Thanks Atanas, I'm waiting for your updated sample

Best regards :)

Dear Alex,

The article I mentioned is not yet ready, but the Silverlight sample that shows how to persist the data in the PivotGrid is ready.

I am attaching a solution that demonstrates how to save and load the states:

1. to and from memory with no serialization (updated save/load sample),
2. to and from an XML file with XML serialization
3. to and from memory with Binary Serialization.

In the third case the only thing that you need to do in order to save the states in a database is to upload the byte array called savedBinaryData with the serialized result to the database. One way to do that is to use a WCF service. More on that in the coming article.

I hope that this is helpful. If you have any questions feel free to ask me!

All the best,

Atanas Dyulgerov

P.S.

The solution comes with sample data generator, FlatDataSource that uses the generator and prepares a nice sample initial view, a custom class that represents each saved view (this is majorly different than the same class from the SamplesBrowser). This last class is also the one that we serialize. In the main page we have a number of buttons that trigger the different serializations and deserializations. The GroupBox class and the contents of the Themes folder implement a helper control to make the UI a bit more useful.

The two files that will help you most are MainPage.xaml.cs and SavedPivotGridView.cs

One more thing, the sample will look for the following assemblies:

• InfragisticsSL4.Controls.Grids.XamPivotGrid.v10.2.dll
• InfragisticsSL4.Controls.Menus.XamTree.v10.2.dll
• InfragisticsSL4.DragDrop.v10.2.dll
• InfragisticsSL4.Olap.FlatData.v10.2.dll
• InfragisticsSL4.Olap.v10.2.dll
• InfragisticsSL4.v10.2.dll

The default location for those after you install the Silverlight NetAdvantage trial package is:

C:\Program Files (x86)\Infragistics\NetAdvantage 2010.2\Silverlight DV\Bin

If you have chosen a different location you have to update the references.

Thanks a lot Atanas

Actually, I have figure it out. In the Fill method:

 // Preseving the order for Measures

foreach(string ms in this.Measures)

 {

    var f = pivotGrid.DataSource.Cube.Measures.Where(p => p.Name == ms.ToString());

    if(f.Count() > 0)

      {

         IMeasure meas = f.First();

         IMeasureViewModel mvm = pivotGrid.DataSource.CreateMeasureViewModel(meas);

         pivotGrid.DataSource.Measures.Add(mvm);

      }

 }

//rows

foreach (string st in this.RowHierarchies)

 {

  foreach (IDimension dim in pivotGrid.DataSource.Cube.Dimensions)

     {

        var rows = dim.Hierarchies.Where(r => r.UniqueName == st);

        if (rows.Count() > 0)

          {

            IFilterViewModel fvm = pivotGrid.DataSource.CreateFilterViewModel(rows.First());

            pivotGrid.DataSource.Rows.Add(fvm);

           }

      }

 }

 //columns order

foreach (string st in this.ColumnHierarchies)

  {

   foreach (IDimension dim in pivotGrid.DataSource.Cube.Dimensions)

   {

    var colmns = dim.Hierarchies.Where(r => r.UniqueName == st);

    if (colmns.Count() > 0)

     {

       IFilterViewModel fvm = pivotGrid.DataSource.CreateFilterViewModel(colmns.First());

       pivotGrid.DataSource.Columns.Add(fvm);

     }

   }

  }

Regards.

Hi Atanas,

I am working with PivotGrid and in my application I save / load grid layout in / from  xml file in the local folder. All looks good and I am using your example to save and load. There is one issue though. If columns, rows or measures on the grid on save in the specific order, this order is not perseved on load. As an example. If columns order on save is Distance, Km Per Liter, Usage, CO2, CO2e. After load from the file the order is CO2, CO2e, Km Per Liter, Distance, Usage. The same issue for rows and measures. Is ther any way to keep the order on load?  Or any grid settings can help with this?

Thank you.

Irina

Hi, Achim,

The XMLA data source works in a different way.

The browser hangs because waitOne suspends the whole thread and the pivotgrid cant expand its members (which means the event that stops the waiting never fires). If you want to use XMLA data source you need to take the expansion process in a separate thread.

I'm attaching a sample that expands filter members up to a certain level with XMLA data source. You will find all you need there.

If you cant figure out what to change in the serialization sample I'll post the modified version tomorrow.

All the best,

Atanas