Rajganesh Writes

VS 2008 Setup Project Upgrade – File replacement rules for versioned and non versioned files

2011-09-18T16:32:00.000+05:30

Recently I ran into an interesting problem in one of a .NET Windows Application's Setup project. This is a .NET 3.5 application, hence used the visual studio 2008 setup project to build MSI installer for the application. While first time installation was successful, upgrade version of the setup project failed to update few of the files.

After going through few MSDN link, got to know when you have an updated application and set RemovePreviousVersions as true in setup project “VS 2005 setup project will create MSI which completely uninstalls the old version and reinstalls the new version” but “VS 2008 will create MSI which updates the old installation based on pre defined file replacement rules”. In simplified terms below are the rules,

Both the old and new files are versioned files (i.e. application’s executable, dll files)
- If your new file is having higher version compared to old files (e.g., old dll version is 1.0.0.1 and new dll version is 1.0.0.2), old file will be replaced by new file
- If your new file is having same or lower version compared to old files (e.g., old dll version is 1.0.0.1 and new dll version is 1.0.0.0 or 1.0.0.1), old file will not be replaced by new file
Both the old and new files are non versioned files (i.e. txt files, help files)
- Installer will compare the created time and the last modified time of the old file. If the times are not same (case when the file is modified after installation in users machine), old file will not be replaced by new file (though your new file could be different than old file). The intent of this logic is to prevent Windows Installer from overwriting files that users have modified on their machine.
- Installer will compare the created time and the last modified time of the old file. If the times are same (case when user haven’t changed the file), old file will be replaced by new file.

A detailed combination of the rules are given in this MSDN link.

As one of my text file (non versioned) was not updated in my upgrade MSI, I ran the MSI with verbose log to understand how the Windows Installer treats my file. Below is the command to run MSI with verbose log in command prompt.

msiexec /i "SampleInstaller.msi" /l*v "d:\SampleInstaller.log"

/l tells Windows Installer to create a log, * tells it to log everything and v tells it to use verbose mode.
This is the entry in log file for one of the file which failed to update.

MSI (s) (60:8C) [09:48:51:735]: Executing op: FileCopy(SourceName=SAMPLE~1.TXT|SampleTextFile.txt,SourceCabKey=_897AF91AADD144A4A32B738927E9B501,DestName=SampleTextFile.txt,Attributes=512,FileSize=6608,PerTick=32768,,VerifyMedia=1,,,,,CheckCRC=0,,,InstallMode=58982400,,,,,,,)
MSI (s) (60:8C) [09:48:51:735]: File: C:\Program Files\Sample Application\SampleTextFile.txt; Won't Overwrite; Won't patch; Existing file is unversioned but modified

The log tells that SampleTextFile.txt as “Existing file is unversioned but modified”, hence Windows Installer “Won't Overwrite; Won't patch;”.

Ouch, SampleTextFile.txt has been modified after it was installed, hence my upgrade MSI is not updating this file. What can I do to solve this?

After googling got to know, there are no direct solutions inside Visual Studio (atleast to my knowledge in VS 2008). Here are the steps to follow to solve this issue.

Download ORCA, a free utility to edit the MSI packages
Open the MSI file in ORCA editor
Look for Property table under Tables
Add a new property named REINSTALLMODE and set its value as amus
Save the MSI and close ORCA

That's It. This property REINSTALLMODE with value amus tells the Windows Installer to uninstall old files and install new files irrespective of rules (similar to VS 2005).

Here are the screen shots of ORCA editor.

HTML5 - Canvas

2011-05-30T15:54:00.001+05:30

Recently got a chance to learn some of the new HTML5 standards. There are lot of new features supported out of box by the browsers adapting the HTML5 syntax.

DIVE INTO HTML5 is a wonderful book/tutorial on HTML5. This article, lists out the top 28 features one must know before coding in HTML5.

There are many new element tags introduced by HTML5 standards. As per HTML5 standards, Canvas element is defined “a resolution-dependent bitmap canvas which can be used for rendering graphs, game graphics, or other visual images on the fly”. Along with strong JavaScript support, Canvas is one of the best competitor to RIA development tools like Flash & Silverlight.

To start with HTML5 learning, I thought of migrating my MIX 10 10K coding puzzle developed in Silvelight to HTML5 Canvas element. Below is the puzzle displayed using Canvas element.

Your browser does not support iframes.

The above puzzle is displayed by the below given source code.

<!DOCTYPE html>
<html>
<head>
    <title>Lights On</title>
    <!--[if IE]>
      <script src="http://explorercanvas.googlecode.com/svn/trunk/excanvas.js"></script>
    <![endif]-->
    <script src="http://ajax.googleapis.com/ajax/libs/jquery/1.6.1/jquery.min.js"></script>
    <style>
      #container { position: relative; background-color:#005B96; width:550px; height:280px; }   
      #surface { position: relative; background-color:#ffffff; margin:9px; }   
      #title {margin-left:10px; font-family:Verdana; font-size:smaller; color:White; font-weight:bold}
      #description {float:left; margin-left:10px; font-family:Verdana; font-size:smaller; color:White; font-weight:lighter;}
      #reset {float:left; margin-left:10px; font-family:Verdana; font-size:smaller; font-weight:lighter;}
      #clickCount {margin-left:5px; font-family:Verdana; font-size:smaller; font-weight:lighter; color:White; vertical-align:bottom;}
      #clickCounter {margin-left:5px; font-family:Verdana; font-size:smaller; font-weight:lighter; color:White; vertical-align:bottom;}
    </style>
</head>
<body>
    <div id="container">
        <table>
            <tr>
                <td style="width:50%; vertical-align:top;">
                    <span id="title">Turn On the lights in all the cells.</span><br><br>
                    <span id="description">Clicking a cell will Turn On / Off its light, as well as its Horizontal & Vertical adjacent cells.</span><br><br><br><br><br><br><br><br><br><br>
                    <input type="button" id="reset" value="Restart" onclick="reset();" />
                    <span id="clickCount">No of Clicks: </span>
                    <span id="clickCounter">0</span>
                </td>
                <td style="width:50%; vertical-align:middle;">
                    <canvas id="surface">Canvas is not supported</canvas> 
                </td>
            </tr>
        </table>        
    </div>
    <script>
    var rowCount;
    var columnCount;
    var blockWidth;
    var blockHeight;
    var padding;    
    var canvasWidth;
    var canvasHeight;
    var ctx;
    var blocks;
    var canvasLeft;
    var canvasTop;
    var clickCount;
    function init() {
        rowCount = 5;
        columnCount = 5;
        blockWidth = 50;
        blockHeight = 50;
        padding = 1;
        clickCount = 0;
        blocks = new Array(rowCount);
        for (i = 0; i < rowCount; i++) {
            blocks[i] = new Array(columnCount);
            for (j = 0; j < columnCount; j++) {
                blocks[i][j] = 0; // Set as not lighted
            }
        }
        
        ctx = $('#surface')[0].getContext("2d");
        canvasWidth = ((blockWidth + padding) * columnCount) + padding;
        ctx.canvas.width = canvasWidth;
        canvasHeight = ((blockHeight + padding) * rowCount + padding);
        ctx.canvas.height = canvasHeight;
        draw();           
        
        canvasLeft = $("#surface").offset().left;
        canvasTop = $("#surface").offset().top;
        $("#clickCounter").text(clickCount); 
    }
    function rect(x, y, w, h) {
        ctx.beginPath();
        ctx.rect(x, y, w, h);
        ctx.closePath();
        ctx.fill();
    }
    function clear() {
        ctx.clearRect(0, 0, canvasWidth, canvasHeight);
    }
    function reset() {
        init();
    }
    function draw() {
        //draw blocks
        for (i = 0; i < rowCount; i++) {
            for (j = 0; j < columnCount; j++) {
                if (blocks[i][j] == 0) {
                    ctx.fillStyle = "#D3D3D3";
                }
                else {
                    ctx.fillStyle = "#FF0000";
                }
                rect((j * (blockWidth + padding)) + padding, (i * (blockHeight + padding)) + padding, blockWidth, blockHeight);
            }
        }
    }
    function onClick(e) {
        var x = Math.floor((e.pageX - canvasLeft) / blockWidth);
        var y = Math.floor((e.pageY - canvasTop) / blockHeight);
        var lightState;
        lightState = blocks[y][x] == 1 ? 0 : 1;
        blocks[y][x] = lightState;
        if (y - 1 >= 0) {
            lightState = blocks[y - 1][x] == 1 ? 0 : 1;
            blocks[y - 1][x] = lightState;
        }
        if (x - 1 >= 0) {
            lightState = blocks[y][x - 1] == 1 ? 0 : 1;
            blocks[y][x - 1] = lightState;
        }
        if (y + 1 < rowCount) {
            lightState = blocks[y + 1][x] == 1 ? 0 : 1;
            blocks[y + 1][x] = lightState;
        }
        if (x + 1 < columnCount) {
            lightState = blocks[y][x + 1] == 1 ? 0 : 1;
            blocks[y][x + 1] = lightState;
        }
        draw();
        clickCount += 1;
        $("#clickCounter").text(clickCount);       
    }        
    $(document).ready(function() {
        init();
        $("#surface").click(onClick);
    });
    </script>       
</body>
</html>

Embedding SilverLight XAP in Blogger

2011-01-27T11:57:00.002+05:30

I just thought of showing my simple "Lights On" puzzle in my blog. I hosted my XAP file in Google sites and tried adding it in <object> tag, like adding in a normal HTML file as shown below.

<object id="SL" data="data:application/x-silverlight," type="application/x-silverlight-2" 
style="width: 550px; height: 280px">
<param name="source" value="https://sites.google.com/site/rajganeshmountbatton/home/blogspotdocuments/25SquarePuzzle.xap" />
<param name="minRuntimeVersion" value="3.0.40624.0" />
<param name="enableHtmlAccess" value="true" />
<param name="initParams" value="someParam=value" />
<a href="http://go.microsoft.com/fwlink/?LinkID=149156&v=3.0.40624.0" style="text-decoration: none;">
<img src="http://go.microsoft.com/fwlink/?LinkId=108181" style="border-style: none"
alt="Get Microsoft Silverlight" />
</a>
</object>

The above code looks good and every one expects it to work fine. But in real, it just displays a white blank SilverLight application. This is an expected behavior of SilverLight and Timheuer talks about the solution in this link.

The solution is to map proper XAP MIME Type, adding few changes in AppManifest.xml file and adding the below param tag inside the object tag as shown below.

Below code block is the change to be done in AppManifest.xml file.

<Deployment xmlns="http://schemas.microsoft.com/client/2007/deployment"
xmlns:x="http://schemas.microsoft.com/winfx/2006/xaml"
ExternalCallersFromCrossDomain="ScriptableOnly">
<Deployment.Parts>
</Deployment.Parts>    
</Deployment>

Param change in object tag

<param name="enableHtmlAccess" value="true" />

AppManifest and Param changes are in our hands and can be modified, but mapping proper MIME type for SilverLight application in Google site cannot be done. Other options are like hosting your SilverLight application in some other MIME configurable hosting provider. But there are no better free SilverLight hosting provider available. Any other options? I heard about Google app engine. So thought of trying to use app engine as host.

Here is good post about embedding SilverLight in app engine. I followed the same and hosted my application in app engine. Then changed the source attribute in the object tag to the app engine URL. What a surprise, the application worked like charm.

Below is the code after hosting in the Google App Engine.

<object id="SL" data="data:application/x-silverlight," type="application/x-silverlight-2" 
style="width: 550px; height: 280px">
<param name="source" value="http://rajganeshsilverlight.appspot.com/25SquarePuzzle.xap" />
<param name="minRuntimeVersion" value="3.0.40624.0" />
<param name="enableHtmlAccess" value="true" />
<param name="initParams" value="someParam=value" />
<a href="http://go.microsoft.com/fwlink/?LinkID=149156&v=3.0.40624.0" style="text-decoration: none;">
<img src="http://go.microsoft.com/fwlink/?LinkId=108181" style="border-style: none"
alt="Get Microsoft Silverlight" />
</a>
</object>

Here is the link to my hosted SilverLight application blog. Enjoy the Light On application and try lighting up 25 squares.

Bing Maps Developer API

2010-09-01T23:30:00.000+05:30

Good link in MSDN about the Bing Map API.

http://msdn.microsoft.com/en-us/library/dd877180.aspx

Visual Studio Keybinding Posters

2010-08-01T16:17:00.001+05:30

Visual Studio team has released the shortcut keys for Visual Studio 2010 in various PDF formats.

Here is the download link

A Small Silverlight Puzzle for MIX10 10K Coding Challenge

2010-01-30T18:12:00.014+05:30

My contribution of a Small Silverlight Puzzle for MIX10 10K Coding Challenge.

http://mix10k.visitmix.com/Entry/Details/225

Will upload the code once the contest completes.

Update 1: After a long break have updated the code used in this small application. You can download it here

Update 2: Tried to show that small application in Blogger, after couple of try able to show it here. Try to solve it, if you can...

Searching Stored Procedures & Functions based on the text

2009-04-13T12:02:00.004+05:30

The below script is used to get the list of Stored Procedures based on their text.

SELECT * FROM INFORMATION_SCHEMA.ROUTINES WHERE ROUTINE_TYPE = 'PROCEDURE' AND ROUTINE_DEFINITION LIKE 'Text To Be Searched'

To search for the functions, the ROUTINE_TYPE needs to be changed to FUNCTION as shown below.

SELECT * FROM INFORMATION_SCHEMA.ROUTINES WHERE ROUTINE_TYPE = 'FUNCTION' AND ROUTINE_DEFINITION LIKE 'Text To Be Searched'

Accessing the GAL (Global Address List) from ASP.NET

2008-12-05T15:56:00.004+05:30

Yesterday I have written an article in CodeProject. The article's link is

http://www.codeproject.com/KB/exchange/AccessGAL.aspx

Migrating Visual Studio.NET project from VisualSourceSafe to ClearCase while retaining history

2008-09-10T09:06:00.013+05:30

This article by IBM has a complete description of migrating the .NET project which is currently under the VSS to ClearCase with history. But still it doesn't much explain about exporting the project / solution from VSS and importing back to the ClearCase by using the clearexport_ssafe and clearimport. The below given section explains them better.

To export the project's / solution's information along with history we need to use the clearexport_ssafe command in the DOS prompt. This article explains about the same, but still it doesn't have any sample to explain. The below given sample batch file explains it better.

This is the batch file commands, which I used when migrating one of my .net project from VSS to ClearCase.

SET PATH=%PATH%;C:\Program Files\Microsoft Visual SourceSafe;(The loaction of the SS.exe in the machine)
SET SSDIR=\\NetwrokVSSDatabaseLocation\VSS DatabaseName (eg., \\TestLocation\TestDatabaseName)
SET SSUSER= VSS User Name (This user should be created in VSS database with blank password)
SS CP $/VSSProjectLocation (eg., $/TestProject)
clearexport_ssafe -r

Running the above batch file, will generate a file in the same folder with the name "cvt_data". This is the file having the complete VSS project data with history.

Then while importing the data back to ClearCase, the command used is

clearimport -pcase "cvt_data file location"

This command should be executed in DOS prompt in the location of VOB where you need to move your data. Consider your project's VOB is mapped to Z: directory in your machine and your project's VOB name is TestVOB. Then you chould first change the directory to "Z:\TestVOB\FolderName". Then execute the above command with the correct CVT_Data file location.

In the above command the switch -pcase is used to tell import utility to preserve the case of the files while importing. If this attribute is missed, by default clearimport will import the files with all lowercase letters.

If you are not the VOB owner, then running the above command will give you a "Permission Denied" error. To avoid this run the command with the switch -nsetevent, like shown below

clearimport -nsetevent -pcase "cvt_data file location"

This will import the files exported using the "clearexport_ssafe" command. Now your VOB would be having the complete VSS project's data along with history.

References:

http://www-01.ibm.com/support/docview.wss?rs=0&uid=swg21222507
http://publib.boulder.ibm.com/infocenter/cchelp/v7r0m0/index.jsp?topic=/com.ibm.rational.clearcase.books.cc_admin.doc/import-export.htm
http://www.astrainfotech.com/rsdc-2006-dilip.pdf

Label Chart Widget for blogger - Using Google Charts API

2008-07-29T14:37:00.011+05:30

Are you interested in showing your blogs label count as a chart, like the one available in my blog? Then follow the below and you can achieve it by some scripts and the Google Chart API.

If you are not familiar with Google Chart API, then check it here.

Then the script for the Widget is as follows.

<b:widget id='LabelChart' locked='false' title='Label Chart' type='Label'>
<b:includable id='main'>
<b:if cond='data:title'>
<h2><data:title/></h2>
</b:if>
<div class='widget-content'>
<div id='LabelChartDiv'/>
<script type='text/javascript'>
var LabelArray = new Object;
var i=0;
<b:loop values='data:labels' var='label'>
var LabelName = "<data:label.name/>";
LabelArray[LabelName] = <data:label.count/>;
i++
</b:loop>
var LabelChartDivObj = document.getElementById('LabelChartDiv');
var ChartLocation = 'http://chart.apis.google.com/chart?chs=230x75&chd=t:';
var j=0;
for (LabelCount in LabelArray)
{
j++;
if(j<i)
{
ChartLocation = ChartLocation + LabelArray[LabelCount] + ',';
}
else
{
ChartLocation = ChartLocation + LabelArray[LabelCount];
}
}
ChartLocation = ChartLocation + '&cht=p&chl=';
j=0;
for (LabelCount1 in LabelArray)
{
j++;
if(j<i)
{
ChartLocation = ChartLocation + escape(LabelCount1) + '['+ LabelArray[LabelCount1] + ']|';
}
else
{
ChartLocation = ChartLocation + escape(LabelCount1) + '['+ LabelArray[LabelCount1] + ']';
}
}
LabelChartDivObj.innerHTML = '<img src=' + ChartLocation + '>';
</script>
</div>
</b:includable>
</b:widget>

To add the above widget script, you need to go to your blogger 'Dashboard'. Then go for 'Layout' and 'Edit HTML' as shown in the below image.

Then go for 'Edit Template' section and check the 'Expand Widget Templates' checkbox.

Now copy the above script and paste it in your template code near your normal label widget. You can find this by searching the below script.

<b:widget id="'Label1'" locked="'false'" title="'Labels'" type="'Label'">

Note: ID of the widget will not be same, it can be different.

The Label widget will be having some standard code and will be end with </b:widget>.

Now you paste the above copied code below this end line and save your template.

Thats all. You have made a colorful chart displaying your label counts.

Caution: Before editing your template, always take a copy of your old template. If some thing goes wrong while editing you can always revert back.

ASP .NET AJAX Timer control : Processing client script on Tick event

2008-07-28T10:28:00.013+05:30

Timer Control is a very handy ASP.NET AJAX Extension control used to 'Performs asynchronous or synchronous Web page postbacks at a defined interval'.

This control has good set of features associated with it. But still i feel there is no option to process the client scripts on the Timer's Tick event.

The below given script explains a workaround to the above issue.

// Access the Timer control.

var TestTimer = $find('<%= TestTimer.ClientID %>');

/* Override the '_doPostback' (method called to fire the server Tick event) 

    method with your own function. */

TestTimer._doPostback = TimerTick;        

// Overridden Timer Tick Function

function TimerTick()

{

    // Here this is called on every timer tick event..

    // So do your client processing here..

    alert('Timer Client Click');

    // Finally do the async postback..

    var ClientTimerId = '<%= tAutoSave.ClientID %>';

    // Now call the Asyncpost back method..

    // Replace the client id's with the Unique Id..

    __doAsyncPostBack(TestTimer, ClientTimerId.replace(/\_/g, '$'),'');

}

// Method to do async post back..

function __doAsyncPostBack(sourceElement, eventTarget, eventArgument)

{

    // Here we need to do an async post back...

    // Get the Form

    var form = Sys.WebForms.PageRequestManager.getInstance()._form;

    // Set Event target and the event arguments..

    form.__EVENTTARGET.value = eventTarget;

    form.__EVENTARGUMENT.value = eventArgument;

    // Set the async post back as true..

    Sys.WebForms.PageRequestManager.getInstance()._postBackSettings.async = true;

    // Set source element..

    Sys.WebForms.PageRequestManager.getInstance()._postBackSettings.sourceElement = sourceElement;

    // Call the form submit..

    Sys.WebForms.PageRequestManager.getInstance()._onFormSubmit();

}

Using CTE and Row_Number() to get the duplicate data in SQL Server 2005

2008-04-08T18:52:00.007+05:30

This is the script to get the duplicate data from employee table, whose employee name is repeating.

With TestData as

(

Select

EmployeeId,

Row_Number() Over( Partition By EmployeeName Order By EmployeeName) AS Rowid

From

Employees

)

Select EmployeeId from TestData Where RowId > 1

If we need to get the duplicate records by two columns, those two columns should be in the Partition By column, and in the order by column.

Using CTE to write recursive query in SQL Server 2005

2008-04-08T15:11:00.003+05:30

CTE (Common Table Expression) is newly introduced in SQL Server 2005. The below given link elaborates the CTE and its advantages.

http://www.4guysfromrolla.com/webtech/071906-1.shtml

http://searchwindevelopment.techtarget.com/tip/0,289483,sid8_gci1278207,00.html

Basically the requirement was, when an employee logs in he needs the list of employees reporting directly under him as well as the employees who are reporting under them.

The database structure was like:

Employee table containing the employee details like Employee Id, Name, etc.,
Supervisor table containing the mapping of employee and their supervisor

I tried out with the CTE to build a recursive function which will get the first level reporting employees and then their reporting employee. The script was given below

Declare@UserId Int

Set @UserId = 36332

-- This is the recursive function..

WITH EmployeeHierarchy (EmployeeID, EmployeeName, HierarchyLevel) AS

(

-- Base case

Select

E.EmployeeId,

E.Name As EmployeeName,

1 As HierarchyLevel

From

Employees E

Where

E.EmployeeId = @UserId

UNION ALL

-- Recursive step

select

E.EmployeeId,

E.Name As EmployeeName,

eh.HierarchyLevel + 1 AS HierarchyLevel

FROM

Supervisors ES

INNER Join Employees E On E.EmployeeId = ES.EmployeeId INNER JOIN EmployeeHierarchy eh ON ES.Supervisor1 = eh.EmployeeID

)

-- This is the final select

Select

E.EmployeeId,

E.EmployeeNumber As EmployeeNumber,

E.EmployeeName

From

EmployeeHierarchy E

Where

E.EmployeeId <> @UserId

Convert Decimal to Hours & Minutes in C# .NET

2008-03-28T14:24:00.010+05:30

Here is a simple code to convert a decimal value to its equivalent hour and minute value.

double Value = 1.75;

double Hours = Math.Floor(Value);

double Minutes = (Value % 1 * 100);

Minutes = Math.Round(Minutes, 2);

if(Minutes >= 60)

{

Minutes = Math.Floor(Minutes/ 60) + (Minutes % 60 / 100);

Minutes = Math.Round(Minutes,2);

}

else

{

Minutes = Math.Round(Minutes / 100, 2);

}

string HoursAndMinutes = Convert.ToString((Hours + Minutes));

Using SMO to generate the script of triggers in SQL server 2005

2008-03-27T14:57:00.012+05:30

The below is the code written in C# to get the scripts of the triggers, stored procedures, views and user defined functions available in the SQL Server database.

This is done using the SMO class exposed by the microsoft. To use SMO in our C# code we need to reference to the Microsoft's SMO dll. This dll can be found in the C:\Program Files\Microsoft SQL Server\90\SDK\Assemblies directory. You'll need to add references to dll as shown below

using Microsoft.SqlServer.Management.Smo;

This is the code which we can add as a method to any of the existing class file

Server oServer = new Server();

try

{

oServer.ConnectionContext.ServerInstance = "TestInstance";

oServer.SetDefaultInitFields(typeof(StoredProcedure), "IsSystemObject");

oServer.ConnectionContext.LoginSecure = false;

oServer.ConnectionContext.Login = "sa";

oServer.ConnectionContext.Password = "password";

oServer.ConnectionContext.Connect();

TextWriter oTextWriter = null;

StringBuilder oSBText = null;

foreach (Database oDB in oServer.Databases)

{

if (oDB.Name.ToUpper().Equals("DB1"))

{

oTextWriter = new StreamWriter(@"D:\DBScripts\DbScripts1.sql");

}

else if (oDB.Name.ToUpper().Equals("DB2"))

{

oTextWriter = new StreamWriter(@"D:\DBScripts\DbScriptss.sql");

oSBText = new StringBuilder();

}

if (oTextWriter != null)

{

foreach (Table oTable in oDB.Tables)

{

TriggerCollection oTriggerCollect = oTable.Triggers;

if (oTriggerCollect.Count > 0)

{

oSBText.Append("\n" + "--" + oDB.Name + "\n" + "--" + oTable.Name);

foreach (Trigger oTrg in oTriggerCollect)

{

try

{

oSBText.Append("\n" + oTrg.TextHeader + "\n" + oTrg.TextBody);

}

catch (Exception Ex)

{

TextWriter oErrorWriter = new StreamWriter(@"D:\DBScripts\Error\Error.txt");

oErrorWriter.Write(Ex.Message);

oErrorWriter.Close();

}

foreach (StoredProcedure oSP in oDB.StoredProcedures)

{

try

{

if (oSP.IsSystemObject == false)

{

oSBText.Append("\n" + "--" + oDB.Name + "\n" + "--" + oSP.Name + "\n");

StringCollection sc = oSP.Script();

foreach (string s in sc)

oSBText.Append("\n" + s);

}

catch (Exception Ex)

{

TextWriter oErrorWriter = new StreamWriter(@"D:\DBScripts\Error\Error.txt");

oErrorWriter.Write(Ex.Message);

oErrorWriter.Close();

}

foreach (View oV in oDB.Views)

{

try

{

oSBText.Append("\n" + "--" + oDB.Name + "\n" + "--" + oV.Name + "\n");

StringCollection sc = oV.Script();

foreach (string s in sc)

oSBText.Append("\n" + s);

}

catch (Exception Ex)

{

TextWriter oErrorWriter = new StreamWriter(@"D:\DBScripts\Error\Error.txt");

oErrorWriter.Write(Ex.Message);

oErrorWriter.Close();

}

foreach (UserDefinedFunction oUDF in oDB.UserDefinedFunctions)

{

try

{

oSBText.Append("\n" + "--" + oDB.Name + "\n" + "--" + oUDF.Name + "\n");

StringCollection sc = oUDF.Script();

foreach (string s in sc)

oSBText.Append("\n" + s);

}

catch (Exception Ex)

{

TextWriter oErrorWriter = new StreamWriter(@"D:\DBScripts\Error\Error.txt");

oErrorWriter.Write(Ex.Message);

oErrorWriter.Close();

}

oTextWriter.Write(oSBText.ToString());

oTextWriter.Close();

oTextWriter = null;

oSBText = null;

}

Console.WriteLine("Successfully Written to the file!!!");

}

catch (Exception Ex)

{

TextWriter oErrorWriter = new StreamWriter(@"D:\DBScripts\Error\Error.txt");

oErrorWriter.Write(Ex.Message);

oErrorWriter.Close();

}

finally

{

if (oServer.ConnectionContext.IsOpen)

{

oServer.ConnectionContext.Disconnect();

}

To Learn more about SMO follow the below given links.

http://msdn2.microsoft.com/hi-in/magazine/cc163409(en-us).aspx

http://www.sqlteam.com/article/scripting-database-objects-using-smo-updated

Masterpage – document.getElementById Issue

2007-12-03T14:29:00.002+05:30

Masterpage is one of the good features of the ASP.NET 2.0. But if you are a beginner in .NET 2.0 and Masterpage, you would stuck up to get the element inside a Masterpage from a JavaScript code.

The issue was, whenever a control is inside a Masterpage the client id of the control would get append with the content placeholder id. So for an element of id “txtTest” the client id would be something like “ctl00_ContentPlaceHolder1_ txtTest”.

So when you use the document.getElementById(‘txtTest’), you will not get the access of the txtTest textbox in JavaScript. Indeed you need to access it by calling document.getElementById(‘ctl00_ContentPlaceHolder1_ txtTest’).

To avoid this we can access the control by using the document.getElementById('<%=txtTest.ClientID%>'). This will give you the access to txtTest text box. Now this will work fine until and unless the script is in line with the aspx page ie., if the script is included as a part of the aspx page. But the same won’t work if you have this script in a separate .js file, and add it to the aspx page.

So in this scenario to get the access to the control, we need to hard code the control’s id. But hard coding is not an ideal way of coding. To avoid this situation what we can do is to, maintain the mapping between the client id and the server id. This can be achieved as shown below.

First we need to declare two arrays and the first array will have the server ids of the required controls, and the second array will have the client ids of those server controls in the same order. Register these two arrays to the client side. Now create a JavaScript function, which will accept the server id and will compare the server id with the available server ids in the array and will give its position in the array. Then the same function would return you the matching client id in the same location of the client array. So we were able to get the mapping client id for the given server id.

The code below shows the declaration of the array and the declaration of the JavaScript function.

public void RenderJSArrayWithCliendIds(params Control[] wc)

{
if (wc.Length > 0)
{
StringBuilder arrClientIDValue = new StringBuilder();
StringBuilder arrServerIDValue = new StringBuilder();
// Get a ClientScriptManager reference from the Page class.
ClientScriptManager cs = Page.ClientScript;
for (int i = 0; i < wc.Length; i++)
{
arrClientIDValue.Append("\"" + wc[i].ClientID + "\",");
arrServerIDValue.Append("\"" + wc[i].ID + "\",");
}

cs.RegisterArrayDeclaration("MyClientID", arrClientIDValue.ToString().Remove (arrClientIDValue.ToString().Length - 1, 1));
cs.RegisterArrayDeclaration("MyServerID", arrServerIDValue.ToString().Remove(arrServerIDValue.ToString().Length - 1, 1));
cs.RegisterStartupScript(this.Page.GetType(), "key", "\nfunction GetClientId(serverId)\n{\nfor(i=0; i<MyServerID.length; i++)\n{\nif ( MyServerID[i] == serverId )\n{\nreturn MyClientID[i];\nbreak;\n}\n}\n} ", true);
}
}

We can make these codes as part of a common class for all the UI pages, so that in each page if need to access any one control in the JavaScript we can simply call the below function with the parameter as the controls to be accessed.

// Here give the coma seperated controls for which the client ids should be rendered
RenderJSArrayWithCliendIds(txtTest,lblTest);

This is one of the simplest ways to have a mapping between the client and the server id. By this way no need to worry about accessing a control inside a MasterPage from an external javascript files.

After this codes we can access the elements which are given in the RenderJSArrayWithCliendIds function in the javascript as shown below

// Get the text box
var txtTest = document.getElementById(GetClientId("txtTest "));

This will slove the issues arrised due to the change in the client id of the control which is placed inside a MasterPage.

Updates: I am not much experienced in VB, but still tried to translate the C# code to VB. Just check the below code.

Public Sub RenderJSArrayWithCliendIds(ByVal wc As Control())
If (wc.Length > 0) Then
Dim arrClientIDValue As StringBuilder = New StringBuilder()
Dim arrServerIDValue As StringBuilder = New StringBuilder()
Dim cs As ClientScriptManager = Page.ClientScript
Dim frm As HtmlForm = Page.Form
Dim i As Integer
For i = 0 To wc.Length
arrClientIDValue.Append("\"" + wc[i].ClientID + " \ ",")
arrServerIDValue.Append("\"" + wc[i].ID + " \ ",")
i = i + 1
Next
cs.RegisterArrayDeclaration("MyClientID", arrClientIDValue.ToString().Remove(arrClientIDValue.ToString().Length - 1, 1))
cs.RegisterArrayDeclaration("MyServerID", arrServerIDValue.ToString().Remove(arrServerIDValue.ToString().Length - 1, 1))
cs.RegisterStartupScript(Page.GetType(), "key", "\nfunction GetClientId(serverId)\nBegin\nfori=0; i<MyServerID.legth; i++)\nBegin\nif ( MyServerID[i] == serverId )\nBegin\nreturn MyClientI[i];\nbeak;\nEnd\nEnd\nEnd\n \nfunction GetObject(serverId)\nBegin\nfori=0; i<MyServerID.legth; i++)\nBegin\nif ( MyServerID[i] == serverId )\nBegin\nreturn document.all[MyClientIDi]];\nbeak;\nEnd\nEnd\nEnd\n", True)
End If
End Sub

parseInt - Javascript issue

2007-10-19T08:20:00.000+05:30

What would be the result of the following javascript

var value = "08";
value = parseInt(value);
alert(value);

When we just go through the above script, we would expect the result as "8" should be alerted. But to all our surprise the result would be "0".

Let us analyze how this happened.

when ever the browser tries parseInt("08"), zero in front tells the browser that this is an octal (base 8) number and hence will be returned as "0".

This is the default behavior of the parseInt function, to read more about the function click here.

So what we need to do in this case?

If we just check the intellisence result for the parseInt, it will be like

So it accepts another one argument called radix, and the definition of radix is as follows

"The radix parameter is used to specify which numeral system to be used, for example, a radix of 16 (hexadecimal) indicates that the number in the string should be parsed from a hexadecimal number to a decimal number."

And if the radix parameter is not supplied, browser assumes the following:

If the string begins with "0x", the radix is 16 (hexadecimal)
If the string begins with "0", the radix is 8 (octal). This feature is deprecated
If the string begins with any other value, the radix is 10 (decimal)

So in our case, browser assumed the radix as 8 and returned a value of "0". To solve this issue whenever we are calling the parseInt we should explicitly specify the radix value.

XML PARSE ERROR: Missing end-tag - Error in Export to Excel

2007-10-18T09:06:00.001+05:30

Last week i was debugging a strange problem which came in the export to excel option in one of our web page. We have a export method which, will accept the dataset and will loop through it and builds the XML schema for the excel.

When we were using this excel method to export some data, it went fine and the response rendered the Open or Save dialog box. But when we try to open the excel file it gave a error stating the following

Problem came up in following areas during load:

Table

This file cannot be opened because of errors. Errors are listed in: C:\Documents and Settings\UserId\Local Settings\Temporary Internet Files\Content.MSO\90797611.log

And when we open the specified 90797611.log file the message it had was,

XML PARSE ERROR: Missing end-tag
Error occurs at or below this element stack:
<ss:Workbook>
<ss:Worksheet>
<ss:Table>
<ss:Row>
<ss:Cell>
<ss:Data>
<ss:Font>
<ss:B>
<ss:I>

By looking in to this i was not able to get any thing, after doing some R&D I came to a conclusion that there were some problem with regarding "<ss:Font> <ss:B> <ss:I>".

Finally i decided to have a look at the dataset which is supplied to the export method, and to my surprise there was a field with the following data in it

<Font colour="green"><B><I>7358</B></I></Font>

As we are building a XML, the <Font colour="green"> got treated as a XML tag and it was expecting a corresponding end tag. This created the whole problem.

After removing <Font colour="green"><B><I> from the data, every thing worked fine.