dp's blog » programming

The Quick and Dirty Guide to a Successful Website Deployment – Part 1: Research

gregg — Sat, 05 Jun 2010 20:26:13 +0000

So, you’ve been charged with the responsibility of giving your company website a facelift. Your logo looks as though it was designed by a six year-old, the overall design screams, “1998,” and the blinking text effects are starting to cause problems with your epileptic readers.

Let’s be honest. You’ve got a lot of work ahead of you and the next 3-4 months are going to suck. But, before firing up the ol’ Photoshop, before nesting 37 div tags into a single wrapper, and before punching your fist through the wall, consider the following bits of advice that may make this project a bit more tolerable.

Research

Every time I embark on a new site design, I always go hunting for prime website examples within the industry at hand. The goal here isn’t to start ripping off the competition’s site design, graphic design, copy or product images. You’re looking for ideas and inspiration.

Perhaps Company “X “ has a great slideshow that engages the user the moment he/she visits the site. Perhaps Company “Y” has some very helpful widgets on the homepage that display their newest and most popular products. Perhaps Company “Z” has an exceptional method of generating leads (such as a non-intrusive “Request for Information” form that’s shoved in an I-frame or Lightbox). You get the idea.

Your goal is to create a general site layout in your head and have a running tab of ideas you can present to the development team BEFORE the coding and design processes even begin.

If you can pull it off, you’ll be hailed as a brilliant project manager that KNOWS how to get things done in a timely manner. If you fail, prepare to be called every name in the book. Prepare for your co-workers to doubt your abilities and prepare for your bosses to doubt your competencies as a manager.

The Quick and Dirty Guide to a Successful Website Deployment: Part 2 – Objectives and Strategies Simply put, objectives are WHAT you plan to achieve and...

Databases and backups

dp — Thu, 06 May 2010 15:05:42 +0000

Image by jeromejtk via Flickr

Yesterday, I managed to delete the primary database that we use at work. I won’t go in to how this happened, other than to say: Fuck you MySQL Workbench. I was trying to get a map of the table relations within the database. Needless to say, it didn’t work. After the immediate pants-wiping, I restored the database from the backup. But the backup was from the previous morning, as the current backup hadn’t finished. Unfortunately, I couldn’t find the files from the mornings backup.

In short, losing data sucks. How often do database “pros” backup their data?

How to fix strange characters in MySQL database – Vinh Pham I admit, I borrowed this from How to fix strange characters...
udev debugging toolbox « Web 0.2 This is a great read about working with udev, specifically,...

udev debugging toolbox « Web 0.2

dp — Mon, 12 Apr 2010 15:32:24 +0000

This is a great read about working with udev, specifically, getting things working after you’ve written your rules, and they still don’t work.

Remember that udev has a different $PATH than everything else, so qualify your paths to mount, etc.

udev debugging toolbox « Web 0.2.

gms8994′s benchmark at master – GitHub

dp — Mon, 11 Jan 2010 18:31:14 +0000

I’ve recently been posting my benchmark scripts to github. You can find them here: gms8994′s benchmark at master – GitHub. I’ve also started posting the results within the wiki, each page with the same name as the script. For example: comparing MySQL inserts with one set of values, or multiple sets of values.

Databases and backups Yesterday, I managed to delete the primary database that we...

Codeblocked: An Adventure in Slow Code

dp — Wed, 16 Dec 2009 20:50:19 +0000

It’s amazing how much content a company can add in a year, especially when that company writes articles for it’s living. A year and a half ago, I wrote this code against Sphinx:

$cl = new SphinxClient ();
$res = $cl->Query ( $q, $index );
foreach ( $res["matches"] as $docinfo ) {
    $excerpt = $cl->BuildExcerpts(array($contents), $index, $q, $options);
}

Now, for those of you that don’t know, Sphinx’s BuildExcerpts function takes an array of textual scalar values, and generates the excerpts from them. If you’ll notice, I’m cutting off BuildExcerpts at the knees.

18 months and a million searches later, and I’m trying to revamp this. It’s not looking much prettier, but it is *much* faster….

Jeremy Zawodny: Recent Sphinx Updates (jeremy.zawodny.com)
Comparison Between Solr And Sphinx Search Servers (Solr Vs Sphinx – Fight!) (beerpla.net)

Cleaning up bad data in an SQL database

dp — Tue, 02 Jun 2009 19:13:16 +0000

Recently, I needed to find all of the data in a table that was being used for HTML output, that had non-printable characters. The following SQL statement should do that for you:

select id, name
from table_name
where name not regexp '^[a-zA-Z0-9_ ,&;\/<>(²):%*#?$|!+.\"\'-]+$;

This code is MySQL specific, but should be applicable anywhere, really.

How to fix strange characters in MySQL database – Vinh Pham I admit, I borrowed this from How to fix strange characters...
Databases and backups Yesterday, I managed to delete the primary database that we...
gms8994′s benchmark at master – GitHub I’ve recently been posting my benchmark scripts to github. You...