Execute MDX Query with TSQL

To start with execution of MDX query with TSQL first we need to create a Linked Server. Let’s discuss few points regarding this

Linked Server

  Linked server enables the SQL Server Database Engine to execute commands against OLE DB data sources outside of the instance of SQL Server. Linked servers are configured to enable the Database Engine to execute a Transact-SQL statement that includes tables in another instance of SQL Server, or another database product such as Oracle, mysql, MS Analysis services etc. Many types of OLE DB data sources can be configured as linked server.

Now let’s proceed with creation of linked server within the sqlserver instance

Create a linked server

·         To create a linked sever you must have admin right of SqlServer.

·         Navigate to Server Objects > Linked Servers

·         Right click on Linked Servers and select New Linked Server.

Let’s take a sample cube database AdventureWorksDW2008R2 and create a linked server on this.

EXEC

sp_addlinkedserver

@server='Test', /* Here specify a name to linked server */

@srvproduct='',/*Is the product name of the OLE DB data source to add as a linked server. product_name is nvarchar(128), with a default of NULL.*/

@provider='MSOLAP',/*Is the unique programmatic identifier (PROGID) of the OLE DB provider corresponding to this data source. provider_name must be unique for the specified OLE DB provider installed on the current computer.*/

@datasrc='localhost',/*'DRIVER={SQL Server};Server=(local); Initial Catalog=master;Integrated Security=SSPI;'*/

       @catalog='Adventure Works DW 2008R2' /*Cube Database Name*/

4.       Once the linked server is set, we need TSQL Rowset OPENQUERY function.

Open Query

Executes the specified pass-through query on the specified linked server. This server is an OLE DB data source. OPENQUERY can be referenced in the FROM clause of a query as if it were a table name. OPENQUERY can also be referenced as the target table of an INSERT, UPDATE, or DELETE statement. This is subject to the capabilities of the OLE DB provider. Although the query may return multiple result sets, OPENQUERY returns only the first one.

Syntax: - OPENQUERY (linked_server,’query’)

Now you can write your Script

USE

AdventureWorksDW2008R2

GO

Declare @MDXExpression as Varchar(MAX)

Create Table  #temp(

                        Country varchar(Max),

                        Sale_Amount  float,

                        Product_cost float,

                        Tax_amount Float,

                        TProduct_Cost Float

                        )

Set  @MDXExpression =

'SELECT

NON EMPTY

      {

      [Measures].[Sales Amount]

      ,[Measures].[Standard Product Cost]

      ,[Measures].[Tax Amount]

      ,[Measures].[Total Product Cost]

      } ON 0,

NON EMPTY

      {

NonEmpty ( [Sales Territory].[Sales Territory Region].[Sales Territory Region])

      } ON 1

FROM

      [Adventure Works]';     

insert into #temp     

Exec ('SELECT * FROM OpenQuery(Test,''' + @MDXExpression + ''')')

/* Here you can join your MDX result set with your respective Sql server database tables and can extract expected result set */

SELECT      SalesTerritoryGroup,  #temp.*

FROM         DimSalesTerritory inner join #temp   on #temp.[Country]= DimSalesTerritory.SalesTerritoryCountry

#ERROR/NAN / error: Input string was not in a correct format issue in SSRS

While working on reports, we usually get “#ERROR”,”N/A” or “error: Input string was not in a correct format”error.

The major problem behind these issues is data type mismatches and some additional reasons that cause these issues.

Let’s talk about how to resolve these issues and why we get these issues.

1.    If you are performing any aggregate operation like SUM, AVG etc  on the report and also you know that your input field might have string value then never use an expression like

=sum(iif(isnumeric(Fields!YourFieldName.Value), cdbl(Fields!YourFieldName.Value),0))

OR

=sum(iif(isnumeric(Fields!YourFieldName.Value), cint(Fields!YourFieldName.Value),0))

OR

=sum(iif(isnumeric(Fields!YourFieldName.Value), iif(isnumeric(Fields!YourFieldName.Value),cdbl(Fields!YourFieldName.Value),0),0))

The reason behind this is ISNUMERIC() method works in a mysterious way. If we will try to explorer this method a bit more in detail then we will find that is ISNUMERIC()returns true for any expression if there will be any possibility to convert the expression in numeric. For example, if you test ISNumeric("1,234.55") will return true whereas SUM("1,234.55")will give you #Error. To resolve this sitution

Always use expression like =SUM(VAL(Fields!YourFieldName.Value))

2.    If you are using cube/MDX report and also using your aggregate expression like above but still getting an error or an unexpected result then you need to check these points

·         Start your sql server profiler (Tools menu-> Sql Server Profiler) with Analysis services as service type. Execute/Preview your report and extract/copy the MDX from the profile. Now execute this MDX on the Analysis Services query browser. Make sure that your report’s aggregate column is available in the output. If you see your column unavailable in the output then you need to play a workaround to get include your missing column in the output.

            The reason behind this missing column is if the resultant of your entire column is NULL and you are using Non empty/Nonempty then query engine exclude the column from the output to increase the query response time.

To resolve this problem, you need to add a calculate Member in your MDX query and append the expression by adding +0 at the end like :

With

Member CalculatedMember as [Measures].[Measure Name]+0

·         Sometime due to mishandling the expression in the MDX query/Calculated measure, we get “Infinity”,”N/A” or scientific numeric format like 1.00E+12 in the MDX output.

To see the expected result at report level, we need to handle these problems at MDX level by using appropriate expression/fields/methods.

Exclude Week End Days From Calendar

Today I found a post at MSDN to exclude Sunday using Datediff(). I worked around the query and found a solution

http://social.msdn.microsoft.com/Forums/en-US/sqlreportingservices/thread/9b3efb7f-2015-432f-8624-c78a5c535f1c

declare @start datetime,

              @end datetime

set @start = '2006-01-01'

set @end =   '2006-01-08'

;

with calendar(date,isweekday, y, q,m,d,dw,monthname,dayname,w) as

(

select @start ,

case when datepart(dw,@start) in (1,7) then 0 else 1 end,

year(@start),

datepart(qq,@start),

datepart(mm,@start),

datepart(dd,@start),

datepart(dw,@start),

datename(month, @start),

datename(dw, @start),

datepart(wk, @start)

union all

select date + 1,

case when datepart(dw,date + 1) in (1,7) then 0 else 1 end,

year(date + 1),

datepart(qq,date + 1),

datepart(mm,date + 1),

datepart(dd,date + 1),

datepart(dw,date + 1),

datename(month, date + 1),

datename(dw, date + 1),

datepart(wk, date + 1) from calendar where  date + 1<= @end

)

select * into #test from calendar option(maxrecursion 10000)

Select  DATEDIFF(DD,@start,@end)+1-Count(case when #test.isweekday=0 then #test.isweekday end )

from #test where #test.date between @start and @end

You can also find the solution in this way

SELECT startDate,EndDate,DATEDIFF(DD,startDate,EndDate) FROM Production.WorkOrder WHERE ((DATEPART(dw, StartDate) + @@DATEFIRST) % 7) NOT IN (0, 1)

How to restrict/set number of records per page in SSRS report

Sometimes we came into a situation where we need to show only a specific number of records on per page.

Let’s see how we can restrict the number of records on per page. I am assuming that you are good with your report design. So I will focus only on the required expressions/steps that will need us to achieve required page break in the report.

Step 1: Add a Parent Group under Row Group section. Now on popup window under “Group by:” dropdown, use this expression:

Image -1

=Int((RowNumber(Nothing)-1)/10)

In this expression, you need to change 10 to

any desire number as you wish to see number of records on per page

Image-1 will help you regarding this step.

Step 2: Open the group properties window(Right Click on group name-> select Group Properties) of this newly added group.             

Image -2

           Go to Page Breaks and checked “Between each instance of the group”.

           Now go to Sorting and delete column/sorting expression (if any).

           You can also delete the 1^st column of your table that has been created automatically on action of add new group by choosing “Delete Columns only” option from the prompt.

           Image-2 will help you regarding this step.

Now you are done. Preview your report and check report output.

You can download the sample report from following link.

Number of records per page Sample Report

An error occurred during report processing

"An error occurred during report processing. Index was out of range. Must be non-negative and less than the size of the collection. Parameter Name: index"

I did copy paste of my existing RDL that was using SSAS as datasource and made my all the required changes in the report. But during the report preview I was getting this error “An error occurred during report processing. Index was out of range. Must be non-negative and less than the size of the collection. Parameter Name: index”.

I thought since this is an index related error so there would be some custom code or DLL reference that was throwing this error. But there was neither custom code nor any DLL reference on my report. After a lot of efforts I have found that:

There was a Filed Source that was bind with two different Field Name. After correcting the Field   Source, this error has disappeared.

I am sharing this issue with you so that it will save your couple of valuable hours.