There’s lots of different techniques for achieving this unfortunately they don’t seem to ever be explained very well as they relate to MySQL. Until now… dun dun dun dahhh.

The two most common technique are the Haversine formula and the Spherical Law of Cosines. The method I’ve chosen to use in my example is the Spherical Law of Cosines. The formula is a bit short and uses fewer mathematical functions.

The example is comprised of two parts. The first is the Spherical Law of Cosines. The second part is defining a search radius. Defining this search radius will significantly narrow the scope of our search. This is a good thing.

Alright enough chitchat. Here’s the sample MySQL query. It’s built in PHP below but can easily be converted to your language of preference.

Note: I’m Canadian so the example below is done using kilometers. If you want miles you’ll need to multiple $earths_radius and $surface_distance_coeffient by 0.6214 to convert them to miles.

// Search parameters
$lat = 45.411572;
$lng = -75.698194;
$radius = 25;
 
// Constants related to the surface of the Earth
$earths_radius = 6371;
$surface_distance_coeffient = 111.320;
 
// Spherical Law of Cosines
$distance_formula = "$earths_radius * ACOS( SIN(RADIANS(latitude)) * SIN(RADIANS($lat)) + COS(RADIANS(longitude - $lng)) * COS(RADIANS(latitude)) * COS(RADIANS($lat)) )";
 
// Create a bounding box to reduce the scope of our search
$lng_b1 = $lng - $radius / abs(cos(deg2rad($lat)) * $surface_distance_coeffient);
$lng_b2 = $lng + $radius / abs(cos(deg2rad($lat)) * $surface_distance_coeffient);
$lat_b1 = $lat - $radius / $surface_distance_coeffient;
$lat_b2 = $lat + $radius / $surface_distance_coeffient;
 
// Construct our sql statement
$sql = <<<SQL
SELECT *, ($distance_formula) AS distance
FROM listings
WHERE (latitude BETWEEN $lat_b1 AND $lat_b2) AND (longitude BETWEEN $lng_b1 AND $lng_b2)
HAVING distance < $radius
ORDER BY distance ASC
SQL;

If your interested in digging deeper into the theory and math behind the Spherical Law of Cosines or the Haversine formula checkout the Movable Type Scrips site they have tons of background information.

At first this appeared to work correctly. But then I tried the app on a touch Blackberry and the touch elements didn’t work properly any more. So I ditched changes recommended by Resilient Coder and went digging into sencha-touch-debug.js myself.

To get Sencha Touch working on both touch and non-touch Blackberry devices you need to make the following changes.

18098
18099
18100
18101
18102
18103
18104

if (!Ext.supports.Touch || Ext.is.Blackberry) {
    Ext.apply(this.eventNames, {
        start: 'mousedown',
        //move: 'mousemove',
        end: 'mouseup'
    });
}

You need to add || Ext.is.Blackberry to line 18098. This enables clicking on Sench Touch ui elements like buttons and tabs with the thumbpad or trackball. You also need to comment out line 18101 to enable scrolling via touch on Blackberry devices like the Torch. If you don’t comment this line out scrolling via touch doesn’t work properly.

The final piece of the puzzle is to enable scrolling using the thumbpad or trackball. This is really more of a work around. It’s based off the code I found at http://pastebin.com/Mih4Ps12. To use this work around you basically just extend your scrollable panels from Ext.NonTouchCompatibleScrollPanel instead of Ext.Panel.

Ext.NonTouchCompatibleScrollPanel = Ext.extend(Ext.Panel, {
    scroll: 'vertical',
 
    initComponent: function() {
        Ext.get(document.body).on('mousemove', function(evt, div, el) {
 
            if (!this.scroller) {
                return;
            }
 
            if (this.scroller.offsetBoundary.top === 0) {
                this.scroller.updateBoundary();
            }
 
            var maxOffset = this.scroller.offsetBoundary[1] - 400;
            var bottom = parseInt(Ext.get(document.body).getStyle('height')) - 15;
 
            if (evt.xy[1] > bottom && this.scroller.offset.y >= maxOffset) {
 
                var offset = -1 * this.scroller.offset.y + 10;
                this.scroller.scrollTo({
                    x: 0,
                    y: offset
                });
 
            } else if (evt.xy[1] < 60 && this.scroller.offset.y <= 0) {
 
                this.scroller.scrollTo({
                    x: 0,
                    y: (-1 * this.scroller.offset.y) - 10
                });
 
            }
        }, this, {element:'el'});
 
        Ext.NonTouchCompatibleScrollPanel.superclass.initComponent.call(this, arguments);
    }
 
});

This command which can be used in either E-TextEditor or TextMate will decode the quote-printable encoding. It operated on either your current selection or the whole document. Internally the command itself uses Python to decode the quoted-printable text.

Download the quoted-printable decoder Sublime Text command
Download the quoted-printable decoder E-TextEditor / TextMate command

It lets you cleanup the super scrunched up code into something more readable. This enables you to recover your source files if you lost your originals or dig through someone elses code and get a better idea of how other people in the real world are coding.

The initial trigger for the message seems to be random. But after it initial message is seems to come back around every 3 minutes. So I did some digging. After poking around in the source I discovered that they’re using an older version of Tako Sano’s jQuery history plugin.

So I added some code to the historyCallback do some logging so I could get a better feel for what was going on.

4411
4412
4413
4414
4415
4416
4417
4418
4419
4420
4421
4422
4423
4424
4425
4426
4427

errorOccurred = lastErrorOccurred = new Date().getTime();
_back = function(h) {
    if (_backOk) {
        if (_backCallback) {
            _backCallback();
            _backCallback = null;
        }
        var hh = parseInt(h);
        var m = $.browser.netscape ? 2 : $.browser.msie ? 3 : 7;
        if (hh > m) {
            errorOccurred = new Date().getTime();
            console.log("hh: "+ hh +", m: "+ m +", delta: "+ (errorOccurred - lastErrorOccurred));
            lastErrorOccurred = errorOccurred;
            alert(PP.msg.man.bac);
        }
    }
};

After letting this run for a while I started to see a pattern. Here’s an excerpt of the results.

• hh: 8, m: 7, delta: 209901 (~3.4 minutes)
• hh: 8, m: 7, delta: 200888 (~3.3 minutes)
• hh: 8, m: 7, delta: 190944 (~3.2 minutes)
• hh: 8, m: 7, delta: 183514 (~3.0 minutes)
• hh: 8, m: 7, delta: 197517 (~3.3 minutes)

For webkit (Chrome/Safari) based browsers the jQuery history plugin manually creates a stack to keep track of the history. So this is where I believe we are encountering the problem. If you look at line 4426 you can see that we create a history with a stack length of 9.

4412
4413
4414
4415
4416
4417
4418
4419
4420
4421
4422
4423
4424
4425
4426
4427
4428
4429
4430
4431

_back = function(h) {
    if (_backOk) {
        if (_backCallback) {
            _backCallback();
            _backCallback = null;
        }
        var hh = parseInt(h);
        var m = $.browser.netscape ? 2 : $.browser.msie ? 3 : 7;
        if (hh > m) alert(PP.msg.man.bac);
    }
};
runHistory = function() {
    $.historyInit(_back);
    z = $.browser.netscape ? 2 : $.browser.msie ? 4 : 8;
    for (var j = z; j >= 0; j--) $.historyLoad(j);
    _SetTimeout(function() {
        _backOk = 1
    },
    3000);
};

But on line 4419 we are working with a stack with a length of 8.

4419

        var m = $.browser.netscape ? 2 : $.browser.msie ? 3 : 7;

Shouldn’t this read as follows?

4419

        var m = $.browser.netscape ? 2 : $.browser.msie ? 4 : 8;

In Firefox type about:config into the address bar. Now type dns into the filter. Double-click on the network.proxy.socks_remote_dns to configure Firefox to send our DNS requests through the proxy. Voila! We can now laugh it up watching YouTube clips to our hearts content.

A great solution for scanning those small sketchy files that all of us end up downloading at some point to get our software to work they way we want it to.

It’s the little things like this that make Windows 7 really cool. I won’t talk about it too much right now but what makes Windows 7 worth while for me is the new Explorer interface. So much better than Vista’s Explorer interface.

First we’ll setup the ssh tunnel. For that we’re going to use an application called Putty. To setup the ssh tunnel you’ll need to have access to a server based in the US with ssh enabled (for example – a Dreamhost account). For the purpose of the tutorial we’ll call the server “example.com”. After launching Putty you need to enter the hostname into the field provided. Now comes the magic. Expand the SSH category and select Tunnels. Under source port enter 8080, select the “Dynamic” and then click the Add button.

Click on the Open button. After click on Open you’ll be prompted for your username and password for, in the case of this example, example.com. After successful establishing your ssh connection all you have left to do is setup Firefox to use this ssh tunnel as a proxy.

Open up Firefox and select “Options” from the Tool menu then click on “Advanced”. Next click on the “Network” tab, then the “Settings” button. Select “Manual proxy configuration”. Now comes the important settings.  Erase the HTTP Proxy, SSL Proxy, FTP Proxy, and Gopher Proxy fields. Enter 127.0.0.1 under SOCKS Host and 8080 under Port.

Click OK and click OK once more. We’re done! Browse to www.pandora.com and crank the tunes.