The problem is that I’m an anal perfectionist and the Flash object was eating 80-90% of a computer’s CPU when being viewed… and me being me, I couldn’t just be okay with that.

…so I rebuilt it with CSS3/HTML5/jQuery. It turned out much nicer in my opinion, only uses about 1% of the resources that Flash needed and as an unintentional bonus, it works on things like iPad and phones/computers without Flash support.

I’m not anti-Flash (as I said, I *wanted* to do the site with Flash), but I AM anti-inefficient. There are still some things you can’t do with HTML5 that you can do with Flash, but those things are becoming fewer and fewer it seems.

Sure enough!

Use memcached… no really, use it.

memcached is a distributed memory caching system that allows multiple servers to access the same shared memory (you can use it just for single local server too of course). You can cache pretty much anything… SQL query results (especially ones that can’t hit indexes), dynamically generated web pages, dynamic RSS feeds, etc.

If you compile the memcache() functions/class directly into PHP, you have a an easy (yet powerful) way to incorporate it.

For example, let’s say I have some existing code that makes a call to some sort of API (I use memcached for the keyword suggestion tool in this way)… you can add a couple lines of code to cache the results.

[code=php]$data = @file_get_contents ("http://some_api_call_url?keywords=$keywords");[/code]

could become this:

[code=php]$memcache = new Memcache;
$memcache->addServer('192.168.1.1');
$memcache->addServer('192.168.1.2');
if (!$data = $memcache->get($keywords)) {
$data = @file_get_contents ("http://some_api_call_url?keywords=$keywords");
$memcache->set($keywords, $data, false, 86400); // 24 hour cache
}
$memcache->close();[/code]

Now we are only hitting the API for the same keywords once every 24 hours… that setup utilizes 2 memcached servers working as a single entity for pooling and failover.

I use memcache for all sorts of things (this blog’s pages are cached for 60 seconds as another example because WordPress sucks, and protects against a massive influx of traffic [front-page digg for example]).

Here’s some things I personally use it for…

• Caching this blog’s pages (for 60 seconds)
• Caching Keyword Suggestion Tool queries (for 7 days)
• Caching AdSense Sandbox queries (for 24 hours)
• Temporary counters in Keyword Tracker (using decrement() for displaying remaining keywords on a “Check All”)
• Recent forum topics that show on all the webmaster tools
• vBulletin forum datastore

Any high-traffic site that is dynamically generated (especially if it includes SQL queries) could benefit from it… you could manage which memcache items need to be updated when you insert/update to the database, or an easy way to do it is just check the cache for the pages that potentially have high traffic… if they aren’t in the cache, put them in the cache with a certain expiration date. Then the page is dynamically generated no more often than the cache expiration time (in the case of this blog, the pages are generated no more than once every 60 seconds).

P.S. - Make sure you block outside access to your memcached port since there’s no authentication used!

Google’s Code Search is *really* handy for programmers. For example, say I wanted to see sample code that uses the preg_match() function in PHP, I could do this query:

preg_match lang:php

Neat!

A couple days ago I migrated a cluster of web servers to PHP 5.1.x (they were running PHP 4.4.x before). One of the things I run on them is a geo targeting mechanism, and in the case where my screw-up was, a function to block a subnet of IP addresses.

[code=php]echo sprintf("%u", ip2long(' 80.1.2.3'));[/code]

That little bit of code was my fuck up… PHP 4.4.x returns “1342243331” (as expected), but PHP 5.1.x would return “0“. PHP4 was more forgiving of the stray leading space before the IP address, while PHP5 was not. The end result was that every IP address before 81.1.2.3 was blocked, instead of blocking FROM that IP address to an end IP address (that was not very many IPs away).

Awesome… so what happened here? I ended up blocking about 40% of the potential traffic for awhile, which equated out to probably about $10,000 in lost revenue.

The lesson learned? Don’t fuck around with spaces, because they will cost you many hours of debugging/tracing as well as ~$10,000.

That would be handy… can someone please make it so?

So here’s a basic PHP script that uses the class to take control of the camera, set the pan/tilt/zoom to a certain position, capture the image and save it to a file… handy for uhm… I dunno… a cron job to take time lapse photography.

[code=php] include ('class_camera.php');

$cam =& new Camera();

$cam->connect('your.camera.ip.address');
$cam->get_control();
$cam->goto( -7463, -548, 4126);
sleep(5); // Wait for movement to finish and lens to focus

$handle = fopen('images/' . time() . '.jpg', 'w');
fwrite ($handle, $cam->image());
fclose ($handle);

$cam->disconnect();
?>[/code]

class_camera.php:[code=php]

// #############################################################################
// Canon Camera Control Class
// http://www.shawnhogan.com/2006/08/canon-vb-50i-php-control-class.html

/**
*
* Class to interface with a Canon VB-C50i or VB-C50iR
*
*/

class Camera
{
/**
* Connects to the specified camera
*
* @param string Hostname or IP address of the camera
*
* @return connection_id
*/
function connect($host)
{
$this->host = $host;
return $this->connection_id = substr(file_get_contents ('http://' . $host . '/-wvhttp-01-/OpenCameraServer?client_version=LiveApplet_4125&image_size=640x480'), 14, 9);
}

/**
* Disconnects from the camera
*
* @return none
*/
function disconnect()
{
file_get_contents ('http://' . $this->host . '/-wvhttp-01-/CloseCameraServer?connection_id=' . $this->connection_id);
}

/**
* Attempts to get control of camera
*
* @return none
*/
function get_control()
{
file_get_contents ('http://' . $this->host . '/-wvhttp-01-/GetCameraControl?connection_id=' . $this->connection_id);
}

/**
* Get camera info
*
* @return array of camera info variables
*/
function get_info()
{
preg_match_all ('#(.*?)=(.*)#', file_get_contents ('http://' . $this->host . '/-wvhttp-01-/GetCameraInfo?connection_id=' . $this->connection_id), $matches);
foreach ($matches[1] as $key => $value) {
$this->camera_info[$value] = $matches[2][$key];
}
return($this->camera_info);
}

/**
* Move camera
*
* @param string Movement type (pan | tilt | zoom)
* @param number Position to move to
*
* @return none
*/
function move($operation, $position = 0)
{
file_get_contents ('http://' . $this->host . '/-wvhttp-01-/OperateCamera?' . $operation . '=' . $position . '&connection_id=' . $this->connection_id);
}

/**
* Set pan/tilt/zoom on camera with one call
*
* @param number pan position
* @param number tilt position
* @param number zoom position
*
* @return none
*/
function goto($pan, $tilt, $zoom)
{
$this->move('pan', $pan);
$this->move('tilt', $tilt);
$this->move('zoom', $zoom);
}

/**
* Set parameter
*
* @param string paramter to set
* @param string value
*
* @return none
*/
function set_value($parameter, $value)
{
$mapping = array (
'back_light' => 'B',
);

$parameter = $mapping[$parameter];

file_get_contents ('http://' . $this->host . '/-wvhttp-01-/OperateCamera?' . $parameter . '=' . $value . '&connection_id=' . $this->connection_id);
}

/**
* Get current image
*
* @return none
*/
function image()
{
return file_get_contents ('http://' . $this->host . '/-wvhttp-01-/GetOneShot?image_size=640x480');
}

}
?>[/code]

Most blog themes float the main body to the left, and make the sidebar on the right “static”. I did the opposite when creating my custom WordPress theme… My sidebar is a DIV that is floated to the right. The reason for this then the content below the bottom of the sidebar wrap around it, which I just think looks better. So here’s the problem… Google weighs content at the beginning of a page higher than content at the bottom, so Google was seeing 99% of my pages as “similar content” because my sidebar needed to be at the top of the HTML source.

So to solve that problem, I just put the sidebar at the bottom of the page in a hidden (with CSS) DIV, then call a simple script to set the contents of an empty DIV (<div id=”sidebar”></div>) at the top of the page, where the sidebar needs to be, to the contents of whatever is in the hidden DIV at the bottom…

For web browsers, the sidebar renders at the beginning of the HTML source, but Google sees the sidebar at the end of the HTML and pages no longer are seen as 99% similar.

It more or less lets you see what’s going on in the forum in realtime.

Check it out: http://forums.digitalpoint.com/spy.php

Unfortunately it also means that I’m still up right now and haven’t gone to bed yet. Goodnight!

I ended up making a datastore class for vBulletin also so I could use it for the forum, so if anyone else is using vBulletin on a server with APC, here you go (if you know what this is for, you will know where it goes ).

[code=php]// #############################################################################
// APC

/**
* Class for fetching and initializing the vBulletin datastore from APC
*
* @package vBulletin
* @version $Revision: 0.0.0.1 $
* @date $Date: 2006/05/08 16:51:06 $
*/
class vB_Datastore_APC extends vB_Datastore
{
/**
* Fetches the contents of the datastore from APC
*
* @param array Array of items to fetch from the datastore
*
* @return void
*/
function fetch($itemarray)
{
if (!function_exists('apc_fetch'))
{
trigger_error('APC not installed', E_USER_ERROR);
}

foreach ($this->defaultitems AS $item)
{
$this->do_fetch($item);
}

if (is_array($itemarray))
{
foreach ($itemarray AS $item)
{
$this->do_fetch($item);
}
}

$this->check_options();

// set the version number variable
$this->registry->versionnumber =& $this->registry->options['templateversion'];
}

/**
* Fetches the data from shared memory and detects errors
*
* @param string title of the datastore item
*
* @return void
*/
function do_fetch($title)
{
$ptitle = $this->prefix . $title;

if (($data = apc_fetch($ptitle)) === false)
{ // appears its not there, lets grab the data and put it in memory
$data = '';
if ($dataitem = $this->dbobject->query_first("
SELECT title, data FROM " . TABLE_PREFIX . "datastore
WHERE title = '" . $this->dbobject->escape_string($title) ."'
"))
{
$data =& $dataitem['data'];
}
$this->build($title, $data);
}
$this->register($title, $data);
}

/**
* Updates the appropriate cache file
*
* @param string title of the datastore item
*
* @return void
*/
function build($title, $data)
{
$title = $this->prefix . $title;

if (!function_exists('apc_store'))
{
trigger_error('APC not installed', E_USER_ERROR);
}
$check = apc_store($title, $data);
}
}[/code]

Just so you know that I am still dorky, here’s some PHP code for you (from my post over here)…

This will put the keywords that someone searched on to reach your site (works with Yahoo, AOL, MSN and Google) into the $keywords variable.

[code=php]$parse = parse_url($_SERVER['HTTP_REFERER']);
$se = $parse["host"];
$raw_var = explode("&", $parse["query"] );
foreach ($raw_var as $one_var) {
$raw = explode("=", $one_var);
$var[$raw[0]] = urldecode ($raw[1]);
}
$se = explode (".", $se);
switch ($se[1]) {
case 'yahoo':
$keywords = $var['p'];
break;
case 'aol':
$keywords = $var['query'];
break;
default:
$keywords = $var['q'];
}
unset($parse, $se, $raw_var, $one_var, $var);[/code]

Weeeeeeeeeee! Can I go to bed now?

Google’s Map API is still better than Yahoo’s IMO, but that’s about it.

I have a couple tables that I was replicating across 3 or 4 MySQL databases because I didn’t think I could reference a table from database A while working with database B.

Today I accidentally tried to do a SELECT on a table from a database I wasn’t even working with and instead of getting an error, it worked. As long as the user you are connected to MySQL as has the proper privileges (duh), it works. You just prefix your table/column name with the database name.

For example:
[code=sql]SELECT database1.tablename.a_column,database2.another_table.another_column
FROM database1.tablename
LEFT JOIN database2.another_table ON(database1.tablename.some_id = database2.another_table.some_id)[/code]

You can do anything you can do with tables in the same database (INSERT SELECT, JOINS, sub-queries, etc.)

Wow, I’m an idiot… its probably been a feature since version 1.0 too. Now if you could just do the same thing across different MySQL servers (something like ipaddress.database.table.field), that would be really handy.

[code=javascript][/code]
This only works with Internet Explorer because of some limitations/bug with Mozilla. I’m too lazy to code a complicated workaround for other browsers (which would involve running JavaScript code whenever the mouse moves which seems really inefficient to me). Hopefully Mozilla will fix the bug (although it’s been pending for 2 1/2 years now).

Basically it will log an AdSense click as a pageview to a non-existent page (http://www.digitalpoint.com/AdSenseClick in my case).

Labels (descriptions, freeway, direction, etc.)
Current Highway Speeds (cross referenced by ID to the Labels file)

Would be nice if they gave you the longitude/latitude of each location, but hey… assign some intern to look them all up.

So let’s see if I could stick the data on Google Maps myself (via API)… I’m fairly lazy, and would rather programatically do something that would otherwise require manual labor (looking up longitude/latitude for a zillion points). Thankfully, I remembered that I looked at Yahoo Maps Beta a few weeks ago, and saw it had live traffic data that was overlayed on the map graphic (which means they probably are making a separate HTTP request for just the traffic data). Sure enough… a packet sniffer reveals this URL for getting traffic data for San Diego:

XML Feed

Okay, that’s not bad, but doesn’t include longitude and latitude. Instead it appears the speed data is relative to the current view of the Yahoo map you are looking at. Hmmm… rather than decode their little proprietary system, let’s see what sfmt is. With some luck, it’s “speed format”. Sure enough… change it to sfmt=0, and you get longitude/latitude of the points! So now we have a nice XML feed with longitude and latitude:

XML Feed

Once you have that data, it takes 2 minutes to make a Google Map with the data…

Traffic In San Diego

Of course it doesn’t really answer my original question… why doesn’t Google already have this data on Google Local/Maps? If I worked at Google, this would be my 2nd “20% time project” (this was the other one).