Revived for SharePoint !

Getting Key Token of a Signed Assembly

2007-01-08T17:34:00.000-08:00

To obtain the public key token from strong-named and signed assembly, I normally use the Reflector tool to view it. Lately, I discovered that you can get the key token from the Visual Studio.NET IDE tool, too. Here are the steps

1. Open Visual Studio 2005, click Tools -> External Tools...
2. Click Add and enter the following inputs

Title: Get Public Key
Command: C:\Program Files\Microsoft Visual Studio 8\SDK\v2.0\Bin\sn.exe
Arguments: -Tp "$(TargetPath)"

* Uncheck all options, except Use Output window\

After clicking OK, go to the "Get Public Key" in the Tools Menu.

And you would get the public key as well as its token in the OUTPUT WINDOW.

ASPMENU Source Code is now available !

2006-12-11T17:00:00.000-08:00

The underivable AspMenu class that shipped with WSS 3.0 is now unsealed ! The class source code has been released lately to allow developers to extend its functionality. Download the MOSSMenu source code here and you can treat the class as your base class for your own custom menu.

Check Customizing the WSS 3.0/MOSS 2007 Menu Control -- MossMenu source code released for more information.

Inflexibility of WSS 3.0 Customization

2006-12-11T07:16:00.000-08:00

Recently I came across the page customization in WSS 3.0 (Creating new custom page that using application.master under _layouts folder), and I found out that the master page is likely not edited for the company branding purpose because the master page is used and shared by all sites in the WSS v3. Any changes upon the application.master would eventually applied to the rest of the sites too.

This would be a PAIN for developers where easier customization feature is not ready for them. The very easy and straightforward solutions to the matter that I can think of are:

1. Clone the application.master and rename it to your own name. Change and apply the custom master page separately.

2. Develop a custom web control that check current SPSite or SPWeb to change master page accordingly.

Perhaps, there are more elegant solutions that can resolve this problem.

Installing Custom Group and Actions in WSS/MOSS Features

2006-12-05T07:06:00.000-08:00

Another customization in WSS/MOSS! You can install your own custom feature other than out-of-box features that offered WSS/MOSS. You can add new actions or menus in any location in WSS/MOSS like Site Settings, Site Actions, Item Context Menu, and etc in Standard Menu.

By default, there have 5 groups - Look and Feel, Users and Permissions, Gallaries, Site Administration, Site Collection Administration.

To install new custom feature, create a new folder in C:\Program Files\Common Files\Microsoft Shared\web server extensions\12\TEMPLATE\FEATURES. I name the folder as MyCustomGroupAndAction. Then we need to create an XML called feature.xml and another xml file that contains the feature definition. (featuredef.xml)

Feature.xml has several elements - <Feature>, <ActivationDependencies>, <ElementManifests> and <Properties> but the important ones are :

1. <Feature> (the root element) - Defines feature ID, scope (site, collection, etc), description and title.

2. <ElementManifests> - defines the feature definition, which links to the xml file that contains the definition.

In feature.xml


<?xml version="1.0" encoding="us-ascii"?>
<Feature  Id="F863C1CF-897D-49c6-8600-7E6F3FC9D653" 
                Title="My New Feature" 
                Description="This is my new feature"
                Scope="Web" 
                xmlns="http://schemas.microsoft.com/sharepoint/">
    <ElementManifests>
        <ElementManifest Location="featuredef.xml" />
    </ElementManifests>
</Feature>

The ID of the feature is unique GUID. You need to generate new GUID for each new feature being created. To generate the GUID, you can go to Visual Studio -> Tools Menu -> Create GUID or launch the tool at C:\Program Files\Microsoft Visual Studio 8\Common7\Tools\guidgen.exe.

The location attribute of <ElementManifests> specifies the location of feature definiton XML file.

In featuredef.xml


<?xml version="1.0" encoding="us-ascii"?>
<Elements xmlns="http://schemas.microsoft.com/sharepoint/">
  <CustomActionGroup
     Id="MyCustomGroup"
     Location="Microsoft.SharePoint.SiteSettings"
     Title="My Custom Group" />
  
 <CustomAction Id="MyCustomFeature"
  GroupId="MyCustomGroup"
  Location="Microsoft.SharePoint.SiteSettings"
  Title="New Custom Feature"
  Sequence="50">
  <UrlAction Url="_layouts/layoutdesign.aspx"/>
 </CustomAction>
</Elements>

The <CustomActionGroup> specifies the definition of new group to be created. Location attribute specifies where the custom group will reside. In this example, "Microsoft.SharePoint.SiteSettings" indicates that the new group will be in the Site Settings menu of WSS sites. Check Location and Group ID in MSDN for more information. Text in the Title attribute is the name of the group.

<CustomAction> is where the custom feature definition set. The location attribute MUST BE same as the location being specified in the group that the custom feature points to (in GroupID attribute). <UrlAction> element merely specifies URL of the custom application page.

It is about to install the feature ! Go to the command prompt, type


C:\Program Files\Common Files\Microsoft Shared\web server extensions\12\BIN>stsadm -o installfeature -name MyCustomGroupAndActions

The feature has just been installed. However, you need to activate the feature before using it. To activate it, you can type


stsadm -o activatefeature -filename MyFeature/feature.xml -url <server url>

or

go to http://<server name>/_layouts/manageFeatures.aspx, and you will see this

Click "Activate" button to activate the feature. Go to "Site Settings" and you will see this

You have completed installation of new custom group and feature !!

Converting String GUID to System.Guid

2006-12-04T06:56:00.000-08:00

Simple matter. I was trying to convert the GUID in string to type of System.Guid and the code below apparently did not work for me.


System.Guid guid = (System.Guid)strGUIDInString;

I was given invalid specified cast error message and I noticed that direct casting wasn't the way. I ended up with


System.Guid guid = new System.Guid(strGUIDInString);

This was the final solution I could think of, so far !

MOSS 2007 and WSS3 Functional Architecture Diagram

2006-11-27T07:38:00.000-08:00

Here is the colourful and comprehennsive diagram illustrating the functional architecture of the most latest of Microsoft Office Sharepoint Server 2007 and Windows Sharepoint Services 3.0.

Release of WSS RTW and MOSS Standard and Enterprise Edition

2006-11-25T04:58:00.000-08:00

YOu can download the Release-To-Web version of WSS (Windows SharePoint Services) and Microsoft Office SharePoint Server 2007 Evaluation version at

1. Microsoft Office SharePoint Server 2007 x86 English Evaluation
2. Windows SharePoint Services 3.0

CD Keys available for standard and enterprise edition are

SharePoint Server Standard Trial: XJMKW-8T7PR-76XT6-RTC8G-VVWCQ
SharePoint Server Enterprise Trial: F2JBW-4PDJC-HKXTJ-YCKRP-T2J9D

For more resources about installation, upgrade, versioning, check this link, from Microsoft SharePoint Products and Technologies Team Blog

Efficient String Concatenation in JS

2006-11-23T07:46:00.000-08:00

In .NET, we use StringBuilder if we need to concatenate strings for better string manipulation performance. For instance,


StringBuilder builder = new StringBuilder();
builder.append("This");
builder.append("is");
builder.append("really");
builder.append("good");
builder.append("for");
builder.append("performance");
  
string str = builder.ToString();

In JS, we can also do the same thing for better memory allocation in string concatenation too.


var builder = new Array();
builder.push("This");
builder.push("is");
builder.push("really");
builder.push("good");
builder.push("for");
builder.push("performance");
 
var str = builder.join("");

This method will perform faster if the number of string concatenation is huge. THe normal += operator will win, otherwise.

Check Efficient JavaScript for more information on optimizing JS code

Invoking Executeable File from ASP.NET Web Services

2006-11-21T17:24:00.000-08:00

Below is the code snippet on how to invoke the executable file (exe) from the cmd.exew being executed.


// Execute the Command Prompt
 ProcessStartInfo psi = new ProcessStartInfo("cmd.exe");
 
// Indicating the input read from the process's standard input
psi.RedirectStandardInput = true;
psi.UseShellExecute = false;
 
// Do not show the GUI of the running app
psi.CreateNoWindow = true;
 
Process process = null;
try
{
   // Execute the cmd.exe
   process = Process.Start(psi);
 
   StreamWriter writer = process.StandardInput;
             
   // Write the command to the command prompt
   writer.WriteLine("notepad");
 
}
catch (Exception ex)
{
  // do something here
}
finally
{
    if (process != null)
    {
                process.Close();
    }
}

The code will simply run the notepad from the cmd.exe. You can just call "notepad" from at line the of ProcessStartInfo psi = new ProcessStartInfo("notepad");, but I am showing how are you going to write the command to the cmd using the StandardInput of Process object.

Similarly, you can call batch file from the cmd to execute command in the batch file too, or store those static command in the resource files (resx) inside the app_GlobalResources folder and get from the key by


Resources.<Resource file name>.<key name>

MSDN Magazine - HTML Help Downloads

2006-11-21T07:39:00.000-08:00

If you want to read MSDN Magazine contents wherever you are, you can download the dozens of MSDN Magazine in portable chm format at http://msdn.microsoft.com/msdnmag/htmlhelp.aspx

Use of ResolveClientUrl() method in getting browser-acceptable relative URL

2006-11-21T05:59:00.000-08:00

In .NET, we always use tilde (~) to refer to root virtual directory of the web application. For instance,


<asp:Image runat="server" ImageUrl="~/images/mypic.gif" />

However, in order to generate a URL that can be used by client-side browser, the URL is actually resolved in the AddAttributesToRender() of the Image server control.


writer.AddAttribute(HtmlTextWriterAttribute.Src, ResolveUrl(ImageUrl));

That is the reason why you can use "~" in the ImageUrl property. But I afraid this might be the case if you do the same thing upon the normal HTML elements or HtmlGenericControl, such as :


<img src="~/images/mypic.gif" alt="" />

because the browser does not recognise what the "~" symbol stands for. In order to workaround with that, this is where the ResolveClientUrl() method comes into place.


<img src='<%= ResolveClientUrl("~/images/mypic.gif") %>' />

Apart from using it in HTML, it can also be used in code-behind that involves in getting the relative path for the URL. For instance, ResolveClientUrl() method is handy when we are trying to register the javascript reference in code-behind,


this.Page.ClientScript.RegisterClientScriptInclude(this.GetType(),"MyScript", this.ResolveClientUrl("~/images/myscript.js"));

Not only that, you will always need this method in your custom server control development.

Debugging JS in VS 2005 using Script Explorer

2006-11-02T07:33:00.000-08:00

Breakpoint is one of the useful features available in Visual Studio 2003/2005 to allow developers to trace the current value of variable at the location where the debugger is currently running at. Nevertheless, you can set any breakpoint in the code-behind of the aspx page, but not in declaractive page. No particular client-side javascript function codes can simply be set as breakpoint in Visual Studio.

I rarely debug the client-side JS codes, or I can say NEVER! I have tried to debug it using Script Explorer today and it impressed me. To debug the client-side JavaScript functions, you can use the only-visible-in-debugging-environment Script Explorer in Visual Studio 2003/2005. There are few steps need to be followed:

1. Enable Client-side Script Debugging

IE -> Tools Menu -> Options Menu -> Advanced Tab -> Untick " Disable script debugging check box" in Browsing category.

2. In VS environment, write a simple JS function in external functions.js


function MyLoad(name)
{
   var message = "Welcome, " + name;
   alert(message);            
}

Then, invoke the function in body's load() event in ASPX page.


window.onload = MyLoad("Alvin");

3. Press F5 to debug the page. In debug mode, go to Debug Menu->Windows-> Script Explorer (this option only appears in debug mode). The window of Script Explorer will be shown.

4. Once you debugging the page, you will see list of JS files/resources shown in Script Explorer in tree view.

Double click the "functions.js" and set the breakpoint at particular line. The debugger will stop at that line when it passing through it.

Additionals
If your javascript functions are in the aspx page, you can add breakpoint by

Debug Menu -> "Add Breakpoint" -> "Break At Function" -> type your method name there and choose right language.

Read more in MSDN
1. How to: Debug a Client-Side Script from Microsoft Internet Explorer
2. How to: Enable Client-Side Script Debugging
3. How to: Set a Function Breakpoint

I need a break....

Accessing GridView's DataControlFieldCell Programmatically

2006-10-31T04:21:00.000-08:00

How would you access the auto-generated control in the HyperLinkField (HyperLink control) and ButtonField (LinkButton or Button control) in RowDataBound event ? You will find out that FindControl() method would no longer be working since there is no concrete ID is assigned to the control, and you are not intelligent enough to know what ID will be assigned automatically by the compiler. There are two workarounds, in which the second one is new in ASP.NET 2.0.

Assuming that the HyperLinkField is at 4th column.

First Workaround


protected void GridView1_RowDataBound(object sender, 
GridViewRowEventArgs e) 
{ 
    if (e.Row.RowType == DataControlRowType.DataRow) 
    { 
         HyperLink hl = e.Row.Cells[3].Controls[0] as HyperLink;
         hl.Text = "Changing Text here";
    } 
}

This workaround is pretty straightward. Cells[3] refers to 4th column (index starting from 0) ; Controls[0] means the first control in that column, which is HyperLink control.

The first workaround is apparently classic way that we used in DataGrid control. But here is the another workaround, which works in different way !

Second Workaround


protected void GridView1_RowDataBound(object sender, 
GridViewRowEventArgs e) 
    { 
        if (e.Row.RowType == DataControlRowType.DataRow) 
        { 
            DataControlFieldCell dField = e.Row.Cells[3] as 
DataControlFieldCell; 
            HyperLinkField hlf = dField.ContainingField as HyperLinkField; 
            hlf.Text = "Changing Text here"; 
            hlf.DataNavigateUrlFormatString = "nextpage.aspx?id={0}";            
        } 
    }

As you can see, I am accessing the HyperLinkField itself, and I am able to assign the DataNavigateUrlFormatString property in the RowDataBound event too. Of course, if you would like to access the control in that field, it can be done by


HyperLink hl = dField.Controls[0] as HyperLink;

Hope this helps...

SQL Server Connection Timeout !

2006-10-31T02:05:00.000-08:00

Frustrated ! I was given an error message of "Timeout expired. The timeout period elapsed prior to completion of the operation or the server is not responding." when I was trying to connect to SQL Server 2000 with VS 2005. The connection string was correct and I tried several connection strings :


1. Server=ALVIN;Initial Catalog=pubs;UID=sa;
2. Server=ALVIN;uid=sa; database=pubs";
3. packet size=4096;user id=sa;data source=ALVIN;persist security info=False;database=pubs

(I used sa account for simplication, don't use it in your production site)

But none of them worked for me. I tried it in VS 2003 and unbelievably it worked! (Don't say there is any discrimination) What the heck.. Without any patience left, I saught the answer from Google and I got this link at http://msdn2.microsoft.com/en-gb/library/ms190181.aspx. It explains several causes of the problem, but one of them seemingly is the cause of my problem, which is :

"Database Engine is not listening on port 1433 because it has been changed, or because it is not the default instance, and the SQL Server Browser service is not running."

I tried to go to Client Network Utility and Server Network Utility to change my port number of TCP/IP to 1433 AND reorder it to be the first before the "Named Pipes".

After restarting the SQL Server, it eventually went fine for me. :)

New Features of my Blog

2006-10-30T07:55:00.000-08:00

I have implemented my blogger with new Blogger Beta, and it is pretty amazing. The template in HTML is different from the classic template style. There are some new features added in this blog, such as post count in the right-hand-side sidebar, new widgets, new navigation bar. Other features such as blogs searching will be added soon. Although it is not completely nice, it nears to perfection.

For those users who use this blogger, you can visit http://beta.blogger.com/ for the version switching, or visit these sites for helps.

1. http://betabloggerfordummies.blogspot.com/index.html
2. http://groups.google.com/group/blogger-help
3. http://betabloggerfordummies.blogspot.com/

Still figuring out its features and specialities....

Changes on my ValidTextBox Server Control

2006-10-28T22:02:00.000-07:00

I have made some changes to my ValidTextBox server control. The changes are:

1. Supports to display error image instead of text error message when the control is invalid.
2. Supports Tooltips for error image.
3. Fixed bug of email address expression, which allows symbols.

Screenshots:

Please check http://mind.com.my/forums/1018/ShowThread.aspx#1018 for more details...

Sample Programming Interview Questions

2006-10-28T06:08:00.000-07:00

I have found there are a set of interview books, relating to Programming - ".NET Interview Questions", "Java Interview Questions", "SQL Server Interview Questions" and so on. These questions are commonly asked by companies during their interview/recruitment process. You may take a look on sample questions at http://www.questpond.com/ , which has a lot of sample questions to be downloaded in zip format.

Some questions are tough!

[ASP.NET] Building Tag Cloud in ASP.NET

2006-10-28T06:04:00.000-07:00

What is Tag Cloud. A tag cloud is a way to display a weighted list such that the weight of each item is reflected by the size of the item's text. Tag clouds provide a quick way for one to eyeball a list and ascertain what items are more prevalent.

Scott Mitchell has just written an article of Creating a Tag Cloud in ASP.NET. That is very cool especially you are dealing with blogging or tagging-related sites. I have tested it, and that's nothing but COOL !

[ASP.NET] Basic of Custom Server Control Development Part II - ParseChildren(), PersistChildren() and PersistenceMode()

2006-10-25T09:01:00.000-07:00

Let us boils down to the three attributes that important in implementing simple properties and subproperties in the custom control development - ParseChildrenAttribute, PersistChildrenAttribute and PersistenceModeAttribute.

Default and Inner Property Persistence
Default persistence is in constrast to Inner persistence, but they are identically same for its functionality. In our control development, we may have complex properties (class). To persist them, we can use hyphenated syntax (Default Property Persistence) or nesting them within your control tags (Inner Property Persistence). Example of Inner Property Persistence,


<asp:GridView ID="GridView1" runat="server">
     <HeaderStyle BorderWidth="12" />
</asp:GridView>

In the example given, the BorderWidth is the subproperties of the HeaderStyle property of GridView. To enable this inner property persistence, you need to set the ParseChildren(true) and PersistChildren(false) in your control class.


[ParseChildren(true)]
[PersistChildren(false)]
public class MyWebControl : WebControl
{ 
  ... 
}

1. ParseChildrenAttribute
This attribute tells the parser to parse the parse the contents within the control tag as properties but not child controls.

To enable the inner default property persistence, you have to specify the inner default property to the second parameter of ParseChildrenAttribute.


[ParseChildren(true, "<DefaultPropertyName>")]
public class MyControl : WebControl 
{ 
  ... 
}

2. PersistChildrenAttribute
On the other hand, this attribute tells the designer whether the child controls of the server control should be persisted as inner child controls. Setting to false indicates inner content corresponds to properties but not to child controls.

* Therefore, the ParseChildrenAttribute and the PersistChildrenAttribute attributes use opposing conventions for the semantics of their argument. Check more difference of these two attributes at Justin Lovell's Blog - The ParseChildren and PersistChildren Attributes

3. PersistenceModeAttribute
This attribute specifies how the property of server control is persisted declaractively in ASP.NET pages. It has 3 enumerations :

EncodedInnerDefaultProperty - Specifies that the property persists as the only inner text of the ASP.NET server control. The property value is HTML encoded. Only a string can be given this designation.

InnerDefaultProperty - Specifies that the property persists in the ASP.NET server control as inner text. Also indicates that this property is defined as the element's default property. Only one property can be designated the default property.

InnerProperty - Specifies that the property persists in the ASP.NET server control as a nested tag. This is commonly used for complex objects, those that have persistable properties of their own.

Picked from http://msdn2.microsoft.com/en-us/library/system.web.ui.persistencemode.aspx


[PersistenceMode(PersistenceMode.InnerProperty)]
public virtual TableItemStyle HeaderStyle { ... }

This indicates that the designer to persist the property as inner property. By default, it is persisted by using hyphenated syntax in control tag. To persist inner default property, you need to have


[ParseChildren(true, "<DefaultPropertyName>")]
public class MyControl : WebControl 
{ 
  ... 
}

and


[PersistenceMode(PersistenceMode.InnerDefaultProperty)]
public virtual ListItemCollection Items { ... }

These three attributes are confusing, but are not undifferentiable.

My ValidTextBox is OUT !

2006-10-23T10:11:00.000-07:00

After spending 3 days of custom control development, I have successfully developed the beta-period ValidTextBox server control. It basically integrates validator controls (RegularExpressionValidator and RequiredFieldValidator) and normal textbox control to reduce the developers' time and efforts. Here are some screenshots of the ValidTextBox server control.

Check more about the ValidTextBox server control at MIND Forum here

[ASP.NET] All About Dynamic Server Controls

2006-10-21T20:03:00.000-07:00

If you want have a more-in-depth reading on the dynamic server controls in ASP.NET, you not gonna to miss these :

1. Working with Dynamically-Created Controls [4GuysFromRolla]
2. Dynamic Controls in ASP.NET [4GuysFromRolla]
3. Dynamic Web Controls, Postbacks, and View State
4. TRULY Understanding Dynamic Controls (Part 1) - Dynamic vs Static
5. TRULY Understanding Dynamic Controls (Part 2) - Dynamic Controls
6. TRULY Understanding Dynamic Controls (Part 3) - Adding Controls to Control Tree

[ASP.NET 2.0] Post-Cache Substitution

2006-10-19T19:24:00.000-07:00

Post-Cache Substitution, a new feature that is not so eye-catching compared to the other new features introduced in ASP.NET 2.0 like Master Pages, Personalization, DataSource Control and etc. Basically, Post-Cache Substitution allows developers dynamically update the contents of output-cached portions (eg. header, footer). When a header user control (header.ascx) is given the following directive


<%@ OutputCache Duration=3600 VaryByParam=none %>

It tells the page to be partially cached for an hour (3600 seconds). If there is current date and time showing in the header section, it is unlikely that the time will be updated for every second since it is cached for an hour long.

There are common 2 ways of implementation of Post-Cache Substitution solves this problem for you - Use of Substitution Control and Substitution API.

1. Substitution Control
This control tells the page not to cache the section for its dynamic contents. Every contents in the control would not be cached.

1. To use this control, drag the Substitution Control from VS 2005's ToolBox and drop it onto your WebForm.

2. The control has a MethodName property, which is assigned to a static method. In the code-behind, define a static method that accepting a page context as parameter.


public static string GetCurrentDateTime(HttpContext context)
{
      return DateTime.Now.ToLongTimeString ();
}

3. Back to Design View. In the control's property window, assign GetCurrentDateTime to MethodName property. This ascertains that subsequent requests to the page call the method again to generate dynamic content.

2. Substitution API
This implementation uses the in-lline code, by calling the Response.WriteSubstitution() method that accepting name of method as parameter. Thus, in your ASPX HTML view, add this


<% Response.WriteSubstitution(new HttpResponseSubstitutionCallback(GetCurrentTime)); %>

It would change client-side cacheability to server cacheability, which the dynamic contents are not cached on clients.

[ASP.NET] Specifying DefaultButton in MasterPage

2006-10-18T20:10:00.000-07:00

In normal ASP.NET 2.0, we can easily assign the DefaultButton property in the


<form runat="server" DefaultButton="btnSearch">

But what if we would like to make two different buttons from two pages that inheriting the MasterPage to be DefaultButton? It might not be working using the way above because MasterPage because it is publicly shared by its content pages. You, however, have to programmatically assign the DefaultButton in each of the content page.

In content Page,


this.Form.DefaultButton = "MyDefaultButton";

This is not the way of assigning the ID of button to DefaultButton property. Use this way,


this.Form.DefaultButton = MyDefaultButton.UniqueID

Cheeers

[ASP.NET] Redirecting Page after Emiting JS in Code-behind

2006-10-18T19:36:00.000-07:00

To emit the JavaScript in code-behind in ASP.NET, we use RegisterStartupScript() or RegisterClientScriptBlock() method. But when we would like to redirect user to a specified page after the JS pop-up, we probably would use this (written in VS.NET 2003, use ClientManager for VS 2005)


string strScriptName = "AlertScript";
string strMessage = "There is an error";
 
string strScript += strMessage + "')<"
strScript += "/"
strScript += "script>"

if(!Page.IsStartupScriptRegistered(strScriptName))
     objPage.RegisterStartupScript(strScriptName, strScript)
 
Response.Redirect("error.aspx");

If you run the code, you will notice that you will be redirected to the error.aspx WITHOUT being prompted the error message dialog box of "This is an error" because Response.Redirect() forces the browser not to render the HTML (sending a header to client to cause client to redirect to specified URL).

To overcome this problem, replace the Response.Redirect("error.aspx"); with


Response.AppendHeader("refresh","0;Url='error.aspx'");

Instead of redirecting the page, the browser would simply refresh the page within 0 second and "redirecting" to the error.aspx, as specified in the Url parameter.

[ASP.NET] Basic of Custom Server Control Development Part I

2006-10-17T06:44:00.000-07:00

Developing custom server control in ASP.NET is fun! You are flexible to develop it either inheriting the existing ASP.NET server controls like Button, HyperLink, BoundField or create you own control by inheriting Control/WebControl classes from System.Web.UI namespace. There are dozens of cool and fabulous custom server controls developed by third-parties like Metabuilder or ExcentricWorld. I like to develop the custom server control because the tailored controls can specifically accomplish some jobs that normal shipped-in ASP.NET server controls can't. It serves the purpose on your needs.

There are basic concepts/rules in custom server control development in ASP.NET like :

1. Derive your control class from System.Web.UI.Control if your control is dealing with non-visual elements or merely render as HTML output. For instance, META tag. Otherwise, your class should inherit from System.Web.UI.WebControls class if your control is built to have visual appearance to the client.

2. Do not override the Render() method of in your control if it is inheriting from System.Web.UI.WebControls class because the method is responsible in rendering the outer tag and control styling to the output stream. Override its RenderContent() method instead!


protected override void RenderContent(HtmlTextWriter writer)
{
  writer.Write("Test");
}

3. When your control inherting the System.Web.UI.WebControls, its outer tag is <span> by default. You can get rid of this and render any outer tags you want by overriding its HtmlTextWriterTag.TagKey or TagName (non-standard HTML tags).


protected override HtmlTextWriterTag TagKey
{
   get {  return HtmlTextWriterTag.Input; }
}


public Button() : base(HtmlTextWriterTag.Input)
{  }

4. Check whether the control is in the <form runat="server"> by


if (Page != null)
    Page.VerifyRenderingInServerForm(this);

in either AddAttributesToRender() or Render()/RenderContent() method.

5. By default, the tag prefix of your custom server control would be auto-generated. For instance, when you drag and drop your control, the tag would be something like


<cc1:CustomMeta ID="CustomMetaTag1" runat="server" />

As you see, the "cc1" gives no meaning. You can add your own tag prefix by adding TagPrefixAttribute in the assemblyinfo.cs file.


using System.Web.UI;
[assembly:TagPrefix("SimpleControl", "Alvin")]

Now, it would become like


<Alvin:CustomMeta ID="CustomMetaTag1" runat="server" />

6. Take note of rendering methods of System.Web.UI.Control class


public void RenderControl(HtmlTextWriter writer) {
    if (Visible) {
        Render(writer);
    }
}

protected virtual void Render(HtmlTextWriter writer) {
    RenderChildren(writer);
}

protected virtual void RenderChildren(HtmlTextWriter writer) {
    foreach (Control c in Controls) {
        c.RenderControl(writer);
    }
}

It is wise to call the base class's methods if you intend to override the method like RenderChildren method to ensure its child controls are recursively rendered.


protected override void RenderChildren(HtmlTextWriter writer)
{
   base.RenderChildren(writer);
  
   // do your stuff
}

7. To restrict the control to have NO CHILD CONTROLS, you need to return EmptyControlCollection(this) to the overidden CreateControlCollection() method.


protected override ControlCollection CreateControlCollection() {
            return new EmptyControlCollection(this);
}

By doing that, no child controls/tags are allowed.


// Allow
<Alvin:CustomMeta ID="CustomMetaTag1" runat="server" />

// Not allow !
<Alvin:CustomMeta ID="CustomMetaTag1" runat="server"></Alvin:CustomMeta>

Go for Google or Microsoft?

2006-10-14T23:15:00.000-07:00

Cliff, who has been offered by Google and Microsoft is in trouble! He is in hestitation whether to work for Google or Microsoft, He posted a dicussion topic of Microsoft or Google? to ask Slashdot-ers about both companies' working environments and their thought about the companies. It is interesting to see the commenters posted their opinions about Google and Microsoft's working environment, working hours, management styles, pressure, corporate culture, benefits and so on, respectively.

Google or Microsoft? Hard to choose even to me.

Tips and Tricks on cloning data-bound controls

2006-10-14T05:45:00.000-07:00

Here are some tips and tricks on cloning the Data-Bound controls.

Cloning DataGrid/GridView
It is common mistake when you would like to copy the assigned DataGrid's DataSource to another DataGrid's datasource like


DataGrid2.DataSource = (DataSet)DataGrid1.DataSource;

You would simply get the NullReference exception. It seems work but it doesn't. A better way is store the needed DataSet to Cache or Session and reuse it for subsequent data binding.


Object objDS = Cache["MyDS"];
if(objDS != null)
{
   DataSet ds = (DataSet)objDS;
   DataGrid1.DataSource = ds;
   DataGrid2.DataSource = ds;
}
else
{
   DataSet ds = new DataSet();
   // open conn, fill the DS;
   DataGrid1.DataSource = ds;
   DataGrid2.DataSource = ds;
 
   Cache["MyDS"] = ds;
}

THERE is another way cloning the content of Grid by rendering its HTML contents to HtmlTextWriter. Try to put this code in the Button's Click event.


StringBuilder strBuilder  = new StringBuilder();
StringWriter writer = new StringWriter();
 
HtmlTextWriter textWriter = new HtmlTextWriter(writer);
strBuilder = writer.GetStringBuilder();
dgrdAuthors.RenderControl(textWriter);
 
string strContent = strBuilder.ToString();
Response.Write(strContent);

Once you click the button, the cloned Grid appears. This code is particularly useful especially you would like to send the content of Grid over email.


.....
objMail.Body = strContent
objMail.BodyFormat = MailFormat.Html
....

Cloning DropDownList/CheckBoxes/RadioButtonList
How do you make two DropDownLists, which bound to a datasource to have identical items? The very basic method is use of DataSet like


DropDownList1.DataSource = ds;
DropDownList1.DataTextField = "MyText";
DropDownList1.DataValueField = "MyValue";
DropDownList1.DataBind();
 
DropDownList2.DataSource = ds;
DropDownList2.DataTextField = "MyText";
DropDownList2.DataValueField = "MyValue";
DropDownList2.DataBind();

It works fine. No complain. What if you use forward-only IDataReader for performance wise? You may need to have two IDataReaders for each DDL's data-binding. Here is another trick where only one DataReader is needed.


reader = cmd.ExecuteReader(CommandBehavior.CloseConnection);
DropDownList1.DataSource = reader;
DropDownList1.DataTextField = "MyText";
DropDownList1.DataValueField = "MyValue";
DropDownList1.DataBind();
reader.Close();
 
// I am cloning the ListItemCollections of DDL1 to DDL2
DropDownList2.DataSource = DropDownList1.Items;
DropDownList2.DataTextField = "Text";
DropDownList2.DataValueField = "Value";
DropDownList2.DataBind();

Much simpler, much faster, much neat !

Browser DOM Supports and Compatibility Issues

2006-10-13T19:28:00.000-07:00

I occasionally get some earful from other people how hard to develop an AJAX application. Well, it isn't that hard if you use on-ready AJAX development tools such as Atlas. Nevertheless, you may be cracking your head on writing remote scripting from ground up because there are few issues need to be taken care.

Other than the checking the types of the browser, you also need to concern about the DOM supports the particular browser can support. For instance, the innerText property of < span > is not supported in FireFox, you have to use textContent property or universal supported innerHTML property. Here are the listing of sites that you may need to check the DOM that the layout engines in each browser can support.

Links
1. Web browser HTML support
http://www.webdevout.net/browser_support_html.php

2. Web browser DOM support
http://www.webdevout.net/browser_support_dom.php

3. DOM Support in Mozilla
http://www.mozilla.org/docs/dom/mozilla/

4. Comparison of layout engines (DOM)
http://en.wikipedia.org/wiki/Comparison_of_layout_engines_%28DOM%29

5. Common Differences Among IE and Mozilla
http://www.htmlgoodies.com/primers/jsp/article.php/3624446

[ASP.NET] Serializing/Deserializing Passing Objects Across Pages with LosFormatter

2006-10-13T11:37:00.000-07:00

On our ordinary basis, we pass the values or object values across two pages using Session, Form Posting, PreviousPage reference in ASP.NET. But however, we could not able to append the object to the query string (unless you use the Session state instead). There is another alternative way - passing the serialized the object values into string presentation in query string.

How ?? First of all, are you familiar with this ?


< input type="hidden" name="__VIEWSTATE" value="dDwxMzIwMTI1MjYwO3Q8O2w8aTwxPjs+O2w8dDw7bDx
pPDE+Oz47bDx0PHA8cDxsPE5hdmlnYXRlVXJsOz47b
DxXZWJmb3JtMi5hc3B4P3ZhbHVlPWJEeDBhR2x6T3o
0PTs+Pjs+Ozs+Oz4+Oz4+Oz5RXn3JAFz8V0ozE1lmu
5B1K2D7xw==" / >

If you do, that's good. It is ViewState, the hidden field that stores control state, which the state values (eg. object values, primitive type values) are serialized and deserialized into base-64 string using LosFormatter. Similarly, this approach can be used to serialize the object and pass it to the next page in query string.

The following example shows you how to pass the serialized ArrayList to the next page using LosFormatter.

Code
Firstpage.aspx


System.IO;
System.Collections;

private void Page_Load(Object sender, EventArgs e)
{
   ArrayList ary = new ArrayList();
   ary.Add("this");

   System.Web.UI.LosFormatter formatter = new System.Web.UI.LosFormatter();
   StringWriter writer = new StringWriter();

   formatter.Serialize(writer,ary);
   HyperLink1.NavigateUrl = "secondpage.aspx?value=" + writer.ToString();
}

Check the URL in the HyperLink control in Firstpage.aspx, it would be something like


http://localhost/WebTest/secondpage.aspx?value=bDx0aGlzOz4=

The ArrayList object has just been serialized ! How to deserialize the object back?

Secondpage.aspx


System.IO;
System.Collections;

private void Page_Load(Object sender, EventArgs e)
{
   System.Web.UI.LosFormatter formatter = new System.Web.UI.LosFormatter();
   ArrayList aryL = (ArrayList)formatter.Deserialize(Request.QueryString["value"]);

   Response.Write(aryL[0]);
}

* Be noted, you need to explicitly specify the System.Web.UI.LosFormatter in the object instantiation. You will get the undefined Serialize()/Deserialize() method, otherwise. Watch out the number of characters supported in the URL. Not all objects are suitable to be serialized, only relatively small-size objects are appropriate to do so.

The LosFormatter class is extremely useful especially you override the SaveViewState() or LoadViewState() of the Page class/ Control, you need this class to do the serialization or deserialization of control states.

[.NET 2.0] Foreach-ing Arrays with Iterator & Generics

2006-10-12T07:44:00.000-07:00

Prior to the .NET 2.0, we have to use for statement to loop each of the element in the array. C# 2.0 offers you another one - Iterator, a familar word that in the Design Pattern, a method that can be used to enumerate through collection of values such as array and return each of the element in turn. For instance,

Code


string[] aryNames = {"Alvin","Westley","Howard"};
 
foreach(string str in IterateMe(aryNames))
  Response.Write(str + " ");
 
 
private IEnumerable IterateMe(string[] str)
{
      int iLength = str.Length;
      for(int index=0;index < iLength;index++)
          yield return str[index];
}

Output :
Alvin Westley Howard

Your Iterator method has to return either IEnumerable or IEnumerator type to the caller function, which in turn contains the element value of current iteration. Make it more usable with generic,

Code:


string[] aryNames = {"Alvin","Westley","Howard"};
int[] iValues = {1,2,15,4,6,9};
 
foreach(string str in IterateMe< string >(aryNames,true))
  Response.Write(str + " ");
  
foreach(int value in IterateMe< int >(iValues,true))
  Response.Write(value + " ");
 
 
private IEnumerable IterateMe< T >(T[] str, bool sorting)
{
    if(sorting && !typeof(T).Name.Equals("Object"))
        Array.Sort(str);
 
    int iLength = str.Length;  
    for(int index=0;index < iLength;index++)
          yield return str[index];
}

Output:
Alvin Howard Westley
1 2 3 6 9 15

My Revamped Blog !

2006-10-11T09:40:00.000-07:00

I have just revamped my blog by changing the page template to my favourite color and font family + size combinations.

Testing..


    Response.Write("Welcome to my newly-revamped blog!");

[AJAX] IE 7 getting Native XMLHTTPRequest object

2006-10-10T09:58:00.000-07:00

Programming AJAX Web application would be in trouble when users turning off the ActiveX support in IE. The sentence of


var xmlHttp = new ActiveXObject("Microsoft.XMLHTTP");

would not be working. But luckily, there is native XMLHTTPRequest object like what non-IE browsers do.


var xmlHttp = new XMLHttpRequest();

No ActiveX object is needed, so far. Here is the code needs to be added to check the support of native XMLHTTPRequest object


if (window.XMLHttpRequest)
{
   // If IE7, Mozilla, Safari, etc: Use native object
   var xmlHttp = new XMLHttpRequest();
}
else 
{
   if (window.ActiveXObject){
          // use the ActiveX control for IE5.x and IE6
          var xmlHttp = new ActiveXObject("Microsoft.XMLHTTP"); 
  }
}

[ASP.NET] ASP.NET's FREE Components List

2006-10-10T09:29:00.000-07:00

Jason'd compiled a long but yet comprehensive list of free components of ASP.NET, ranging from data-bound controls, graphing controls, navigation controls to templating and reporting controls..

Check Free Components for ASP.NET for more detailed info. (Check user comments for some other useful controls too)

Hunting for more !

[My Project] LVUtilities Lib v1.109 Updates : XmlHelper class

2006-10-09T01:43:00.000-07:00

I have just done the following changes in this LVUtilities Lib v1.109

New XMLHelper class
1. ReadValuesByXPath()
2. ReadElementValues()
3. ReadElementList()
4. ReadSingleElementValue()
5. ReadAttributeValue()

6. UpdateValuesByXPath()
7. UpdateSingleValueByXPath()

8. InsertAttributeByXPath()
9. InsertElementByXPath()

10.RemoveAttributeByXPath()
11.RemoveElementByXPath()

Changelog
1. Overloaded Generic IsDBNullValue<>() method
2. Improve the string manipulation performance in string replacement

Download the library + examples here

Updates : SecurityHelper and ValidationHelper classes

2006-10-07T04:57:00.000-07:00

I have enhanced the SecurityHelper and the ValidationHelper classes to support :

1. ValidationHelper class
The IsValidStateNames(), IsValidPhoneNO() , IsValidHPNO(), IsValidICNO(), IsValidPostalCode() are no longer limited to Malaysia context. For instance, to validate the phone number of US,


if(ValidationHelper.IsValidPhoneNO(strInput,"(999)999-9999"))
  // true or false

in which the "(999)999-9999" is the input mask pattern that used to validate the US phone number format.

2. SecurityHelper class
The CompareValueToHash() method can now accept both string parameters of user input and hashed string. For instance,


string strPassword = SecurityHelper.GenerateHashInString("abc");
if(SecurityHelper.CompareValueToHash("abc",strPassword))
  // true or false

Download Examples and Library here

[.NET] Debugger, dont STEP INTO my method !

2006-10-06T08:16:00.000-07:00

Had you experienced whenever you debug your application using the step-into feature (F11), the debugger will go through the implementation of the methods being called, which you might not want it to. How do you "tell" the debugger to skip the method and move on to next statement? Here is the answer.

using System.Diagnosis;

[DebuggerStepThrough]
public void MyMethod()
{
....
}

Be noted, the [DebuggerStepThrough] attribute can be used in the get/set of property, too.

public string MyProperty
{
[DebuggerStepThrough]
get { }
}

MyMethod : "Thanks for skipping over me, Debugger !"

[AJAX] AJAX Tutorials !

2006-10-05T22:16:00.000-07:00

Planning to develop the AJAX app on the fly? Here are AJAX tutorials links that you can't afford to lose....

1. 130+ AJAX Tutorials [AjaxMatters]
2. Top 126 AJAX Tutorials
3. 60 Tutorials of AJAX [Max Kiesler]

[My Project] My Helper library is OUT !

2006-10-05T10:37:00.000-07:00

I have been working on creating a simple library which consists of 5 major helper classes in this version as following:

1. ValidationHelper class

a. IsValidFileName()
b. IsValidNumeric()
c. IsDBNullValue()
d. IsValidDateFormat()
e. IsAlpahbet()
f. IsAlphaNumeric()
g. IsValidUserName()
h. IsValidName()
i. IsValidStateName() // Malaysia Context only
j. IsValidPhoneNo() // Malaysia Context only
k. IsValidHPNo() // Malaysia Context only
l. IsValidICNo() // Malaysia Context only
m. IsValidPostalCode() // Malaysia Context only
n. IsValidEmail()

2. SecurityHelper Class

a. GenerateHash() // MD5, SHA1, SHA256, SHA384, SHA512
b. GenerateHashInString()
c. CompareValueToHash()
d. TDESEncrypt() and TDESDecrypt() // Triple DES

3. MailHelper Class

a. SendMessage()

4. HttpPostHelper Class

a. SendRequest()

5. ClientScriptHelper Class (ASP.NET only)

a. ShowMessage()

* I am working on the XmlHelper and FileHelper classes in coming release.

Download
Download Library DLL and Examples : http://www.megaupload.com/?d=1G96G4U9

[ADO.NET] 3 ways closing DataReader & Connection..

2006-10-04T19:03:00.000-07:00

These are the 3 common ways I use to close the XDataReader and DB Connection...

1. Use CommandBehavior.CloseConnection

SqlDataReader reader = null;

try
{
   reader = cmd.ExecuteReader(CommandBehavior.CloseConnection);

   // do stuff
}
catch(Exception ex)
{
   // logging error
}
finally
{
   // conn will be closed when reader is closed
   reader.Close();
}

2. Explicitly Close DataReader and Connection

SqlDataReader reader = null;

try
{
   reader = cmd.ExecuteReader();

   // do stuff
}
catch(Exception ex)
{
   // logging error
}
finally
{
   // conn will be closed when reader is closed
   reader.Close();
   conn.Close();
}

3. Use Using Statement

using(SqlConnection con = new SqlConnection(<connection string>))
{
   using(SqlCommand cmd = new SqlCommand(<query>,conn))
   {
      using(SqlDataReader reader = cmd.ExecuteReader())
      {
         // do stuff
      }
   }
}

The first one is my preference, what about you?

[.NET 2.0] New Methods / Operator in .NET 2.0

2006-10-04T08:09:00.000-07:00

I summarize some of the useful new methods and operator introduced in .NET 2.0 here...

1. String.Contains()

Old way in .NET v1.X

string str = "Today is monday";

if(str.IndexOf("is")!= -1)
// do stuff

New way in .NET 2.0

string str = "Today is monday";
if(str.Contains("is"))
// do stuff

2. String.StartsWith()

Old way

string str = "Alvin Chooi";
if(str.IndexOf("Al")==0)
// do stuff

New way

string str = "Alvin Chooi";
if(str.StartsWith("Al"))
// do stuff

3. String.EndsWith()

Old way

string str = "Alvin Chooi";
if(str.IndexOf("Al")== str.Length-2)
// do stuff

New way

string str = "Alvin Chooi";
if(str.EndsWith("Al"))
// do stuff

4. String.IsNullOrEmpty()

Old way

string str = "Alvin Chooi";
if((str!=null) && (str.Length > 0))
// do stuff

New way

string str = "Alvin Chooi";
if(!String.IsNullOrEmpty(str))
// do stuff

5. TryParse()

Old way

int iValue;
try
{
iValue = Int32.Parse("2x");

return true;
}
catch(Exception ex)
{
return false;
}

New way

int iValue
bool isValid = Int32.TryParse("2x",out iValue);

6. Non-null operator (??)

Old Way

string str = null;

if(str!=null)
Response.Write(str);
else
Response.Write("It is null");

New way

Response.Write(str?? "It is null");

7. Eval() & Bind() in ASP.NET 2.0

Old way

<%# DataBinder.Eval(Container.DataItem,"MyField") %>

New way

<%# Eval("MyField") %> // one-way binding
<%# Bind("MyField") %> // two-way binding

Codes become lesser and lesser... (But why our work will never be lessen?, ha)

[ASP.NET] RequiredFieldValidator + OnClientClick issue

2006-10-03T22:15:00.000-07:00

OnClientClick property was introduced in ASP.NET 2.0 to eliminate the need of developers to add the client-side "onclick" to a control programmatically. For instance,

JS script

function ConfirmMe()
{
return confirm("Do you want to proceed?");
}

ASPX

<asp:TextBox id="txtName" runat="server"/>
<asp:Button id="btnSubmit" OnClientClick="return ConfirmMe()" Text="Submit" runat="server"/>

Well, that is pretty straightforward. BUT, it goes weird when you have a validator control (eg. RequiredFieldValidator) that is used to validate the "txtName" textbox server control. For instance,

<asp:TextBox id="txtName" runat="server"/>

<asp:RequiredFieldValidator id="rq1" ControlToValidate="txtName" ErrorMessage="Name cannot be blank" Display="Dynamic" runat="server"/>

<asp:Button id="btnSubmit" OnClientClick="return ConfirmMe()" Text="Submit" runat="server"/>

Whenever you press the button with no textbox value, the client-side confirmation dialog will be invoked first before the validator message is able to show up. This isn't what we expected it to behave. I tried several ways to overcome this problem, including using CLIENT CALLBACK, disabling the CauseValidation, but it failed. Finally, I was able to find a solution by adding JUST ONE line in the JS script.

function ConfirmMe()
{
   if(Page_ClientValidate())
      return confirm('Do you want to proceed?');

   return false;
}

Another discovery today !

Validator for CheckBox/CheckBoxList

2006-10-03T18:45:00.000-07:00

By default, there is no validator server control available to validate the validity of checkbox in ASP.NET. On the basis practice, I have to use CustomValidator control or using pure client-side JS or the postback checking on server-side. But to ease the job, you can try Custom Checkbox Validator, created by Microsoft MVP, Scott Mitchell.

URL : http://aspnet.4guysfromrolla.com/articles/092006-1.aspx

IsValid? True !

About SqlDataSource control

2006-10-03T01:38:00.000-07:00

I've been running a problem of losing the selected items in a listbox, which is bound to the SqlDataSource control on every postback. The code is :

<asp:ListBox ID="lstBox" runat="server" CssClass="listbox-flat" DataSourceID="objSQL" DataTextField="userFullName" DataValueField="RID" SelectionMode="Multiple" Width="335px" Enabled="False"></asp:ListBox>

<asp:SqlDataSource ID="objSQL" runat="server" ConnectionString="<%$ ConnectionStrings:dbConnectionString %>" SelectCommand="SELECT * FROM Table1"> </asp:SqlDataSource>

As far as I remember, the selections of ListBox SHOULD be retained on the postback since the ViewState of the control will be saved in the SaveViewState event after the binding, but the thing goes weird unexpectedly :(. Since the automatic databinding performed by the SqlDataSource takes place in the Page_PreRender event (You can check it by using Page Tracing). Here is the screenshot:

* The ListBox is bound to the DataSource control.

Eventually, I used another way round by programmatically binding the ListBox in that PreRender event with IsPostBack checking. (No choice, this is the best way I could think of, I will find out the cause of the problem later)

Steps
1. Set the DataSourceID property of ListBox to empty (So that it would not be bound to the SqlDataSource until runtime)

2. Obtain the SqlDataSource's IEnumerable collection by calling the DataSource's Select() method with no argument.

3. Call the ListBox's DataBind() method

Code

protected void Page_PreRender(Object sender, EventArgs e)
{
if(!IsPostBack)
{
   lstBox.DataSourceID = String.Empty;
   lstBox.DataSource = objSQL.Select(DataSourceSelectArgument.Empty);
   lstBox.DataBind();
}
}

Good Practice
For ordinary data binding operations like binding DataGrid, ListBox, DropDownList, etc, we normally perform in Page_Load() event. Actually, we can do the binding in the Page_PreRender event (bind the data-bound controls as late as possible) before the controls are about to be rendered.

Do always check IsPostBack before performing any data binding to reduce the roundtirp to server.

Digging into detailed observations...

Patterns & Practices Guidance Explorer

2006-09-29T02:30:00.000-07:00

ScottGu recommended a very useful tool,namely Patterns & Practices Guidance Explorer, which contains hundreds lists of security & performance best practices and patterns. The usage scenarios are

"1. Improve the security and performance of your application with guidelines and checklists that match your application exactly.
2. You can build custom sets of guidance and share with your team as recommended practice.

3. You can add new guidance to the library to share with your team, your company or the larger development community. "

Hope this helps...

Free PDF Converter in .NET

2006-09-21T01:37:00.000-07:00

I have been struggling looking for non-commercial PDF library to convert my ASP.NET pages to PDF for reporting purpose. Finally, I decided to choose iTextSharp v3.1.5 library among other free libraries. I have found out that the library is extremely handy with few cool features such as document watermarking and encryption, user permission on copying, printing the document and the list goes on.

Anyway, there is an annoying problem - the library has iTextSharp.text.Image class, which conflicts with the System.Drawing.Image class. Hence, the class will be ambiguous if both namespaces are imported in your project. Anyway, this matter is bearable for me. :)

Library Project Homepage : http://sourceforge.net/projects/itextsharp/
Library Download : http://prdownloads.sourceforge.net/itextsharp/itextsharp-3.1.5-dll.zip?download
Tutorial (C#) Download : http://prdownloads.sourceforge.net/itextsharp/iTextSharp.tutorial.01.zip?download
Tutorial (VB) Download : http://prdownloads.sourceforge.net/itextsharp/iTextSharp.tutorial.VB.NET.01.zip?download

Cool !

No more DbCommandWrapper class

2006-09-19T22:23:00.000-07:00

I'm testing the Enterprise Library January CTP 2006, and I have been searching for the DbCommandWrapper class and the GetXCommandWrapper() methods of Database class but they do not exist. Eventually, I found out that they are actually removed from the DAAB in this new version.

Example of code for old versions of DAAB prior to January 2006 release

Database db = DatabaseFactory.CreateDatabase();
DBCommandWrapper dbCommand = db.GetStoredProcCommandWrapper("SelectAuthors");
dbCommand.AddInParameter("AuthorID", DbType.String, strAuthorID);
DataSet dsAuthors = db.ExecuteDataSet(dbCommand);

Example of code for DAAB January CTP 2006

using System.Data.Common;

Database db = DatabaseFactory.CreateDatabase();
DbCommand dbCommand = db.GetStoredProcCommand("SelectAuthors");
db.AddInParameter(dbCommand,"AuthorID", DbType.String, strAuthorID);
DataSet dsAuthors = db.ExecuteDataSet(dbCommand);

Be informed that some online examples are using DbCommandWrapper class. thus, make some modification like above. Visit Intro to DAAB January 2006 for more version migration issues.

EBook on Threading in .NET

2006-09-18T23:36:00.000-07:00

Thinking of learning Threading in .NET? I've found out a very handy ebook that explains fundamentals about .NET Threading. Worth to read !

Web : http://www.albahari.com/threading/
Book (PDF) : http://www.albahari.com/threading/threading.pdf

Sharing~~

Corrections of ClientCallBack Example

2006-09-18T20:02:00.000-07:00

I have tested the example of ASP.NET 2.0 new feature - ClientCallBack at ASP.NET 2.0's Client Callback Feature, but found not working due to:

1. There is no this.GetCallbackEventReference()
2. RaiseCallbackEvent() does not return string data type, from ICallbackEventHandler
3. Missing of GetCallbackResult(), that defined in ICallbackEventHandler

Finally I get it done by making some corrections:

1. Use this.ClientScript.GetCallbackEventReference()
2. Define the RaiseCallbackEvent() that returns no data type
3. Define GetCallbackResult() to return the result to the client-side

Corrected code:

protected void Page_Load(object sender, EventArgs e)
{
       sCallBackInvocation = this.ClientScript.GetCallbackEventReference(this, "message", "ShowServerTime", "context", "OnError",true);
}

public void RaiseCallbackEvent(string eventArgument)
{
       sCallBackInvocation = DateTime.Now.ToString();
}

public string GetCallbackResult()
{
       return sCallBackInvocation;
}

Recommendation
1. Check whether the browser does support callback feature in Page_Load() event

if (!Request.Browser.SupportsCallback)
// error message

Page Redirect After Login in Forms Authentication

2006-09-18T00:30:00.000-07:00

Normally, we will be redirected to the originally requesting page, with ReturnUrl query string appended in the URL after successful user login in Form Authentication using FormsAuthentication.RedirectFromLoginPage() method. Otherwise, the we will be redirected to default.aspx, by default.

The question is, what method do we need to use if we want to redirect the users to different page other than default.aspx? Here I demonstrate how it can be accomplished.

bool isAuthenticated = true;

if(isAuthenticated) // after user is authenticated
{
   if(Request.Params["ReturnUrl"] != null)
   {
      FormsAuthentication.RedirectFromLoginPage(txtUsername.Text, false);
   }
   else
   {
      FormsAuthentication.SetAuthcookie(txtUsername.Text, false);
      Response.Redirect("yourpage.aspx");
   }
}

Get rid of default ones.... !

VIEWing your controls STATE

2006-09-17T22:37:00.000-07:00

ViewState is one of the state managements in the ASP.NET and it is paramount especially in custom server control development. Proper understanding and use of ViewState could definitely save your life, notably in postback model in ASP.NET.

Here I recommend some articles of ViewState for your readings :

1. Understanding ASP.NET ViewState [MSDN]
2. TRULY Understanding ViewState [InfiniteLoop's Blog]
3. Understanding View State in ASP.NET [aspAlliance]
4. ViewState: All You Wanted to Know
5. The ASP.NET ViewState [MSDN Magazine]

You already mastered another ASP.NET 's state management ;)

ASP.NET 2.0 Full Page Life Cycle Diagram

2006-09-17T20:27:00.000-07:00

Here is the big yet comprehensive diagram illustrating the page life cycle in ASP.NET 2.0

To recall the Page Life Cycle in ASP.NET v1.x

ASP.NET 2.0 ones much complex !

Checking Existence of Available SQL Server

2006-09-17T19:39:00.000-07:00

Of late, I had read an article about determining list of SQL Servers that are available to your application.

.NET 1.1
There is no way but you need to use COM-based SQLDMO to have an interop call.

using SQLDMO;

try
{
NameList objSQLList;
ApplicationClass objSQLApp = new SQLDMO.ApplicationClass();

objSQLList = objSQLApp.ListAvailableSQLServers();

foreach(string name in objSQLList)
Response.Write(name + "
");
}
catch(Exception ex)
{

}
finally
{
objSQLApp.Quit();
}

* You have to reference the SQLDMO.dll, which located at :\Program Files\Microsoft SQL Server\\Tools\Binn by default.

.NET 2.0
It is much easier. You can retrieve the list of servers available using System.Data.Sql namespace, by enumerating each of the server instance.

SqlDataSourceEnumerator enumerator = SqlDataSourceEnumerator.Instance;
DataTable datatable1 = enumerator.GetDataSources();
foreach (DataRow row in datatable1.Rows)
{
Response.Write("Server Name:" + row["ServerName"] + "
");
Response.Write("Instance Name:" + row["InstanceName"] + "
");
}

Sometimes, these methods are better trying to connect the DB Server to see whether it is succesfully connected, but the output not always accurate due to :

1. Invalid Login / User credentials
2. Timeout due to network congestions

Neat !

Storing Static JavaScript in Resource File [ASP.NET 2.0]

2006-09-12T22:49:00.000-07:00

In ASP.NET , we normally embedded the script string in the code-behind and invoke the RegisterStartupScript() to emit the JavaScript. The normal way we do like

ClientScriptManager csManager = Page.ClientScript;

StringBuilder strScript = new StringBuilder();
strScript.AppendLine("<script>");
strScript.AppendLine(@"var response = confirm('{0}, do you want to continue ?');","Alvin Chooi");
strScript.AppendLine("if(response)");
strScript.AppendLine(" alert('OK')");
strScript.Append("</script>");

if (!csManager.IsStartupScriptRegistered("ScriptTest"))
csManager.RegisterStartupScript(this.GetType(), "ScriptTest", strScript.ToString());

The script is less readable and messy. What the worst is that would be a slightly string manipulation overhead. In ASP.NET 2.0, you can store this static script in the resource file resx, which is stored in the App_GlobalResources folder.

Screenshot

You could able to reference the script string from the resource file without hard-coding the script in the code-behind by

ClientScriptManager csManager = Page.ClientScript;

if (!csManager.IsStartupScriptRegistered("ScriptTest"))
{
csManager.RegisterStartupScript(this.GetType(), "ScriptTest", String.Format(Resources.script.AlertMeScript,"Alvin Chooi"));
}

Neat and manageable !

Microsoft Interview Questions

2006-09-12T21:04:00.000-07:00

Recently, I'd read the post of Jason Looney about questions asked Microsoft Interview. Amazing and extraordinary !

Worth to read
1. Microsoft Interview Questions Guide
2. ASP.NET Interview Questions (The only one I could answer well ;p)
3. .NET Interview Questions
4. The Guerrilla Guide to Interviewing

Hunting for more....

Fastest Ever Browser ?

2006-09-08T10:51:00.000-07:00

Recently, I had read an article of What is the world's fastest browser, and it claimed that optimized firefox (swiftfox) is the fastest browser among others in 64-bit environment. Whereas, Konqueror is crowned best-performance browser in Linux.

Accessing Text in DataGrid's TemplateField

2006-09-07T20:32:00.000-07:00

If you try to access the content of TemplateField

<asp:TemplateColumn>
<ItemTemplate>
<%# DataBinder.Eval(Container.DataItem,"au_fname") %>
<%# DataBinder.Eval(Container.DataItem,"au_lname") %>
</ItemTemplate>
</asp:TemplateColumn>

using DataGrid1.Items[0].Cells[0].Text (I assume it is first column), you would probably get nothing. It is because the text would be placed in a special literal control called DataBoundLiteralControl, which is auto-generated by the compiler IF you place the data-bound expression in the ItemTemplate. Therefore, to access its content. You can use this,

((DataBoundLiteralControl)dgrdAuthors.Items[0].Cells[0].Controls[0]).Text

You get it !!!

Capitalizing Every First Letter of Words

2006-09-07T20:07:00.000-07:00

How would you do that in your first thought? Would you create your own user-defined function for that purpose in .NET? Please do not reinvent the wheel because there is built-in method available.

using System.Globalization;

string str = "cool strINg";
string str1 = "COOL STRING";

TextInfo objTextInfo = new CultureInfo("en-US",false).TextInfo();
str = objTextInfo.ToTitleCase(str); // Cool String *CORRECT
str1 = objTextInfo.ToTitleCase(str1); // COOL STRING *WRONG
str1 = objTextInfo.ToTitleCase(str1.toLower()); // Cool String *CORRECT

* For FULL CAPITALIZED words, you need to convert them to lower case before passing it to the ToTitleCase() method.

Dynamically Adding Columns to GridView

2006-09-07T19:27:00.000-07:00

I was asked in ASP.NET Forum why the dynamically inserted BoundField would be lost from the GridView's column collection on postback (eg. Button click in GridView)?

Many people would dynamically create controls in the Page_Load event. But I afraid it isn't a right place for the dynamic creation because the data in the control may not be pertained. (ViewState lost on postback). If possible, do the dynamic control creation in Initialization event (eg. Page_Init).

One of the differences between Page_Init and Page_Load is Page_Init event is only fired once and it is not fired on each postback, whereas the Page_Load event is always fired on every postback.

Back to the topic, hence, to avoid the dynamic GridView's fields gone from its GridView's collection on postback, you SHOULD always create it in GridView's Init event. For instance,

Protected Sub GridView_Init(sender as Object, e as EventArgs) Handles myGridView.Init
Dim field As New BoundField()
field.DataField = "data"

myGridView.Columns.Insert(0, field) 'To be first column in GridView
End Sub

The related discussion can be found here

Good luck !

Recovering DB with LDF only

2006-09-07T18:56:00.000-07:00

Of late, I witnessed the loss of mdf (due to corrupted or accidentally removed/overwritten) is painful. It is easy to recover the data from the .bak (database backup) or mdf(database file), but how would it be done the same thing just using ldf (log file) ? Here are the 10 steps that needs to be followed in order you have data restored.

1. Find out your previous database .bak file (backup file).

2. Backup your updated .MDF file and the latest LDF file

3. Delete your current database
- If you cant access to SQL Queries Analyzer no worries just open with master
- type :

DROP DATABASE SAMPLE_DB
GO

p/s - Make sure your infected MDF and LDF fully deleted from C:\Program Files\Microsoft SQL Server\MSSQL\Data

4. Create a new database with FULL Mode
- Open SQL Analyzer type following codes:
- type :

USE master
GO

CREATE DATABASE myDB
ON(NAME='myDB',
FILENAME='C:\Program Files\Microsoft SQL Server\MSSQL\Data\myDB..mdf')
LOG ON(NAME='myDB',
FILENAME='C:\Program Files\Microsoft SQL Server\MSSQL\Data\myDB_Log')
GO

ALTER DATABASE myDB
SET RECOVERY FULL
GO

EXEC sp_dboption 'myDB', 'autoclose', true
GO

5. Delete existing data and log files
- type :

EXEC master..xp_cmdshell 'del C:\Program Files\Microsoft SQL Server\MSSQL\Data\myDB.mdf
EXEC master..xp_cmdshell 'del C:\Program Files\Microsoft SQL Server\MSSQL\Data\myDB_Log'
GO

- Make sure no more .MDF and .LDF in your C:\Program Files\Microsoft SQL Server\MSSQL\Data

6. Stop your whole database

7. Copy your previous infected .MDF and .LDF to C:\Program Files\Microsoft SQL Server\MSSQL\Data

8. Backup the log with NO_TRUNCATE
- Note that backup will error due to inaccessible data file, but log will still be backed up.
- type :

BACKUP LOG myDB
TO DISK='C:\Backups\myDB.bak'
WITH NO_TRUNCATE, INIT
GO

- Make sure backup success, check the .bak file if the file size too small if compare with your .LDF file size then might be error, maybe your infected .LDF file stored wrongly, but normally no problem.
- This is very important file, no error here then fine already.

9. Restore your database

- Restore your previous backup database. Doesn't matter if your data not up to date.
- type :

USE Master

RESTORE DATABASE DatabaseName
FROM DISK = 'c:\Backups\old_myDB.BAK
WITH NORECOVERY

- p/s Remember use WITH NORECOVERY

10. Restore your backup log

- type:

RESTORE LOG myDB
FROM DISK = 'C:\Backups\myDB.bak'
WITH RECOVERY

- p/s Remember use WITH RECOVERY

- Check your data again.
- Done !!!

JSON Reader

2006-09-07T08:27:00.000-07:00

Today is a boring day, I decided to create a lightweight JSON reader for JSON manipulation using JavaScript (After I have not been touching JSON and JavaScript prototype since my FYP). My intention of creating this JSONReader is simply to manipulate/read the JSON value with string path supplied. To start off, I use a very simple JSON in my example.

var myJSON = {"Image": {
"Width":800,
"Width":200,
"Height":600,
"Title":"View from 15th Floor",
"Thumbnail":
{
"Url":"http:\/\/scd.mm-b1.yimg.com\/image\/481989943",
"Height": 125,
"Width": "100"
},
"IDs":[ 116, 943, 234, 38793 ]
}}

Now, we "instantiate" the object of JSONReader with the JSON format passed as parameter to its constructor.

var reader = new JSONReader(myJSON);

Then at the frontend, we allow the users to key in the string path in the TextBox.

<input type="text" id="txtPath" name="textfield">
<input type="button" onclick="GetValue()" name="Submit" value="Get the Value">

<span id="lblMessage"/>

The string path provided by user will be passed to the FindValueByKeyPath() method of JSONReader in the GetValue() function to obtain the value in JSON. The $F() is the method in the prototype.js that used to retrieve the value of specified HTML element.

var path = $F("txtPath");
lblMessage.innerHTML = reader.findValueByKeyPath(path);

Test Results
User Input : Image\Thumbnail
Output : {"Url":"http://scd.mm-b1.yimg.com/image/481989943","Height":125,"Width":"100"}

User Input : Image\Thumbnail\Url
Output : http://scd.mm-b1.yimg.com/image/481989943

Screenshot

Additional Libraries Needed
prototype.js
$F(), class.Create()

json.js
parseJSON(), toJSONString()

The complete example can be downloaded here

Still boring ~~ :|

ASP.NET Best Practices

2006-09-02T08:53:00.000-07:00

I have been collecting ASP.NET BEST PRACTICES articles for quite some time. Here are some articles/guidelines that may be handy to you.

ASP.NET Performance Best Practices
1. Improving ASP.NET Performance [MSDN]
2. Developing High-Performance ASP.NET Application [MSDN2]
3. Performance Tips and Tricks for .NET Applications [MSDN]
4. 10 Tips for Writing High-Performance Web Applications [MSDN Magazine]
5. Improving String Handling Performance in .NET Framework Applications [MSDN]
6. Performance Strategies for Enterprise Web Site Development [Code Project]
7. Improving SQL Server Performance [MSDN]
8. Developing High-Performance ASP.NET Applications [aspalliance]

ASP.NET Security Best Practices
1. Building Secure ASP.NET Applications: Authentication, Authorization, and Secure Communication [MSDN]
2. Improving Web Application Security: Threats and Countermeasures [MSDN]
3. ASP.NET Security: 8 Ways to Avoid Attack [devx]
4. SQL Injection Attacks - Are You Safe? [sitepoint.com]
5. Security Practices: ASP.NET 2.0 Security Practices at a Glance [MSDN]
6. An Introductory Guide to Building and Deploying More Secure Sites with ASP.NET and IIS [MSDN Magazine]

ADO.NET Best Practices
1. Best Practices for Using ADO.NET [MSDN]
2. ADO.NET Best Practices [devx]
3. ADO.NET Best Practices, Part II [devx]
4. ADO.NET Best Practices [code-magazine]
5. Using Data with ASP.Net - 10 of my 'Best Practices'
6. Optimized ADO.NET [theserverside.net]
7. ADO.NET and SQL Server Performance Tips [sql-server-performance]

Error Logging / Exceptions Best Practices
1. Best Practices for Handling Exceptions [MSDN]
2. Exception Handling Best Practices in .NET [codeproject]
3. Exception Handling in Enterprise Applications [devcity]
4. Exception Handling Best Practices in .NET

Naming/Standards Guidelines
1. Naming Guidelines [MSDN]
2. C# Coding Standards and Best Programming Practices [dotnetspider]
3. SSW's Naming Conventions [SSW]

I'll post more useful ASP.NET links in the future. Hope this helps... :)

Displaying 2 Field Names in 1 Column of DataGrid

2005-02-27T00:58:00.000-08:00

In the previous articles, I used the Authors table from Pubs database to display data in the datagrid. As you can see the screenshots, the author's firstname (au_fname) and author's lastname (au_lname) are displayed in the different columns. Perhaps, you would be wondering how to combine both au_fname and au_lname fields into 1 column. Now, I will demonstrate two ways to do that. First method using new combined DataColumn, and second method using Data Binding in the ItemDataBound event. To avoid this article from becoming lengthy, I combine both methods into 1 solution. [As you will notice that there are two same columns, which have author's full names.]

In the aspx page,

<asp:datagrid id="dgrdAuthors" onItemDataBound="dgrdAuthors_ItemDataBound" .......>
<Columns>
     <asp:BoundColumn DataField="au_id" HeaderText="Author ID"/>
     <asp:BoundColumn DataField="FullName" HeaderText="Name (DataColumn)"/>;      <-- First method

      <asp:TemplateColumn HeaderText="Name(DataItem)">      <-- Second method
          <ItemTemplate>
               <asp:label id="lblFullName" runat="server"/>
          </ItemTemplate>
     </asp:TemplateColumn>

     <asp:BoundColumn DataField="phone" HeaderText="Phone"/>
     <asp:BoundColumn DataField="city" HeaderText="City"/>
     <asp:BoundColumn DataField="state" HeaderText="State"/>
</Columns>
</asp:datagrid>

In my DataGrid, there are total number of 6 columns, which 2 of columns are the same (Name(DataColumn) and Name(DataItem) columns). As you can notice, one of the columns used is TemplateColumn. This column is customizable and provide more flexibility in formatting the appearance of the displayed data. The purpose of using this TemplateColumn here is I want to customize the data (combining both au_fname and au_lname) at run time.

Definitions : TemplateColumn class | BoundColumn Class | ItemDataBound Event

See Also : Customizing DataList Items at Run Time [MSDN]

In the <script> tag or code behind ,

For the first method, I have created a new datacolumn, which has FullName column name and string data type. The Expression property of DataColumn can be used for calculating values from different data fields, combining columns and so on. At here, I'm combining the au_fname and au_lname fields as 1 field (column). After that, I add this new column to the DataColumnCollection of "Authors" DataTable, and then bind this DataSet to DataGrid. Eventually, the DataSet has 7 fields.

In the ItemDataBound event, I check the data rows using the If..... Statement, and for each data row, I use the FindControl() to search and locate the Label server control that is placed in the TemplateColumn there by its ID. Of course, I have to cast the returned result of the FindControl() to Label type. Let us proceed to next line, which will be a bit complicated.

The e.Item.DataItem is an object that represents the data item in the DataGrid. It can only be placed in the data rows like Item, AlternatingItem, SelectedItem and EditedItem. Whereas the DataRowView represents each displayed data row. The ((DataRowView)e.Item.DataItem)["au_fname"] is an explicit casting. Alternatively, you can change to the DataBinder.Eval(e.Item.DataItem,"au_fname"), which is the equivalent to the ((DataRowView)e.Item.DataItem)["au_fname"]. The reason I use the explicit casting is because it offers better performance than DataBinder.Eval() in avoiding the cost of reflection.

TRY it and GET it !

Definitions : DataColumn.ColumnName Property | DataColumn.DataType Property | DataColumn.Expression Property | Type Class | DataColumnCollection.Add() | DataGridItem.DataItem | DataRowView Class

See Also : Improving ASP.NET Performance [MSDN]

Showing number of records at footer in DataGrid (Pt. 2)

2005-02-26T23:44:00.000-08:00

This is my second part of the "Showing number of records at footer in DataGrid". In the first part, I showed how to do it when you using SqlDataReader. And now, I show how can it be done if you are using DataSet. Compared to the SqlDataReader, it is much easier for you using DataSet.

DataSet

DataSet object is always a disconnected recordset. It is the core of the ADO.NET disconnected architecture. Unlike DataReader, it is not connected to the database. It is filled by the SqlDataAdapter (a bridge between the connected and disconnected objects), which associates with the Database. Behind the scenes, a SqlDataReader is created implicitly and the rowset is retrieved one row at a time in succession and sent to the DataSet. Once all the data is in the DataSet, the implicit SqlDataReader is destroyed and the SqlConnection is closed.

DataSet is fully navigable. It can traverse forward and backward, which unlike the SqlDataReader that moves forward only. Also, the DataSet is flexible since the DataTables in the it can be sorted, filtered and searched. Also, it is fully bindable. it can be bound to the data source of data controls every time. The DataSet can reduce the roundtrip to the database server, but, however, it increases the memory footprint in where it is stored. If you are retrieving 1 million of records, these records will be stored in the memory, which occupies the system resources.

Therefore, you have to be careful which one you are using. If you just merely want to view or display the data, always stick with the SqlDataReader ; if you would like to sort, filter , search the retrieved data, you can consider the DataSet. Nevertheless, there are some situations where you must use DataSet. For instance, when doing paging in DataGrid, the DataSet is needed because it requires your data source implements ICollection interface.

See Also : DataSet Vs. DataReader [sitepoint] | Contrasting the ADO.NET DataReader and DataSet [MSDN Magazine]

Definition : ICollection interface [MSDN]

Here, I am not going to explain the code line by line since I explained it in my previous article. Instead of using the SqlCommand, I use the SqlDataAdapter, which in turn will fill the DataSet with the DataTable named "Authors". After that, I bind the DataTable to the data source of the DataGrid. Note that I declare the DataSet outside of the Page_Load() event because it can be used in the ItemDataBound event later. Simple ?

In the ItemDataBound event,

protected void dgrdAuthors_ItemDataBound(Object sender, DataGridItemEventArgs e)
{
     if(e.Item.ItemType == ListItemType.Footer)
     {
          TableCellCollection cells = e.Item.Cells;
          int iCount = dsAuthors.Tables["Authors"].Rows.Count; // obtaining total rows in the data table.

          cells.RemoveAt(0);
          cells.RemoveAt(0);
          cells.RemoveAt(0);
          cells[0].ColumnsSpan= 4;

          if(iCount!=0)
          {
               cells[0].HorizontalAlign = HorizontalAlign.Right;
               cells[0].Text = String.Format("{0} records found",iCount.ToString());
          }
          else
          {
               cells[0].HorizontalAlign = HorizontalAlign.Center;
               cells[0].Text = "No record found";
          }
     }
}

Is this code looked simpler? Yes, we do not need to count the number of data in the datagrid one by one because we are able to the get the total rows in the rows.count property of the "Authors" DataTable of the DataSet. Instead of calling the e.Item.Cells everytime, I assign it to the variable cells with type of TableCellCollection. Therefore, it is more readable with less typing errors. For the subsequent codes, it is similar to what the previous article does.

Definitions : TableCellCollection [MSDN]

- The End -

Showing Number of Records at Footer In DataGrid (Pt. 1)

2005-02-26T09:57:00.000-08:00

Showing number of records in DataGrid

When displaying the data in the DataGrid, you likely want to show the total records that are listed in the grid. When there is no data available, the ASP.NET will generate a blank datagrid that has column names but does not have any data inside, which isn't what we wanted.

In this article, I will show you how to display the "XX records found" at the footer, and show "No record found" when there is no data in the datagrid. Same thing, I use the Pub database and Author table in my following example. In this example, I will use SqlDataReader and DataSet to accomplish this job.

SqlDataReader

First, let me talk about the introduction of SqlDataReader. The SqlDataReader, which implements the IDataReader interface, is connected, read-only and forward-only (also called firehose cursor of ADO.NET). It is a stream of data that is returned from query of database. It can read one row at a time from database and can only move forward. Those rows retrieved from database by SqlDataReader can be read, but they cannot be edited. However, the SqlDataReader has to maintain its connection to the data source (stay connected) and it is partially bindable. (It is bindable for only one time, unless you resetting the cursor to the beginning)

Back to the matter, here is my code, which can be placed in the script tag or in the code-behind of the ASP.NET page. For me, I put it in the script tag.

At the very first line of the Page_Load event, I use the if statement the check whether it is IsPostBack. Normally, we only need to bind the data from data source to DataGrid once - first time the page is loaded, but not everytime the page is requested. It could reduce the roundtrip to the database server. In the following line, I use the using block, that is, when the the code exits the using block, the SqlConnection object will be disposed automatically. [However, some saying that it is inefficient to wait until the GC to dispose the SqlConnection object. Erm.. it depends on you, if you dont use this method, then call the Close() method manually. * Close the connection as soon as possible. ]

Then, I assign the ConnectionString property with the correct value. Next, I retrieve 4 fields in the Author table, open the connection and run the bind the a data reader which is returned by result of ExecuteReader() to the datagrid.

Definitions : The "Using" Statement in C# | using Statement [MSDN] | SqlDataReader class

In the ItemDataBound event,

int iCount = 0;      // count the number of items
protected void dgrdAuthors_ItemDataBound(Object sender, DataGridItemEventArgs e)
{
     if(e.Item.ItemType == ListItemType.Item || e.Item.ItemType == ListItemType.AlternatingItem)
     {
           // .... do other things here
          iCount++;
     }
     else if(e.Item.ItemType == ListItemType.Footer)
     {
          e.Item.Cells.RemoveAt(0);
          e.Item.Cells.RemoveAt(0);
          e.Item.Cells.RemoveAt(0);
          e.Item.Cells[0].ColumnsSpan= 4;

          if(iCount!=0)
          {
               e.Item.Cells[0].HorizontalAlign = HorizontalAlign.Right;
               e.Item.Cells[0].Text = String.Format("{0} records found",iCount.ToString());
          }
          else
          {
               e.Item.Cells[0].HorizontalAlign = HorizontalAlign.Center;
               e.Item.Cells[0].Text = "No record found";
          }
     }
}

Repeating what the ItemDataBound event is, the ItemDataBound event is fired when the data is bound to each row of the datagrid. Therefore, this event is the best place for us to count the number of data in the grid. In the code above, I declare an integer-type variable with value of 0. Its value increments to 1 whenever a data is bound to the row, so that we could know how many data are actually in the datagrid. Remember, I exclude the header and footer, which are not counted in.

At the footer, I merge the 4 cells in the row by calling the removing the cells 3 times before merging it. For more information about merging cells, you can read my previous article Merging Cells in DataGrid. Then, I check whether there is any incrementation of value of iCount. If yes, that means there is data in the DataGrid. I set to the HorizontalAlign to Right, which will display the "XX records found" on the right hand side. Otherwise, it will be showing the "No record found" at the center of the footer row.

Since this article is quite lengthy, I will separate the one using DataSet in part 2 of this article.

Definitions : OnItemDataBound event | DataGrid's ItemTypes

Merging Cells in DataGrid

2005-02-26T07:27:00.000-08:00

Merging Cells in DataGrid

By default, all the data bound to the DataGrid are displayed in the tabular form, which is like an ordinary table. But, however, sometimes we neeed to adjust or change the appearance of the grid so that the displaying data would be presented nicely and clearly. And here is the reason this article comes about. I would like to show you how to merge the cells (columns in a row) in the DataGrid.

As what I said in my first article before, the onItemCreated event is the great place for us to edit and change the appearance of the datagrid. ( not to forget about the ItemDataBound event, too).

In the example below, I use the Pubs as my database, retrieving au_id, au_fname and au_lname fields frome the Authors table. Then, I would like to merge the 2nd and 3rd cells at the header, and all cells at the footer there.

protected void dgrdAuthors_ItemCreated(Object sender, DataGridItemEventArgs e)
{
     // if it is header
     if(e.Item.ItemType == ListItemType.Header)
     {
          e.Item.Cells.RemoveAt(2);      // Removing 3rd cell
          e.Item.Cells[1].ColumnSpan=2;      // Merging 2nd and 3rd cells
          e.Item.Cells[0].Text = "Author ID";
          e.Item.Cells[1].Text = "Author Full Names";
     }
     else if(e.Item.ItemType == ListItemType.Footer)
     {
          e.Item.Cells.RemoveAt(0);      // Removing the first cell
          e.Item.Cells.RemoveAt(0);      // Removing the second cell
          e.Item.Cells[0].HorizontalAlign = HorizontalAlign.Center;
          e.Item.Cells[0].Text = " End of Records";
     }
}

Header part

In the code above, first, I check whether the current row created is Header of the DataGrid. If it is header, then I remove the 3rd cell (which is au_lname field) of the row by using the RemoveAt(). This method will remove the particular cell at the specified position. I call RemoveAt(2), which removes the 3rd cell (always start with 0). Then, I merge the column 2 and 3 (au_fname and au_lname) by assigning the ColumnSpan property to 2, which takes up 2 cells in the row.

As a result, the header eventually has 2 cells. 1 is cell for Author ID field, and another is merged cell for the Author Full Name. Then, I assign the text to the appropriate cells.

Footer part

At the footer, I would like merge all the 3 cells into 1 cell. Hence, I remove the first cell by calling the RemoveAt(0). And you would wondering why I call this method twice. The reason is, when I remove the first cell of the 3 cells, the second cell becomes the first cell. Therefore, I have to call the RemoveAt(0) once again to remove the second cell, which is already become the first cell. By now, I can merge all the cells by assigning 3 to ColumnSpan property. Also, I set the HorizontalAlign to center so that the text would be displayed at the center.

Definitions: ColumnSpan | RemoveAt()

~~ Happy Programming ~~~

Changing the Row's Background Color in DataGrid When Hovering

2005-02-26T04:50:00.000-08:00

Binding data source to the DataGrid data control is simple, but adding the visual-effect on the DataGrid isn't hard too. Here, I show you the code that changing background color of the particular row when the users move the mouse pointer over the row. Since the DataGrid data control does not have the properties to let us specify the rows' backgroundcolor, thus, we have add the onmouseover and onmouseout attributes when it (DataGrid) is created (OnItemCreated event) or its data are bound to the rows.(OnDataBound). First, add the onItemCreated attribute to the DataGrid data control.

<asp:datagrid id="dgrdAuthors" onItemCreated="dgrdAuthors_ItemCreated" ........ />

then, in the <script> tag or in the code-behind, add this event (I put it in the script tag in my example). As what I have said, it can be done by putting the code in the ItemDataBound event, so just change the event name instead.

How does this code work? The ItemCreated event is fired when each row is created in the datagrid, just before the data is bound to the row. Therefore, it is commonly used to set the appearance of the datagrid's rows here. Whereas, the ItemDataBound event is fired when each data from the data source bound to the row.

The e.Item simply represents the current data row of the datagrid. And, a datagrid consists of header, data items and footer, which are under the ListItemType enumerations. The Item ItemType represents the odd-number rows in the datagrid, and the AlternatingItem ItemType represents the even-number rows.

In the code above, the ItemCreated event will be fired when each row is created. Since we do not want add the visual effect on the header and footer, we check the ItemType whether it is Item or AlternatingItem. If it is, then we add the current row with the onmouseover and onmouseout attributes, which will apply the specified styles to it.

Definitions : OnItemCreated event | OnItemDataBound event | DataGrid's ItemTypes

Alternative way :

Apart from the doing it in the ItemCreated and ItemDataBound events, there is another way to accomplish the job above. In the Page_Load() event, add this

protected void Page_Load()
{
     if(!IsPostBack)
     {
          foreach(DataGridItem item in dgrdAuthors.Items)
          {
               item.Attributes["onmouseover"] = "this.style.backgroundColor='#f7f7f7'";
               item.Attributes["onmouseout"] = "this.style.backgroundColor='#ffffff'";
          }
     }
}

Simple, The DataGridItem represents the data items in the datagrid. Therefore, I add the onmouseover and onmouseout attributes to every data item in the grid.

Definition: DataGridItem

Okay, I have done my first article regarding to the DataGrid Data Control. And soon, I will write more articles, basically more on the data controls in asp.net. :)

Welcome to My Blog

2005-02-24T23:27:00.000-08:00

Hello, in very first place, thanks for visiting this blog. Basically, I will write my own articles regarding to the ASP.NET ( more on Data Controls). Of course, if my information given incorrect, feel free to drop comments here.