To begin I set a CSS class on the error messages that were displayed in the td with the form element. An error message could be in the form of

<span class=".valError">You have selected an invalid date.</span>

Now all we have to do in jQuery is find all elements which match the class, remove any classes that may be on any of the td tags, then add a class to the tr in which the error exists.

$(".valError").each(function() {
    // remove all classes from any td's of the below the parent row
    $(this).parent().parent().find("td").removeClass();
    // set the class on the tr
    $(this).parent().parent().addClass('ms-informationbar');
})

That’s it.

You could optionally set the class of the td rather than or as well as the tr. The following line would help you achieve this.

$(this).parent().parent().find("td").removeClass().addClass('myErrorClass');

For a while I have been trying to work out how I could export Crystal Reports in various formats such as Excel and PDF without having to render the report in the web browser first. Some of our customers wanted the data to put into their spreadsheet application so they could manipulate the data to their hearts content and I wanted to still use the power of Crystal Reports as it meant that the people who are good at writing the reports could do it and save me the work.

I’d seen it done in Crystal 8 but had not seen anything that pulled it all together in XI. I think I was looking a little too hard, the solution was in fact pretty straight forward.

The method employed is pretty similar to the post I made on displaying Crystal Reports in the browser, with only a couple of exceptions.

Firstly, you don’t need to employ the ReportViewer control as the report is not being rendered to the browser.

// Declare variables
private ReportDocument document1 = new ReportDocument();
private TableLogOnInfos infos1 = new TableLogOnInfos();
private TableLogOnInfo info2 = new TableLogOnInfo();
private ConnectionInfo info1 = new ConnectionInfo();
 
//  default export type
ExportFormatType expFormat = new ExportFormatType();
expFormat = ExportFormatType.PortableDocFormat;
 
// load the report
document1.Load(Request.PhysicalApplicationPath + “myReport.rpt”);
 
// Add the parameters
document1.SetParameterValue(“param1″, “value1″);
document1.SetParameterValue(“param2″, “value2″);
 
//Set the login info for the database
info1.ServerName = ConfigurationSettings.AppSettings[“server_name”];
info1.DatabaseName = ConfigurationSettings.AppSettings[“database_name”];
info1.UserID = ConfigurationSettings.AppSettings[“user_id”];
info1.Password = ConfigurationSettings.AppSettings[“password”];
Tables tables1 = document1.Database.Tables;
 
//Apply login info to the report
foreach (CrystalDecisions.CrystalReports.Engine.Table table1 in tables1)
{
    info2 = table1.LogOnInfo;
    info2.ConnectionInfo = info1;
    table1.ApplyLogOnInfo(info2);
}
 
// Export the document
HttpResponse Response = HttpContext.Current.Response;
document1.ExportToHttpResponse(expFormat,Response,true,”Exported Report”);
document1.Close();

It’s worth mentioning that I don’t use the Crystal Reports extensions for Visual Studio, I have Crystal Server running on the box. I can’t see how this method wouldn’t work through the Visual Studio extensions.

Code Infront

Register the assembly in the top of your aspx page

<%@ Register TagPrefix="cr" Namespace="CrystalDecisions.Web" 
Assembly="CrystalDecisions.Web, Version=11.0.3300.0, Culture=neutral, PublicKeyToken=692fbea5521e1304" %>

Place the control where you want it to appear in the page.

<cr:crystalreportviewer id="crReport1" runat="server" hasviewlist="False"
    hastogglegrouptreebutton="False" hassearchbutton="False"
    hasgotopagebutton="False" hasdrillupbutton="False" hascrystallogo="False"
    autodatabind="true" width="350px" displaygrouptree="False"
    height="50px" hyperlinktarget="_blank">

Code Behind

// declare the variables
protected CrystalReportViewer crReport1;
private ReportDocument document1 = new ReportDocument();
private TableLogOnInfos infos1 = new TableLogOnInfos();
private TableLogOnInfo info2 = new TableLogOnInfo();
private ConnectionInfo info1 = new ConnectionInfo();
 
// load the report
document1.Load(Request.PhysicalApplicationPath + "myReport.rpt");
 
// Add the parameters
document1.SetParameterValue("param1", "value1");
document1.SetParameterValue("param2", "value2");
 
//Set the login info for the database
info1.ServerName = ConfigurationSettings.AppSettings["server_name"];
info1.DatabaseName = ConfigurationSettings.AppSettings["database_name"];
info1.UserID = ConfigurationSettings.AppSettings["user_id"];
info1.Password = ConfigurationSettings.AppSettings["password"];
Tables tables1 = document1.Database.Tables;
 
//Apply login info to the report
foreach (CrystalDecisions.CrystalReports.Engine.Table table1 in tables1)
{
    info2 = table1.LogOnInfo;
    info2.ConnectionInfo = info1;
    table1.ApplyLogOnInfo(info2);
}
 
// Bind the report
crReport1.ReportSource = document1;
crReport1.DataBind();

You also need to remember to close the document after it has been rendered to free up memory. This should ideally be done in the Page_Unload method. See this post on MSDN for more information on why it should be placed in Page_Unload.

I’ve used Subversion for years and didn’t really want to move to Visual SourceSafe. Subversion does everything I need it to and I’ve never felt the need to move and be locked in to anything else. Unfortunately running windows meant I needed to find a way of installing Subversion on a windows box and running it as a server, I didn’t want to run Apache just for the sake of it. So here it is, my install technique for installing, running and managing repositories on Windows. One point to make, is that this guide uses Tortoise SVN to manage the repository. I could probably have done it just as easily from the command line but I have Tortoise installed so why make life hard.

1. Download and install the latest version of SVN setup.
2. Create a directory on the c: drive to hold the repository (i.e c:\svnrepo)
3. Extract SVNService.exe to the same directory as svnserve.exe
4. Install SVNService as a service using the following command SVNService -install -d -r c:\svnrepo
5. Start the service from Windows Manage Computer

Creating a repository

1. Create folder inside your SVN Repository (c:\svnrepo) in this example we’ll create folder called test.
2. Create a temporary drive locally containing 3 folders; trunk, branches, tags
3. Right click on the parent folder and got to TortoiseSVN > Import
4. In the URL box at the top enter the path to the repository, i.e. svn://<server>/test
5. This will import the 3 folders into the chosen repository.

Adding a Project to a repository

If you already have some files for a project that you have started that you wish to import into your new repository, please follow these steps.

1. Browse to the parent folder of the files you wish to install.
2. Right click this folder and choose TortoiseSVN > Import.
3. This time we want to import the file into the main trunk, so the path to the repository will be svn://<server>/test/trunk

Checking out the repository.

Now we have the project imported into SVN we need to check out a local copy to work on.

1. Browse to the parent directory of where you want to checkout the project to.
2. Right click and choose SVN Checkout
3. This will bring up the Checkout window. Enter the URL of the Repository in the top box (remember we want to checkout the trunk so this needs to go at the end). The checkout directory will default to the directory we clicked in 2, add the name of the subdirectory you wish to check the files out to at the end.
4. Click OK, if the directory you specified does not exist you will be prompted to create the directory, do so.
5. All the files will be checked out into the folder you specified. There will also be hidden folders beginning with .svn created in the directories, DO NOT delete these as they are used by Subversion. You can now work on your local copy.