify the files in your buckets. Well that’s just changed!

Over the weekend I finally gathered up enough free time to add support for creation and deletion of files/buckets. While I was in there I made a few other modifications to improve performance and clean things up a bit.

Download the latest version (and the source: Github repo)

Adding a bucket
Adding a bucket
Deleting a bucket
Delete link when hovering over buckets
Adding files
Multi-select file browser for uploading files
Uploading Files
The uploading window
File delete
The confirmation you see when deleting files
Deleting a bucket with files in it
Error you see when attempting to delete a bucket with files in it

Read More »

Posted (Updated ) in Database

Earlier today I completed my first project in Django and it came time to do some database optimization. I wanted to get a list of SQL queries executed for each page and a bit of Googling let me to this script on DjangoSnippets. It did everything I needed it to do, however I noticed it interfered with dynamically generated binary file outputs (such as the images made with django-simple-captcha). For this I needed to check if the output was in binary, and if so just return it without attempting to print the SQL log. I found what I was looking for here and after combining the two had the perfect SQL logger! Below is my finished code.

from django.db import connection
from django.template import Template, Context
import string 
# SQLLogMiddleware:
    def process_response ( self, request, response ):
        #Don't print SQL queries for binary outputs!        if istext(response.content) == 0:            return response 
        time = 0.0
        for q in connection.queries:
            time += float(q['time'])
        t = Template('''
            <p><em>Total query count:</em> {{ count }}<br/>
            <em>Total execution time:</em> {{ time }}</p>
            <ul class="sqllog">
                {% for sql in sqllog %}
                    <li>{{ sql.time }}: {{ sql.sql }}</li>
                {% endfor %}
        response.content = "%s%s" % ( response.content, t.render(Context({'sqllog':connection.queries,'count':len(connection.queries),'time':time})))
        return response
# istext(s):
    if "\0" in s:
        return 0
    if not s:  # Empty files are considered text
        return 1
    # Get the non-text characters (maps a character to itself then
    # use the 'remove' option to get rid of the text characters.)
    t = s.translate(string.maketrans("", ""), "".join(map(chr, range(32, 127)) + list("\n\r\t\b"))) 
    # If more than 30% non-text characters, then
    # this is considered a binary file
    if float(len(t))/len(s) >= 0.30:        return 0

To get this working on your site just add it to your MIDDLEWARE_CLASSES in and make DEBUG is set to True.

Read More »

Posted (Updated ) in PHP

While aimlessly browsing my Twitter feed last night, I came across this tutorial on using CodeIgniter‘s Zip class. Kohana 3.1 doesn’t seem to come with similar functionality out of the box (at least from the brief glance I took), so I figured I’d write up a quick patch to get CI’s class working with it.

Download the module with example controller and test files here.

You can read up on how to use the class in CodeIgniter’s documentation or the tutorial linked to above. I’ve included a controller and 2 test files in the download above as a kind of ‘quick start guide’. To access them simply go to /koziptest and /koziptest/invalid for a successful and unsuccessful download respectively.

There isn’t really any more to say about this one other than the fact that I made the methods in my module chainable for convenience and consistency with other Kohana classes. Note that this means the read_file, read_dir and archive methods will now throw exceptions if you use an invalid file/directory path or your input or output files aren’t readable/writable – so remember to use a try..catch block! I’ve also made the download method take a Request variable as its first argument – simply pass your controllers current request to it like so:

$Zip->download( $this->response )

Read More »

Page 10 of 16« First...89101112...Last »