innovation initiative

Dave, the co-creator of our Steve Jobs Facts project has just launched IE Death March.

“Internet Explorer 6 will be SEVEN years old on August 27th. It came out a few weeks before the Twin Towers fell. It came out before the Nintendo GameCube. It came out before the first iPod.

It’s time to put a deadline on dropping IE6, and I say that time is now, and the deadline should be soon… say like, March 2009. That’s roughly a little more than 6 months.”

So, we’ve officially stopped supporting IE6 regarding SJF. From this point on we wont be testing any changes to the site in IE6. Not that it ever worked properly in IE6, but it does (for now) work at least.

Both me and Dave are, like IE Death March states, planning on halting all support for IE6 in our development efforts by March 2009. Hopefully IE Death March catches on and inspires more developers, and front line web-services to also halt supporting IE6, and encouraging more users to update to IE7 or switch to a better more modern browser.

IE6 has for many years a been filter of creativity with web development for me, and many others, here’s me hoping it wont be so in six months time.

Yepp, its what it sounds like, its stevejobsfacts.com.

And with that I’m ending my six months of silence. It’s been a crazy, tiring and exhausting six months. But I’m not gonna go for months again without a word here.

Anyway, to get back on topic. Steve Jobs Facts is pretty much what it sounds like. It’s a satirical fact index about Steve Jobs. You vote on your favorite facts, and submit your own. Think of it as the bastard child of Chuck Norris Facts and bash.org with fancy sexiness sprinkled on top like fairy dust.

SJF is a project me and Dave (@sxtxixtxcxh) started last summer. Yes, I did say last summer. But due to circumstances, hiccups, and laziness, we never got the project ready for a launch. But this week we both decided to get SJF live before WWDC even if it killed us (not really, but sounds better than “if we fail, we suck!”, or “you suck” like Dave said it… lol).

The project is still very much beta/alpha though. We didn’t implement user registrations till 2 hours after we launched the site. Right now its very simple, but very usable at the same time.

We’ll be adding more features over the next few days, and some code cleanup behind the scenes to make things runs slightly faster and smoother. Not that anyone will notice the speed improvements though unless we get slashdotted or something *keeps fingers crossed and makes cute puppy eyes*

Today I released a small and quick side-project called LiteMySQL. It’s basically a PHP5 class which is designed to automate the boring repetitive tasks of opening and managing database connections, looping through the query result resource to get an array and so on.

I created this for a small project I’m working on which basically needs database access on less than a handful of pages. And so I’d have a ready to use MySQL solution other small projects in the future.

And sorry for the lack of information on the project page as of now, but I gotta finish off the original project which motivated me to create LiteMySQL. Once done I’ll update the project page with relevant information

Example Usage:

# general usage
$sql = new litemysql('host', 'username', 'password', 'database', 'table');
$rows = $sql->find_all();

# conditions
# - the following three uses of the find
#   function all produce identical results
$result = $sql->find(3);
$result = $sql->find(array('id' => 3));
$result = $sql->find('`id` = 3');

# insert a single row
$sql->insert(
   array(
      'title' => 'hello world',
      'body' => 'my first blog post :D',
      'author' => 'John Doe'
   )
);

It’s been a long while since i last updated here, life’s been crazy. But i’m not gonna fall into a bottom less pit ranting about everything thats been going on and keeping me busy.

Instead, I’d like to announce parseCSV 0.3.0, my fully featured CSV parser for PHP.

I recently created a Google Code project for parseCSV to bring the class some more attention and spotlight than it’s been getting here dug down in my blog somewhere. I also updated the script to fix a small bug (just plain stupid code), and add some more functionality. Namely, offset and limit options when parsing a CSV file/data. And a basic but very useful conditions option which uses SQL-like syntax to match parsed data for easy filtering.

You can download the script from the project page. The wiki pages are still not really complete, at all, but i’ll get there

With all the general grumble about the Safari-is-the-SDK-for-the-iPhone on the internet, I figure that… hey, if the internet *is* where all the iPhone apps are, all web apps are iPhone apps. So, I made a badge.

Feel free to include this image on your Safari Tested™ blog (or web app) if you feel like Steve is brushing you off like a cheap $2 hooker.

Download (the badge, you perv, you can find your own hookers on your own time).

It would seem I am the only one who’s ever seen the ‘Made for iPod’ badge, so don’t feel bad if the joke is lost on you.

Reflections are all the buzz right now, if you don’t have reflections, you’re just not cool, and if you do have reflections, you’re just plain stupid for making out like everyone else… hehe…

Eitherway i’m a huge reflections addict — yes i admit it, i’m an addict — and i tend to use them a bit to much even sometimes, even in webdesign

This brings us to an awesome javascript which generates reflections for images on the fly, many of you (two people who’ll read this… lol) already know all about Cow’s reflection.js script. The one thing i’ve wanted from it tho, is being able to set fixed height reflections, rather than the reflections being X% the height of the image in question.

Hence i spent this afternoon haxx0ring reflection.js to add a rheightpx option, which with a little logic, you’ll figure what it does ;)… while i was at it, i also added a rgap option which puts a gap of X pixels between the image and the reflection.

As for bugs, well, the rheightpx option works fine in all browsers except Opera, which for reasons thats completely beyond my understanding, causes some very weird effects… after 2 hours of testing and messing around, i can’t find any reason for why it does it, but maybe Cow will know, and hopefully he’ll implement my changes into his script *hint*

Also, the rgap option is a bit sketchy between browsers unfortunately

Download:

And if you’re wondering about what you could possibly use a fixed height for, check this…

…which is the main reason i put effort into this modification today

If you’ve had a chance to check out my 48 hour search project from about 2-3 weeks ago (i don’t really remember, and i’m to lazy to double check, live with it…hehehe), you might have noticed that the results from Google and/or Windows Live Search isn’t always correct. And if you’ve used Internet Explorer to attempt to view the site, well shit, i haven’t even dared doing that cause of the fear! the fear man!

Anyway, jokes aside (or so you are lead to believe), the classes handling the results from the different search engines are far from perfect. The Yahoo class works alright cause of Yahoo’s extremely easy to use search API, but Google and Windows Live’s API’s are far greater pains in the ass to use…… Hence, i didn’t have time to implement them properly (48 hour side project and all, remember? ), but i will be fixing it at some point soon… (hopefully before i mature enough to require reading glasses just to see my own hands)

As for Internet Explorer compatibility, it’s coming, with one or two or all UI elements removed since IE just doesn’t render PNG’s properly even with any and all available fixes (during the limited one page load test i conducted on a PNG heavy page with the fix and IE).

So if you’re lucky, there might be a TITLE tag left for IE users when i’m done removing incompatible elements

And in other news, OMFG i want (need) a MacBook Pro !!! (anybody feel like donating their kidney (i’m seriously joking ))… zee ol’ PowerBook is just not fast enough to run itunes and toast my breakfast at the sametime anymore… darn advances and updates to technology!

over and out, hopefull you haven’t died from brain damage while reading my rant

Mac OS X is an excellent web-development platform if you ask most (except if you ask Windows fanatics). It’s UNIX based, comes with Apache 1.3 pre-installed and ready to rock ‘n’ roll basically. But the default factory settings for Apache and more generally don’t fill all the requirements of the average web-developer, so you change settings, install this, install that, customize a few more things, and before long you’re not sure what you did and where sometimes. And this becomes a problem if/when it comes time to reinstall OSX on your little baby, or simply when your old baby is to old and ugly and you throw it out the window from the 4th floor head first in favor of a brand new and cute little baby. *evil grin*

Whatever the reason, it’s always a pain in the ass to reinstall a system, specially when you’ve installed and reconfigured a lot of background daemons and services. Hence I thought I’d share my method of configuration which makes it quite easy just lifting it all from one system to another.

What I’m gonna outline here is installing PHP 5, MySQL and additional tools, and reconfiguring apache the way I’ve done it.

So lets start by downloading PHP 5 for your system from here:
http://www.entropy.ch/software/macosx/php/

It’s a no brainer, just extract the tar.gz archive and double click the .pkg installer package and follow instructions. This will leave you with a working PHP install located in “/usr/local/php5/”.

Then lets move on to MySQL, download the package for your system here:
http://mysql.org/downloads/mysql/5.0.html#Mac_OS_X_(package_format)

Again, its an easy .pkg installation. Optionally, you can install the preference pane which is an easy way to start and stop MySQL. If you want the MySQL daemon to run at system startup, use the .pkg startup item installer, as the option in the preference pane doesn’t seem to work at all

The MySQL GUI Tools are very useful, and are available here:
http://mysql.org/downloads/gui-tools/5.0.html#OSX

As for Apache, there’s a few things to do, first of all you should make sure you’ve turned on “Personal Web Sharing” under the Sharing preference panel in System Preferences. Next we’re gonna add a single line to Apache’s configuration file. Use any text editor of your choice, to open “/etc/httpd/httpd.conf”, you can use the terminal based pico editor by running this in a terminal:

sudo pico /etc/httpd/httpd.conf

Then right at the bottom of the file, add the following on a new line (if you’re using pico, you can quickly scroll down using ctrl+v):

Include /Users/username/Library/httpd/*.conf

Obviously, make sure to replace “username” with your own username. The idea here is that we have apache automatically include all .conf files located in your own “~/Library/httpd/” folder.

Now, before we do anything else, I recommend you open your “~/Sites/” folder with the finder, create a new folder called “_public”, and move everything inside your Sites folder into the newly created _public folder. We’re gonna reconfigure your _public folder to be Apache’s root directory rather than “/Library/WebServer/Documents”.

Then open the “~/Library/” folder in your home folder, and create a folder named “httpd”, in there lets create “_main.conf”, and put this in it (the “Listen” settings you don’t really need unless you want them):

DocumentRoot “/Users/username/Sites/_public”
Listen 80
Listen 3000
Listen 8080

Again, replace “username” with your own username. Save the file, and run the following command in the terminal and enter your password when asked to relaunch Apache.

sudo apachectl graceful

Open http://localhost/ in a web browser, and you should see the contents of your own _public folder

Then you can continue to create an “aliases.conf”, “virtual_hosts.conf” and more in “~/Library/httpd/” and apache will automatically include them. This way, you only need to do a single modification to the bottom of httpd.conf on any new system aswell, and just place your config files in the correct dir, and you’re ready to rock ‘n’ roll

What I’ve also done myself, is that I create virtual hosts for the different projects I’m working on. So lets say I’ve got a new project now called “Banjo”, then I would create “~/Sites/banjo/” and I’d create a virtual host entry in “virtual_hosts.conf” that looks like this:

<VirtualHost *:80>
   ServerName banjo
   DocumentRoot “/Users/username/Sites/banjo”
</VirtualHost>

Then I open “/etc/hosts” with a text editor, and I create a new entry which looks like this:

127.0.0.1 banjo

Once all is saved and done, I restart Apache with “sudo apachectl graceful”, open http://banjo/ in a web browser, and I’m all set to create another world dominating merciless enemy business killing machine of a site… Heh

Myself I’ve got phpMyAdmin installed in “~/Sites/_public/mysql/”, which makes it accessible at http://localhost/mysql/. Also, I have a virtual host configured, and a entry in the hosts file so I can also access from http://mysql/

Then of course, I use TextMate, and I’ve got a TextMate project with all the Apache and PHP configuration files, which I can quickly open using Quicksilver by pressing cmd+space and typing “httpd” and pressing return cause the project file is called “httpd”. But the rest of my workflow is a story for another time

I hope some of you have found this useful, saionara fili moy…

Ok, so this is gonna sound like i’m a complete n00b, but i’ve recently fallen in love with Exception handling in PHP5 .

The reason why i never even looked at it before, is cause i’ve always strived for PHP4 compatibility, but with this latest project (more on it at a later date ), PHP4 has gone out the window .

As for updates, not much, but i did recently upload a new quick PHP function to Code Yard. It’s called dir_exists(). It’s useful for checking if a directory exists, and if not, it will create it recursively, and return true or false. It works in PHP 4 and 5, in 4 it simply loops to create all required dirs since mkdir() doesn’t support the recursive argument.

Thats it for now, more updates soon, i’ve just been really busy the last few weeks with shit… GRRR

I have to get some real work done, that i can post about, cause this is starting to feel like i’m a broken record player just, cause all i’ve posted about lately is parseCSV{}… lol

Obviously having major hard drive problems haven’t helped

Anyway, i spent the most part of today updating parseCSV{}, finishing an automatic delimiter character detection, cause the original project i created it for, requires such a function. The reason being so users don’t really have to know what a delimiter character is, or which one the file their attempting to upload uses. Eitherway, its a very handy function, and hardly slows down the script at all.

The original reason i started working on it cause i realized some versions of MS Excel for Windows uses ; instead of , as a delimiter no matter which CSV option you choose. And the script was being used to upload that from Excel, which caused issues, so to simplify matters, some type of auto-detection was required…

The auto() function works by simply analyzing a specific number of rows (15 by default) from the beginning of the csv file/data, and eliminating characters that simply can’t be the delimiter cause it doesn’t exist on every row, or the the number of times a character appears in a row isn’t the same on all rows. Rarely will any other character than the actual delimiter get past this stage of elimination, but if more then one character does, the script goes on to further analyze details to choose which character is the most likely to be the delimiter.

I just updated and added a few PHP scripts to Code Yard that might be useful for some

• Updated and fixed some bugs in parseCSV{} and parseINI{}.
• speedometer{} — Class for simplifying calculation of script execution time.
• walkDir{} — Class for recursively getting a list of directory contents. It returns both a multi-dimentional array, and a plain array with full paths, both types of output are optional increase performance. The number of directory levels to walk down into is also configurable.
• rfile() & wfile() — Functions for reading and writeing to local files. I wrote these about 3-4 years ago, and i’ve used them ever since, they’ve undergone some slight modifications since then, but their still basically the same. Their quite flexible, fast, and save a lot of time.
• randpass() — Function with a quite self-explanatory name. Generates a random password of any desired length using the character ranges you specify. It can be made more flexible, but its perfect for most situations when you need to generate a random password.

If you’re using SVN for keeping your source safe for web apps, you may notice uploading the working folder also uploads all those hidden .svn folders. Not ideal.

Here’s the shell command to delete all those .svn files from your working directory:

find -d "your/working/directory" -name ".svn" -exec rm -r '{}' ; -print

The -d flag (-depth) means the find command will “process each directory’s contents before the directory itself” so the rm (delete) command won’t complain it can’t find a file in the structure it just deleted.

(I am not responsible if this screws up your project, I assume if you’re a developer and you’re using SVN you *should* know what you’re doing. However, just in case you don’t, once you run this command you won’t be able to update your working copy, nor commit or anything else, so… [insert cautionary advice here])

Today i’ve setup a new project, called Code Yard. All it really is, is a Subversion repository for me to collect all my small random pieces of code (classes, functions, etc.) which i reuse every now and then. It’s purpose is to let me more easily manage all the small pieces of code currently spread out a bit all over the place in old and current projects, and to allow others who might find my code useful, or just save them 5 minutes of work

And with that explained, the first piece of code i’ve added is a PHP class called parseCSV. I’ve recently written this class for a small project i’m working on, i was quite surprised to find that there was a quite lack of CSV support in PHP, and an even larger vacuum of 3rd party classes available that fully supports the most common types of CSV files.

Hence i wrote parseCSV to be fully compatible with the Wikipedia article available on the topic. Which also seems to be the exact CSV formatting MS Excel uses for example. I based my class on the processing concept of Ming Hong Ng’s CsvFileParser class.