Web 2.0, the big hit. Its everywhere, everyone is doing it. Microsoft has made it very easy for us developers to implment some very cool features of AJAX through the ASP.NET Ajax.NET framework. Simply install, drag and drop and deploy, blammo, you're done.

But there in lies the problem. What about developers who take this for granted? This turns our profession into a melting pot of people who think they can code but actually can't do much of anything other than drag and drop. I'm talking about the developers who implement this type of solution and run with it. They have no concern about how it works, why it works the way it does, it just solves the problem. Little did they know, they went through more work getting this done than is required. Continue on...

Introducing the ASP.NET AJAX Confirmation Button Extender...

This extender enhances usability by giving the user a choice of what to do. For example...

I'm all for ease of use, but sometimes even the smartest people can complicate something VERY SIMPLE. We've all suffered from Analysis Paralysis at one point in our career and well, I think the Confirmation Button Extender is a product of that symptom.

Why would I want to use the AJAX>NET framework to implement a simple confirmation box when I could use the OnClientClick with a fraction of the code?

Here's what I'm saying.

AJAX.NET Implementation

<asp:LinkButton runat="server" id="LinkButton1" Text="test" />

<ajaxToolkit:ConfirmButtonExtender ID="cbe" runat="server"
TargetControlID="LinkButton1"
ConfirmText="Are you sure you want to click this?" />

ASP.NET Implementation with JavaScript:

<asp:LinkButton runat="server" ID="testButton" OnClientClick="return confirm('Are you sure you want to click this?');" Text="test" />

These two methods DO THE EXACT same thing. Yes, the markup output is different, but honeslty, the latter of the two is much easier to implement. To implement this in AJAX.NET you have include the dll and js files, blah blah blah.

The problem I have with the AJAX.NET implementation is that its not helping me gain anything. I have to type more code to get the same thing accomplished. I see this as teaching developers its ok to lean heavily on the design time . This is a problem that we've been solving for years in web dev. How to send a confirmation to a user, and it seems, just now, a lot of people are just figuring it out due to AJAX.NET and they think its "the only way", mainly because ... well ... it worked from a drag/drop perspective. Unfortunately it takes more code than necessary and when the developer finally figures out what it does, they usually say "well why did Microsoft implement it that way? This is soo much eaiser doing by using OnClientClick".

My answer: "So you could get it to work, quickly. If developers cant figure it out on their own, then its time abstract the problem out into another level, and thats exactly what they did. Removed the knowledge of how and why, and replaced it with an easy to implement solution."

We call them R.A.D. Tools and sometimes in this industry its a love/hate relationship with them. :)

I ran into an interesting error today and for a minute I didnt understand why it occurred.

Exception
c:\Windows\Microsoft.NET\Framework64\v2.0.50727\Temporary ASP.NET Files\exampleapp\571e5472\336672e2\App_Web_login.master.cdcab7d2.gh4qw4cs.0.cs(115): error CS0030: Cannot convert type 'ASP.login_master' to 'System.Web.UI.WebControls.Login'
http://localhost/exampleapp/login.aspx?ReturnUrl=/exampleapp/default.aspx
/exampleapp/login.aspx

The site ran fine when running under the local cassini web web server but when we pre-compiled the website through our continuous integration system and then dropped it on our test server, everything went hay-wire. This error started popping up.

After a few moments of reviewing the error I figured it out.

The problem

The master page code behind class was called "Login". ASP.NET 2.0 also has a class by the name of Login. The .NET Framework was trying to convert one type (the master page type) to the actual Login class type.

The Fix

You can fix this one of two ways.

1. Give the full type name in the inherits attribute of the master page.

      e.g.: NameSpace.Type

2. Change the code behind class name from "Login" to something like "LoginMaster".

Conclusion

As I found (after the fact of solving this) Rahul Soni to say ... "dont give a class the name of 'Login'".

Please note, I've also see this problem with class names that are of the following:

• Error
• View
• Wizard

I'm sure there are more, but these are the ones I've see cause this problem before.

If you're new to IIS 7 (you probably are) you might receive this nice little gem when you first start working with it:

HTTP Error 500.19 - Internal Server Error
Description: The requested page cannot be accessed because the related configuration data for the page is invalid.
Error Code: 0x80070021
Notification: BeginRequest
Module: IIS Web Core
Requested URL: http://localhost:80/ExampleApplication/
Physical Path: C:\inetpub\wwwroot\ExampleApplication\
Logon User: Not yet determined
Logon Method: Not yet determined
Handler: Not yet determined
Config Error: This configuration section cannot be used at this path. This happens when the section is locked at a parent level. Locking is either by default (overrideModeDefault="Deny"), or set explicitly by a location tag with overrideMode="Deny" or the legacy allowOverride="false".
FONT color=#a52a2a>
Config File: \\?\C:\inetpub\wwwroot\ExampleApplication\web.config

Config Source:

184: </modules> 185: <handlers> 186: <remove name="WebServiceHandlerFactory-Integrated"/>

IIS 7 implements "Configuration Locking". This is to help with IIS administration. The IIS Administrator can lock down Configuration Sections, Section Elements and Attributes at the IIS level. This allows the administrator to have a more granular level of control on the system. Read this link to see more about it.

Important Note:  In order make these changes, you must be an administrator on the machine where the config file resides. If its on your local machine, you must be an administrator on your machine. 

Since I'm on my own development machine, and since this is a development machine, I have decided to change the global setting (the config section itselft). To fix this error I had to go to the applicaitonHost.config file and set the overrideModeDefault to "Allow".

Here's how:

Open the applicationHost.config file, located here: %windir%\system32\inetsrv\config\applicationHost.config

In my instance, I need to edit the "handlers" section.

Change this line:

<section name="handlers" overrideModeDefault="Deny" />

To:

<section name="handlers" overrideModeDefault="Allow" />

Thats it. :)

You might receive more errors after you enable this, such as the same error for modules, but just follow the same steps above, and you should be good.

Note: You can accomplish this same thing through the command line by using the appcmd.exe application (%windir%\system32\inetsrv\appcmd.exe). Like this:

(Download Excel Spreadsheet example at the bottom.)

Sometimes when developing I need to work with data. Sometimes this data is from external sources such as other partners in the business. Unfortunately when I get this data does not follow a consistent format therefore I'm left with formatting it for entry into a data store, usually a SQL Server.

In my last post about how to create store locator, I had this same problem. I had to import a list of company addresses for that tutorial. The only place I knew where to get a good list of locations for a store locator would be off of Starbucks Store Locator site. Through the trusty copy/paste method, I was able to grab about 200 store results in a matter of minutes from the website.

Now back to the original problem, the data format. The data looked like this after is was copied...

1. 1385 Metropolitan Ave (UCO)
1385 Metropolitan Avenue
Bronx NY, 10462
United States

I needed it to look like this (for ease of use, plus its easier to debug):

1385 Metropolitan Ave (UCO)
1385 Metropolitan Avenue
Bronx
NY
10462
 
Which will finally transform into this:

INSERT INTO dbo.Locations(LocationName, Address, City, State, Zip) VALUES ('1385 Metropolitan Ave (UCO)', '1385 Metropolitan Avenue', 'Bronx', 'NY', '10462')

Hmmm... problem. Now I could take my time, write an SQL insert script for each and every one of these, but that would take forever.

Instead we can use Excel to do this for us.

First off, all credit for this sweet little trick goes to Chris Lawson (a developer I worked with during my time at Todd McFarlane Producitons, SPAWN.COM). He's the one who showed it to me, I'm just showing it to you. :)

To get the data formatted into SQL insert scripts we need to use a few functions from Excel. We'll mainly use LEFT, MID, RIGHT, TRIM and SEARCH. Using those 4 methods you can pretty much get anything you want out of the string, and thats what I've done.

Note: It might be very helpful if you downloaded my sample at the bottom of this post and read this while looking at the sample, just so you can see how it was done.

Step 1: Separate the Data

We're going to want to take our data from this format:

1. 1385 Metropolitan Ave (UCO)
1385 Metropolitan Avenue
Bronx NY, 10462

and organize it to:
 
1385 Metropolitan Ave (UCO)
1385 Metropolitan Avenue
Bronx
NY
10462

Which will eventually get transformed into SQL, like this ...

INSERT INTO dbo.Locations(LocationName, Address, City, State, Zip) VALUES ('92nd & 3rd', '1642 Third Avenue', 'New York', 'NY', '101283618')

a) Get the Title to look like: 1385 Metropolitan Ave (UCO)

Tasks:
 - Get rid of the "1. " 

This will return the text in A1 without "1. " in it.

This takes the contents in cell A1 (the text with the 1. 1385 Metro...) and gets rid of the 1. through some simple text manipulation. The SEARCH method takes a parameter to search for. Utilizing the single character wildcard "?" I'm able to find where the numbering 'ends' and then get the text after that utilizing the LEN and RIGHT functions.

I'm going to use the same method for each line...
 

Address

City

State

Zip

Step 2. - Creating the SQL

Now that we have all the fields separated, we can easily create a SQL Script off of this.
Simply concatinate them together.

(click for larger view)

Now that we have all of this done, we can copy, and paste the this code (or click and drag the bottom right corner) throughout the sheet.
This will create the fields and SQL for us.

Then, two lines down...

Which gives us...

Now, select the entire "C" COLUMN and copy it ...

Now move that to any column down the line from column "C", lets say, "E". Right click and click "Paste Special"

Then click "Values"

Step 3 - Extracting the SQL

Now all the actual text values will be placed into the columns. You can now select Column "E" and copy all of those values.

The reason we have to paste special is because otherwise we'd be copying the column caluculations, instead of a pretty SQL script, we'd get a row with no data, such as:

INSERT INTO dbo.Locations(LocationName, Address, City, State, Zip) VALUES ('', '', '', '', '')

By pasting special, we're telling Excel we want the VALUES of the cells, not the actual calculations.

Now that you've "pasted as special" with values, you can now copy column "E" and then paste that into your SQL Management Studio and run. For simplicity sake, here are a few rows from the actual Excel file that is attached to this post.

INSERT INTO dbo.Locations(LocationName, Address, City, State, Zip) VALUES ('1385 Metropolitan Ave (UCO)', '1385 Metropolitan Avenue', 'Bronx', 'NY', '10462')
INSERT INTO dbo.Locations(LocationName, Address, City, State, Zip) VALUES ('Whitestone', '138-11 20th Avenue', 'Whitestone', 'NY', '113562451')
INSERT INTO dbo.Locations(LocationName, Address, City, State, Zip) VALUES ('Target Flushing T-1150', '13505 20th Ave', 'Flushing', 'NY', '11356')
INSERT INTO dbo.Locations(LocationName, Address, City, State, Zip) VALUES ('LGA - Delta Flight Ctr-Delta Term', 'La Guardia International Airport', 'Flushing', NY', '11371')
INSERT INTO dbo.Locations(LocationName, Address, City, State, Zip) VALUES ('Flushing - Main Street', '41-02 Main Street', 'Flushing', 'NY', '11372')

Then execute your script and you're on your way.

Conclusion

When is it appropiate to use this method? IT DEPENDS. If you have 10 records, no. If you have 1000 records? Yes. Its all about productivity. Ask yourself, is it going to be faster for me to spend 5 minutes writing this in EXCEL or is it going to take me 5 minutes to import them manually? This is up to you to answer. Hopefully this helps you on your next project. :)

Download File

starbucks-locations-xls.zip (52.71 KB)

Full Visual Studio 2005 solution included at the bottom of this post.

Many organizations have multiple locations throughout the country. A lot of the time the organizations customers will want to know where the closet location of that store is. We’ve all become accustomed to going to a companies website and finding the store locator and locating the store that’s closest to us. This is a simple thing for a user to do. But building a store locator is far from simple and can become quite complex, real quick. I’ve put together a simple example that will enable you to add store locator functionality to your ASP.NET website in no time.

Let’s start by covering the basics….

Requirements for store locator functionality: 
   - Database to store your store locations
   - Each store location needs a Longitude / Latitude 
   - A method to retrieve Latitude / Longitude, also known as Geocoding
   - A Web Front end that allows users to search and view the results.

Screen Shots

Search Screen:

Results Screen (click for larger view):

The Database

The database is very simple. We have one table, two sprocs, and 3 functions. It is in the App_Data folder of this solution.

The Location table holds the store locations.

The GetNearbyLocations sproc gets the locations that are closest to the user through a mathematical calculation that is inside of the sproc.

The InsertLocation sproc inserts a new location into the Location table (we’ll get to that near the end).

The three functions: XAxis, YAxis, ZAxis are used in the distance calculation that is inside of the sproc.

Important Note: I’m not the genius that wrote this mathematical calculation. Therefore, I can’t take credit for it. The database table, functions and distance sproc are all modeled after this post on MSDN. I have altered quite a few things especially the GetNearbyLocations sproc where I actually calculate the Earths X,Y, and Z axis’ at runtime. 

The Geocoding – Getting the Latitude / Longitude

The latitude and longitude are needed to perform the distance calculation in the GetNearbyLocations stored procedure. The Latitude and Longitude of each store is stored in the Locations table (see the database image above) with the location record itself.

How to do we get the Latitude / Longitude?

That’s where the Google Maps API comes into place. Google Maps allows you to sign up for a free API key.

Important Note: If you are going to test this locally, you will need to get a API key that is associated with your local machine, e.g.: http://localhost:port/  where “port” is your port number, such as: localhost:4688, localhost:1637, etc).

Where put your Google Maps API Key
The key is located in two places (I know, its not best practice) but, it’s the only way I could get it to work without spending hours on figuring it out. It’s located in the web.Config and in the Default.aspx page.

Default.aspx page (click for larger view)

web.Config

In a previous post I had built a class that retrieved the latitude and longitude of an address. By using this class while a new location is being added, (AddNewStore.aspx), the Lat/Lon coordinates are then retrieved and saved.

protected void addNewStoreButton_Click(object sender, EventArgs e)
{
   string address = String.Format("{0}, {1}, {2} {3}", 
   addressTextBox.Text,
   cityTextBox.Text,
   stateTextBox.Text,
   zipTextBox.Text);
   Coordinate addressCoordinates = Geocode.GetCoordinates(address);
   ... // Other code ommitted
}

The Web Front End

The web front end is simple (errrrrrrr… kind of). The application requires 2 things:
1. An address
2. A distance to search within.

The user enters this information, it’s passed to the database, the calculation is performed and the records that match the distance compared to the address are returned. Simple, right?

Now its time for the “somewhat” confusing part: the task of integrating the Google map with clickable markers. Google doesn’t offer markers that are numbered (from what I could find). So I found some on the net and included them in this project. Which leads me to the next point…

A maximum of 100 results will be returned.

This is set within the stored procedure. The reason we only return 100 records is because we only have 100 icon images (1 through 100). Plus, placing more markers on the page would have a negative impact on the memory and processor on the user’s machine. Furthermore, who’s really going to look at 100 results? The user is probably only looking for the closest couple of stores. Not the top 100.

Google Marker Placement

In order to place markers on the screen we needed lat/lon points to pass to Google’s API. This poses a problem because our data is in the DB, but Google Maps API needs it on the client to process it. Therefore, while on the server, while we have the data in a dataset (below) ...

/// <summary>
/// Gets the store locations and the coordinate for the "from" area (the address the user entered).
/// </summary>
/// <param name="coordinate">A coordinate.</param>
/// <param name="data">A LocationsData dataset.</param>
private LocationsData GetLocationData(Coordinate coordinate)
{
SqlConnection connection = new SqlConnection(ConfigurationManager.ConnectionStrings["StoresDb"].ConnectionString);
connection.Open();

string SQL = "GetNearbyLocations";
SqlCommand command = new SqlCommand(SQL, connection);
command.Parameters.Add(new SqlParameter("@CenterLatitude", coordinate.Latitude));
command.Parameters.Add(new SqlParameter("@CenterLongitude", coordinate.Longitude));
command.Parameters.Add(new SqlParameter("@SearchDistance", distanceDropDown.SelectedValue));
command.Parameters.Add(new SqlParameter("@EarthRadius", 3961)); // In Miles
command.CommandType = CommandType.StoredProcedure;

SqlDataAdapter da = new SqlDataAdapter(command);
LocationsData data = new LocationsData();
da.Fill(data.Location);
return data;
}

... we have to create a string representation of a JSON object array that holds all this info, this is done in the GetJSONLocations method. Please see the method for more documentation on how that variable is created.

Info that the JSON object holds: 
   - Location Name
   - Address 
   - Url Encoded Address
   - Latitude
   - Longitude

This info was written to the screen using the RegisterClientScriptBlock.

Using the same method of as before we create another JSON variable (homeSpatialInfo) that holds the “home info”. This info is the address that the user typed in.

This information is used for the popup marker, as well as the link that is provided within the popup marker. Clicking on the “directions” link (within the popup marker) will take you to Google’s map site with the “to” and “from” address parameters already populated, therefore giving you a directional map from the location you typed in, to the location that you clicked on the map. Smooth, eh?

(click for larger view)

Adding a new Location

To add a new location, fire up the AddNewStore.aspx page and type in the required info (in this case I’m requiring all fields, but this can be changed to fit your own needs) and submit. If the address cannot be found, (Google Maps API returns “0” for both the Lat/Lon coordinates that it cannot find) it will inform you. Otherwise the address will be added and then upon your next search, it will be returned in the results (if the location is within the distance specified).

Database Note: The locations that are in the Stores.mdf that come with this solution are 200 Starbucks stores that are within the New York Area. This is not a comprehensive list, this is just a list I happened to have for testing purposes.

And that’s all folks!

Download and enjoy. :)

StoreLocator.zip (521.92 KB)