Blog posts by Jonas Lindau

Does your edit mode show up blank after upgrade?

Wed, 19 Sep 2018 08:04:45 GMT

I’ve seen quite a few users who ended up with just the top bar in edit mode after upgrading EPiServer lately. Most of them seems to have to do with the Visual Studio Web Accessibility Checker, or with Browser Link. For me, however, none of them solved the issue, and it took me a while … Continue reading →

Does your edit mode show up blank after upgrade?

Wed, 19 Sep 2018 06:04:45 GMT

Recycle bin won’t empty when using EPiServer Relate

Fri, 04 Nov 2016 11:43:33 GMT

A few days ago i found i strange behaviour on our intranet site. The waste basket didn’t empty! The scheduled job ran just fine, but the log always contained the same message: 0 content items were deleted from recycle bin Thats a new one! So where to look? The logs are, off course, a great … Continue reading →

Resolving links in XhtmlStrings, EPiServer 7.16.1

Mon, 21 Dec 2015 14:05:19 GMT

Did you ever try to store a XhtmlString in a variable and use it in your code? Maybe you noticed that some stuff, like links, isn’t very useful because they are internal. I noticed this when i upgraded a site from CMS6, and on that site i had a working code that rendered dynamic content … Continue reading →

Exporting lots and lots of files

Fri, 02 Oct 2015 10:14:05 GMT

Exporting lots of files from EPiServer CMS 6 can be a pain. Picking the files, one at a time, can take ages and when your done and press “Export”-button, session has probably times out and you have to do it all over again. On top of that, if you upload-folder are huge (this one is … Continue reading →

EPiServer CMS 6: Exporting lots and lots of files

Fri, 02 Oct 2015 10:14:05 GMT

EPiServer 7.5 – List icons for blocks

Fri, 17 Apr 2015 22:14:59 GMT

OK, maybe this is obvious for everyone else, but it wasn’t to me. So this is part sharing and part to remember my self. 🙂 And for the record, this may not be the most correct way, but it’s the way i figured out. What i wanted to do What i wanted to do, was … Continue reading →

EPiServer CMS 6R2, .NET 4 and Google Analytics Plugin

Tue, 17 Jun 2014 14:01:57 GMT

Do you have a site running this combination and get this screen when clicking config –> Google Analytics?

In that case, i bet you are using Page themes on your site!

The solution is to simply find this section in web.config:

   1: <location path="Modules">

And add a system.web section to it:

   1: <system.web>

   2:   <pages theme="" />

   3: </system.web>

And tada! You’re good to go!

Relate+ - Get distinct values from attributes

Fri, 14 Mar 2014 13:03:05 GMT

Recently I had to figure out a way to get all distinct values from a certain attribute in a Relate+ project. The site uses a lot of attributes on IUser to store local copies of attributes in Active Directory. These attributes are then used mainly for presentation, but know they want to filter on these attributes. Filtering on an attribute is quite easy, the issue showed to be getting distinct values for a certain attribute to for example populate a dropdown.

The only solution I found was to use LINQ on a UserCollection. It works great, but the downside might be performance. Because of this, I figured I create a InitializationModule which load all distinct values from the needed attribute and store them in the cache. Then, all I have to do is getting the list from the cache when needed.

There are many good articles on how to create InitializationModules, so I’m going to leave that part out. But here is how I get all distinct values for the attribute “OrganizationAbbreviation”:

   1: UserCollection Users = CommunitySystem.CurrentContext.DefaultSecurity.GetUsers();

   2: string[] Units = (from CurrentUser in Users where CurrentUser.GetAttributeValue<string>("OrganizationAbbreviation") != null select CurrentUser.GetAttributeValue<string>("OrganizationAbbreviation")).Distinct().OrderBy(x => x).ToArray<string>();

Now I can easily populate all my dropdown-lists with the organizational units!

Error 404 on images/files when running EPiServer CMS 6.0 on ASP.NET 4.0

Tue, 22 Oct 2013 11:10:54 GMT

I found this excellent post regarding ASP.NET 4.0 and EPiServer CMS 6:

http://world.episerver.com/Articles/Items/EPiServer-CMS-60-now-supporting-ASPNET-40/

I followed all steps and, surprise: No images, no working downloads… hmm… Luckily, it turned out to be a simple fix! The site had some handlers registered for certain file types, and they had a runtime-version specified this way:

   1: <add name="*.jpg_GET,HEAD" path="*.jpg" verb="GET,HEAD"

   2: type="EPiServer.Web.StaticFileHandler, EPiServer"

   3: preCondition="integratedMode,runtimeVersionv2.0"/>

Just remove the attribute preCondition and it works fine!

Users disappear from club’s access groups when using AD and Relate–a workaround…

Fri, 31 May 2013 13:04:01 GMT

I’ve seen a few forum posts from users having this issue. One of the best threads covering this subject is this one: http://world.episerver.com/Modules/Forum/Pages/thread.aspx?id=63028

This thread actually contains a good way to reproduce the error and also two different solutions that might work. The first solution is to specify the groups to transfer instead of using *. This worked for me at first, but suddenly started to act weird.

But since I wanted to know more about this, I started with running SQL Profiler to see who, and when, accessed the database table tblEPiServerCommonGroupUser. It turned out that what’s happening is that all rows for the current user gets deleted when the new session starts. After that, the users group belongings gets inserted back to the table, BUT it seems like it never inserts non-AD groups, like “Group for club with id: xx”. This is what happens:

   1: DELETE FROM tblEPiServerCommonGroupUser WHERE intUserID = @intUserID

   2: INSERT INTO tblEPiServerCommonGroupUser (intGroupID, intUserID) VALUES (@intGroupID, @intUserID)

   3: INSERT INTO tblEPiServerCommonGroupUser (intGroupID, intUserID) VALUES (@intGroupID, @intUserID)

   4: INSERT INTO tblEPiServerCommonGroupUser (intGroupID, intUserID) VALUES (@intGroupID, @intUserID)

   5: INSERT INTO tblEPiServerCommonGroupUser (intGroupID, intUserID) VALUES (@intGroupID, @intUserID)

   6: INSERT INTO tblEPiServerCommonGroupUser (intGroupID, intUserID) VALUES (@intGroupID, @intUserID)

   7: INSERT INTO tblEPiServerCommonGroupUser (intGroupID, intUserID) VALUES (@intGroupID, @intUserID)

   8: INSERT INTO tblEPiServerCommonGroupUser (intGroupID, intUserID) VALUES (@intGroupID, @intUserID)

   9: INSERT INTO tblEPiServerCommonGroupUser (intGroupID, intUserID) VALUES (@intGroupID, @intUserID)

It seems that who’s responsible for this is EPiServer.Common.Web.Authorization.Integrator and it may be possible to do something about this behavior. For now, I have no solution. Instead I tried to find a way to insert the correct groups back to the table. This stored procedure restores the table with correct information for the current user. It could be executed in Session_Start to make sure the user belongs to the correct groups the current session:

   1: SET ANSI_NULLS ON

   2: GO

3:

   4: SET QUOTED_IDENTIFIER ON

   5: GO

6:

   7: -- =============================================

   8: -- Author:        Jonas Lindau, MSB

   9: -- Create date: 2013-05-30

  10: -- =============================================

  11: CREATE PROCEDURE [dbo].[spCustomEpiServerFixGroups]

  12:     @intUSerId int

  13: AS

  14: BEGIN

  15:     SET NOCOUNT ON;

16:

  17:     DECLARE cClubs CURSOR

  18:     FOR

  19:         SELECT intClubID FROM tblEPiServerCommunityClubMember where intUserID = @intUserId

20:

  21:     OPEN cClubs

22:

  23:     DECLARE @intClubId int

  24:     DECLARE @intGroupId int

25:

  26:     FETCH NEXT FROM cClubs INTO @intClubId

  27:     WHILE @@FETCH_STATUS <> -1

  28:     BEGIN

  29:         SET @intGroupId = (SELECT intID FROM tblEPiServerCommonGroup WHERE strName = 'Group for Club with Id: ' + CAST(@intClubId as varchar))

30:

  31:         IF NOT EXISTS(SELECT * FROM tblEPiServerCommonGroupUser WITH(UPDLOCK, ROWLOCK, HOLDLOCK) WHERE intGroupID=@intGroupId AND intUserID=@intUserId)

  32:         BEGIN

  33:             --PRINT 'Values inserted for user: ' + CAST(@intUserId AS VARCHAR) + ' and group: ' + CAST(@intGroupId AS VARCHAR)

  34:             INSERT INTO tblEPiServerCommonGroupUser(intGroupID, intUserID) VALUES(@intGroupId, @intUserId)

  35:         END

  36:         /*

  37:         ELSE

  38:         BEGIN

  39:             PRINT 'Values already exists for user: ' + CAST(@intUserId AS VARCHAR) + ' and group: ' + CAST(@intGroupId AS VARCHAR)

  40:         END

  41:         */

  42:         FETCH NEXT FROM cClubs INTO @intClubId

  43:     END

  44:     CLOSE cClubs

  45:     DEALLOCATE cClubs

  46: END

47:

  48: GO

The procedure above can be executed like this:

   1: protected void Session_Start(Object sender, EventArgs e)

   2: {

   3:     try

   4:     {

   5:         DatabaseHandler.Instance.ExecuteNonQuery("spCustomEpiServerFixGroups", CommunitySystem.CurrentContext.DefaultSecurity.CurrentUser.ID);

   6:     }

   7:     catch { }

   8: }

If you need to correct the entire table in one job, you could try the script below.

What it does is extract a list of all users that is a member of any club and then insert the correct GroupId and UserId into tblEPiServerCommonGroupUser. I have tried this for a couple of days and it seems to work, but be warned! It may do stuff it not intended to, so use at own risk!

This is a one-time fix.

   1: SET NOCOUNT ON;

2:

   3: DECLARE cClubMembers CURSOR

   4: FOR

   5:     SELECT DISTINCT intUserID FROM tblEPiServerCommunityClubMember

6:

   7: OPEN cClubMembers

8:

   9: DECLARE @intUserId int

10:

  11: FETCH NEXT FROM cClubMembers INTO @intUserId

  12: WHILE @@FETCH_STATUS <> -1

  13: BEGIN

  14:     DECLARE cClubs CURSOR

  15:     FOR

  16:         SELECT intClubID FROM tblEPiServerCommunityClubMember where intUserID = @intUserId

17:

  18:     OPEN cClubs

19:

  20:     DECLARE @intClubId int

  21:     DECLARE @intGroupId int

22:

  23:     FETCH NEXT FROM cClubs INTO @intClubId

  24:     WHILE @@FETCH_STATUS <> -1

  25:     BEGIN

  26:         SET @intGroupId = (SELECT intID FROM tblEPiServerCommonGroup WHERE strName = 'Group for Club with Id: ' + CAST(@intClubId as varchar))

27:

  28:         IF NOT EXISTS(SELECT * FROM tblEPiServerCommonGroupUser WITH(UPDLOCK, ROWLOCK, HOLDLOCK) WHERE intGroupID=@intGroupId AND intUserID=@intUserId)

  29:         BEGIN

  30:             --PRINT 'Values inserted for user: ' + CAST(@intUserId AS VARCHAR) + ' and group: ' + CAST(@intGroupId AS VARCHAR)

  31:             INSERT INTO tblEPiServerCommonGroupUser(intGroupID, intUserID) VALUES(@intGroupId, @intUserId)

  32:         END

  33:         /*

  34:         ELSE

  35:         BEGIN

  36:             PRINT 'Values already exists for user: ' + CAST(@intUserId AS VARCHAR) + ' and group: ' + CAST(@intGroupId AS VARCHAR)

  37:         END

  38:         */

  39:         FETCH NEXT FROM cClubs INTO @intClubId

  40:     END

  41:     CLOSE cClubs

  42:     DEALLOCATE cClubs

43:

  44:     FETCH NEXT FROM cClubMembers INTO @intUserId

  45: END

46:

  47: CLOSE cClubMembers

  48: DEALLOCATE cClubMembers

49:

If you know a way to really solve this issue, please let me know!

Texts missing inTinyMCE?

Thu, 16 May 2013 13:37:48 GMT

If all captions and tooltips are missing in tinyMCE, there’s good chance you are using themed CSS. Over the years I’ve learned that EPiServer make some assumptions, that not always applies to your reality. One of them is that it’s assumed your not using themed CSS on your site. There has been some minor issues with themed CSS over the years, and a new one seems to apply to tinyMCE.

When I first configured the site to use tinyMCE, it looked like this:

Firebug to the rescue! Error 500 when loading TinyMCEi18n.aspx canätbe good, so whats the reason?

So how to solve this?

Well, you could edit the aspx-page and turn of theming, but chances are you apply a hot fix or service pack that overwrites your custom changes. I thought a better way was to turn theming off for that specific location using web.config. Since the file resides in /util/, you can solve this by editing the <location path="util">-section as below:

Save the file, and then tinyMCE should look great!

The annoying “error CS1031: Type expected” after upgrade site from CMS5R2 to CMS6

Tue, 30 Apr 2013 11:01:43 GMT

A few days ago a colleague upgraded a EPiServer site using Deployment Center and got this strange error. According to the dump the error was related to /Shell/Views/Shared/Site.Master. I knew the site was working correctly on my development machine, and I also knew the site was working fine in the test environment, so I ruled out any code-issues. Since there was already a lot of sites running on the production servers I also ruled out any missing dependencies. So what’s left? The Web.config of course!

The natural approach to find the error was to compare the non-working web.config in the production environment, with the working web.config from the test environment. That turned out to be an easy task since the only thing that was different was three lines, all which had to do with MVC!

For some reason Deployment Center didn’t care to replace the version of MVC on the production servers. It worked for the same site on my development machine and for the test servers. It also always worked for other sites on out production servers, but not this time.

So what I did was change the version numbers for System.Web.Mvc as below:

This is how it looked after upgrade

This is what I did change to make it work

And the error is gone!

Getting “Error– is not a valid user or group”, when upgrading?

Mon, 29 Apr 2013 11:20:08 GMT

I’ve upgraded a lot of sites in my days, but I still run into new errors. Even when you do it the right way, by upgrading the test-environment and collect as much information you can, there is still a chance it won’t work on the production servers. This error was one of those times…

At first I tried all the obvious things, like check if account was locked, did it have all the correct permissions etc. When all of this was confirmed I really didn’t have much more ideas. But then it hit me! I checked the account in Active Directory once more, and saw that the account name was so long, the “pre-Windows 2000”-name was truncated. For some reason, EPiServer Deployment Center uses this legacy login name during upgrade, and when the name is truncated it won’t work!

So this is what I did:

1. After starting Deployment Center and executed the correct upgrade, this error occurred:

2. Check the account in Active Directory to find out the name is truncated:

3. Rename the account using a shorter name.

4. Modify the application-pool so it uses the new username

5. Run upgrade again.

Hiding files with attribute hidden, system or temporary from file listing

Thu, 28 Mar 2013 11:29:19 GMT

On our sites we use pretty much the same file listing functions that comes with the example site in EPiServer. It’s simple and have worked well for many years, but recently one of our users realized that hidden files were visible in the listing. My first thought was that it should be the easiest thing ever to hide these files, surely the FileSystemDataSource would have a nice property called “ShowHiddenFiles” or something, ready use. It turned out i couldn’t be more wrong!

so what to do? After testing and thinking of several ways to solve my issue, I ended up with a quite simple solution where I used TreeNodeDataBound to determine if the current node should be rendered or not. In the sample unit this method is used to add different functionality for files and directories. Just add a few more lines of code to it like below:

protected void FileTree_TreeNodeDataBound(object sender, System.Web.UI.WebControls.TreeNodeEventArgs e)
        {
            VirtualFileBase virtualFile = e.Node.DataItem as VirtualFileBase;

            if (virtualFile.IsDirectory)
            {
                e.Node.SelectAction = TreeNodeSelectAction.Expand;
            }
            else
            {
                try
                {
                    VersioningFile RealFile = VirtualPathHandler.Instance.GetFile(virtualFile.VirtualPath, true) as VersioningFile;
                    string RealPath = RealFile.LocalPath;
                    FileInfo RealFileInfo = new FileInfo(RealPath);

                    if (RealFileInfo.Exists)
                    {
                        bool IsHidden = (RealFileInfo.Attributes & FileAttributes.Hidden) == FileAttributes.Hidden;
                        bool IsSystem = (RealFileInfo.Attributes & FileAttributes.System) == FileAttributes.System;
                        bool IsTemporary = (RealFileInfo.Attributes & FileAttributes.Temporary) == FileAttributes.Temporary;

                        if (IsHidden || IsSystem || IsTemporary)
                            FileTree.Nodes.Remove(e.Node);
                    }
                }
                catch { }

                e.Node.NavigateUrl = e.Node.DataPath;
            }
        }

Now it’s safe to list files from folders where hidden, system or temporary files are sored!

Link validation with EPiServer CMS 6R2 behind a proxy

Thu, 21 Mar 2013 18:39:02 GMT

Are you stuck with the error: “The link validation job was stopped due to 101 consecutive broken external links. There seem to be a problem for the server to access external resources.” when you enable Link validation? Are your server behind a proxy?

Then you should make EPiServer Link Validator aware of your proxy to make it work!

Open up the websites Episerver.config in Notepad
Locate the node <episerver>
Under this node, add a new child called <linkValidator> and add your proxy.
Restart site

A typical proxy configuration would look like:

There are in fact a few more things you can change, read more at:

http://world.episerver.com/Documentation/Items/Tech-Notes/EPiServer-CMS-6/EPiServer-CMS-60/Link-Validation/

MIME-types and VPP

Fri, 01 Apr 2011 13:36:06 GMT

Yesterday i ran into a problem when i was moving an old JAVA-application from an Apache-server into i VPP-folder on our EPiServer site.

What i wanted to do was to host the JNLP-file and the JAR-files in a VPP-folder, and simple create a hyperlink to the JNLP-file from a web page, rather then have a separate Apache that only serves JNLP-files.

I knew it would be a bit of an issue to set the correct MIME-types for my JNLP-file when stored on a VPP-folder, so what i wanted was a flexible solution that would allow me to add other MIME-types if needed. Here is what i came up with:

First of all i added a handler in global.asax that is triggered when i file transmission starts.

   1:  void Application_Start(object sender, EventArgs e)

   2:  {

   3:      UnifiedFile.UnifiedFileTransmitting += 
               new UnifiedFileEventHandler(UnifiedFile_UnifiedFileTransmitting);

   4:  }

Also in global.asax:

   1:  void UnifiedFile_UnifiedFileTransmitting(UnifiedFile sender, 
           UnifiedVirtualPathEventArgs e)

   2:  {

   3:      if (sender == null || sender.Name == null || sender.Extension == null)

   4:          return;

5:

   6:      UnifiedFileManager.HandleMimeType(sender.Extension);

   7:  }

As you can see i don’t add a lot of code in global.asax, i prefer to create manager-classes. The UnifiedFileManager-class looks like this:

   1:  public static class UnifiedFileManager

   2:  {

   3:      public static void HandleMimeType(string Extension)

   4:      {

   5:          string HandledMimeTypes = 
                   ConfigurationManager.AppSettings["HandledMimeTypes"].ToString();

6:

   7:          foreach (string HandledMimeType in HandledMimeTypes.Split("|".ToCharArray()))

   8:          {

   9:              string[] ThisMimeType = HandledMimeType.Split(";".ToCharArray());

  10:              if (ThisMimeType[0].Equals(Extension))

  11:                  HttpContext.Current.Response.ContentType = ThisMimeType[1];

  12:          }

  13:      }

  14:  }

And the final step is to add a line in appSettings. This is where you store the MIME-types and their extensions:

<add key="HandledMimeTypes" value=".jnlp;application/x-java-jnlp-file" />

To add more MIME-types, simple separate them with a |-character, like this:

<add key="HandledMimeTypes" 
    value=".jnlp;application/x-java-jnlp-file|.pdf|application/octet-stream" />

Now, i was able to launch my JNLP-file without any problems.

XForms: Collecting data and labels and pass it to an existing webservice

Thu, 24 Mar 2011 16:30:54 GMT

Objective

Let service desk staff create forms containing support question using XForms and post it to an existing webservice. They should be able to create as many forms as they needed, with different questions.

So it begins…

I really hadn’t done much work with XForms before, just some templates similar to the demo template. I knew there ware an option to post forms to a specified URL, so i thought this would be really easy. Just build some forms, set it to post to an URL and we should be fine.

Well, it wasn’t that easy… i found out that the option “Post to specified URL” really does nothing unless you add the code for it yourself. Fair enough, it should be easy.

Adding some code to the BeforeSubmitPostData and AfterSubmitPostData, collect all data and post it to the URL stored in e.FormData.CustomUrl. should solve this. Or?

Well, i could be enough most times, but i needed to get my hands on the text inside the labels. The webservice wanted me to send in all text in one string, something like this:

Label 1\n
Value inside the textbox\n
\n
Label 2\n
Value inside the textbox\n

Getting the labels proved to be the main issue here. Is far as i know, the labels are nowhere to be found within the XForm-classes. So, what to do?

I turned out i had to parse the forms XML to get my hands on the labels. This is what i added to BeforeSubmitPostedData to get all textfields:

protected void XForm_BeforeSubmitPostedData(object sender, SaveFormDataEventArgs e)
{
XFormControl control = (XFormControl)sender;

XmlDocument Doc = new XmlDocument();
Doc.LoadXml(control.FormDefinition.Document.InnerXml);

XmlNamespaceManager ns = new XmlNamespaceManager(Doc.NameTable);
ns.AddNamespace("xforms", "http://www.w3.org/2002/xforms");

XmlNodeList InputFields = Doc.SelectNodes("//xforms:input", ns);

foreach (XmlNode thisNode in InputFields)
{

//Inside this loop, you can access the label by using
//thisNode.InnerText
//
//The value of textfield can be accessed by using:
//e.FormData.GetValue(thisNode.Attributes["ref"].Value)

}
}

So i just stored all label values and textfield values inside an array of KeyValuePair, used a stringbuilder to put it all together into a string with the expected format and passed it on to the webservice.

It wasn’t quite as easy as i thought, but in the end it wasn’t really that hard either.

XForms and SSIS (Integration services)

Thu, 28 Jan 2010 09:01:53 GMT

Objective

Collect and transform data collected by XForms using SSIS.

Recently i ran into a situation where i needed to collect data from EPiServer (XForms) using SSIS. At first, i had a few possible ways to solve this task.

Use SQL and collect the data direct from database.
Create my own Data Flow Source in SSIS using EPiServer assemblys.
Simply create an XForm which posts to a custom made ASPX-page.

After a few hours i knew none of the above would do the trick for me because:

The data is binary serialized and i couldn’t find a way of deserialize it.
The assemblies wouldn’t run on my website, so it would be useless.
My network is extremely segmented, so the ASPX-page wouldn’t be able to store data on my SSIS server.

So, what to do?

I decided to create a generic handler on my site which exposed the data as XML. It would than be an easy task for SSIS to get the data, live, using the XML Source component.

Lets begin

I found this excellent blog entry by Dominik Juszczyk: http://blog.juszczyk.biz/2008/05/get-posted-data-xformdata-from-xform/ which describes how to get the data. This turned out to be a good start for me.

I created a class called XForms containing two methods. One method which return the entire XForm, and one which return the posted data.

public static IList<XFormData> GetXFormData(XForm xForm)
{
return xForm.GetPostedData();
}

public static XForm GetXForm(PageReference pageReference, string xFormPropertyName, string xFormFolderPath)
{
if (pageReference == null || pageReference == PageReference.EmptyReference)
return null;

PageData pageData = global::EPiServer.DataFactory.Instance.GetPage(pageReference);

    PropertyData xFormProperty = pageData.Property[xFormPropertyName];
    if (xFormProperty == null || xFormProperty.Value == null)
        return null;

if (!(xFormProperty is PropertyXForm))
return null;

    IList<XForm> xFormFolder;
    if (xFormFolderPath != String.Empty)
        xFormFolder = XFormFolder.GetForms(xFormFolderPath);
    else
        xFormFolder = XFormFolder.GetForms();

    foreach (XForm xform in xFormFolder)
    {
        Guid formGuid = new Guid(xFormProperty.Value.ToString());

        if (formGuid.CompareTo((Guid)xform.Id) == 0)
        {
            return xform;
        }
    }

return null;
}

Create the handler

The next step to take were to create the handler. I simply created a new generic handler on my site.

The first thing to do is to make sure the handler works. I created a folder at root level called Modules. For the handler to work at all, i had to add a location element in web.config.

When this issue were solved, the rest is quite easy. All i had to do was to make sure the handler produces XML. A also needed it to take some parameters, like PageReference to the page containing the form, XForm property name and XForm folder. The folder parameters isn’t really needed, but if you have a lot of forms, you don’t want to loop through them all… The lines XForms.GetXForm and XForms.GetXFormData calls the methods in the class described above.

public void ProcessRequest(HttpContext context)
{
context.Response.ContentType = "text/xml";

            PageReference PageLink;
            string XFormsPropertyName;
            string XFormsFolderPath;
            try
            {
                PageLink = PageReference.Parse(context.Request.QueryString.Get("PageLink"));
                XFormsPropertyName = context.Request.QueryString.Get("XFormsPropertyName");
                XFormsFolderPath = context.Request.QueryString.Get("XFormsFolderPath");

XForm xform = XForms.GetXForm(PageLink, XFormsPropertyName, XFormsFolderPath);

                StringBuilder sb = new StringBuilder();
                sb.Append("<XForm>");
                sb.Append("<FormName>").Append(xform.FormName).Append("</FormName>");
                sb.Append("<AllowAnonymousPost>").Append(xform.AllowAnonymousPost).Append("</AllowAnonymousPost>");
                sb.Append("<AllowMultiplePost>").Append(xform.AllowMultiplePost).Append("</AllowMultiplePost>");

                IList<XFormData> xformdata = XForms.GetXFormData(xform);
                foreach (XFormData data in xformdata)
                {
                    sb.Append("<DataItem>");
                    sb.Append("<Id>").Append(data.Id).Append("</Id>");
                    sb.Append("<DatePosted>").Append(data.DatePosted).Append("</DatePosted>");
                    NameValueCollection dataparams = data.GetValues();
                    foreach (string dataparamname in dataparams)
                    {
                        sb.Append("<").Append(dataparamname).Append(">").Append(dataparams[dataparamname]).Append("</").Append(dataparamname).Append(">");
                    }
                    sb.Append("</DataItem>");
                }
                sb.Append("</XForm>");

                context.Response.Write(sb.ToString());
            }
            catch {
            }
        }

Result

Time for testing. I logged on my EPiServer and created a XForm with the elements Lastname, Firstname and Email. After some posting to the form i was anxious to try my new handler. I opened up my browser and typed the URL http://localhost/Modules/XFormsDataExport.ashx?PageLink=131&XFormsPropertyName=XForm&XFormsFolderPath=

The result was exactly what i needed

<XForm>
<FormName>My test form</FormName>
<AllowAnonymousPost>True</AllowAnonymousPost>
<AllowMultiplePost>True</AllowMultiplePost>
<DataItem>
<Id>883a4e9c-0e53-47d6-9d7d-bade1b699cd4</Id>
<DatePosted>2010-01-26 11:34:37</DatePosted>
<Lastname>Doe</Lastname>
<Firstname>John</Firstname>
<Email>john@doe.net</Email>
</DataItem>
<DataItem>
<Id>f6501f48-9301-4ce8-8443-9d79de6b03fb</Id>
<DatePosted>2010-01-26 11:44:02</DatePosted>
<Lastname>Robinsson</Lastname>
<Firstname>Mrs</Firstname>
<Email>mr@nowhere.net</Email>
</DataItem>
</XForm>

Consume the XML in SSIS

This is a really simple task. All i had to do was to add an XML Source, enter the URL to my handler, generate an XSD and map the columns. Voila!

Conclusion

For security reasons i locked down the access to my handler. I simply used IIS IP security to do this. I also made a lot of data validation in my SSIS package to make sure i didn’t get any dirty data.

Error 500 when using advanced fileupload (drag and drop)

Tue, 20 Oct 2009 14:13:34 GMT

If you are using Themes in your EPiServer CMS templates, chanses are good you cannot use the drag and drop functionallity in EPiServer File Manager.

Lets say you create a theme called myTheme. Then you’ll probably add this to web.config’s pages-element instead of add it to all templates.

After adding this to web.config, drag and drop stop working. You can easily solve this by adding a pages-element to your UI-location:

Now you can start dragging thoose files :)