MDX Essentials: Basic Member Functions: The .Item() Function - Page 2

August 9, 2004

Practice

Preparation

To reinforce our understanding of the basics we have covered so far, we will use the .Item() function in a manner that illustrates its operation. We will do so in simple scenarios that place .Item() within the context of meeting a business need.

To begin, we will construct a SELECT query with a clearly defined set, then put .Item() to use in manipulating that set to meet the business needs of a group of hypothetical information consumers. The intent is, of course, to demonstrate the operation of the .Item() function in a straightforward manner.

Let's return to the MDX Sample Application as a platform from which to construct and execute the MDX we examine, and to view the results datasets we obtain.

Let's assume, for our practice example, that we have received a call from the Finance department of the FoodMart organization, requesting some information surrounding units shipped for a given time frame upon which they plan to perform some basic analysis. The Finance information consumers specifically wish to know the total units shipped figure for the US store states, for each of the quarters of 1998, having become focused on this state / time combination initially to do some per unit analysis of various logistics costs.

We will first create a basic query that returns the units shipped totals for each of the US store states, over the quarters under consideration. It is from the results dataset that is returned that the consumers state that they may want to narrow their request, once they get a look at overall figures, to more focused view(s) of specific state(s) only.

Let's construct a simple query, therefore, to return the units shipped information, presented by quarters (as columns) and the store states (as rows).

-- MDX022-1, Preparation for Use of .Item() Function, Numerical Index, 
-- in a    Basic Query
SELECT
   {[Time].[1998].Children} ON COLUMNS,
   {[Store].[All Stores].[USA].Children}  ON ROWS
FROM
   [Warehouse]
WHERE
   ([Measures].[Units Shipped])

The Results pane is populated by Analysis Services, and the dataset shown in Illustration 2 appears.

We see the four quarters of 1998 populating the columns across, and the Store States (from the Stores dimension) appearing on the row axis.

Next, let's say that our information consumers are provided with the full set of data that we have generated. They examine the metrics, and then request that we narrow the report scope to "Oregon-only" stores.

We realize, from looking at our first results dataset, and perhaps in Analysis Manager, etc., that Oregon is the second of the three members of the tuple we have defined within our initial query. We decide that the .Item() function, with an index setting of 1 ("0" being the first member position, and "1" the second) will serve to generate the new limitation on the results.

We take the following actions to meet the new consumer request:

-- MDX022-2, Use of .Item() Function, Numerical Index, in a  Basic Query

   {[Store].[All Stores].[USA].Children}  ON ROWS

                 {[Store].[All Stores].[USA].Children.Item(1)} ON ROWS

The Query pane appears as shown in Illustration 3, with the new addition circled.

The Results pane is populated, and the dataset depicted in Illustration 4 appears.

We have used the .Item() function, in its member incarnation, with a simple numerical index. Let's see an example where we use a string within the index, based upon our original query above, where we selected all three store states.

-- MDX022-3, Use of .Item() Function, String Index, in a  Basic Query

            "WA"

The Query pane appears as shown in Illustration 5, with changes.

The Results pane is populated, and the dataset depicted in Illustration 6 appears.

We deliver the new results to the Finance consumers, who accept our work thus far with stated satisfaction. They promise to "get back to us" with follow-up requests in short order.

We next hear from a representative of Finance a couple of days later, with a request that contains elements similar to the specifications so far. The group needs the units shipped totals for the Washington stores, from the perspective of non-consumable goods sold only, for the first three quarters of 1998.

We recognize instantly the opportunity, in this case, to combine both the member and tuple versions of the .Item() function into a single query, and set out to do so in the following steps.

-- MDX022-4, Use of .Item() Function, Member and Tuple (MultiString Index)
SELECT
   {[Time].[1998].Children.Item(0): [Time].[1998].Children.Item(2) } ON COLUMNS,
{CROSSJOIN([Store].[All Stores].[USA].Children, [Product].
   [Product Family].Members).Item("[WA]","[Non-Consumable]")} ON ROWS
FROM
   [Warehouse]
WHERE 
   ([Measures].[Units Shipped])

As we can see, we are using the member variation of the .Item() function to return the three quarters (indexes 0, 1, and 2, respectively), the range of which we manage conveniently with the colon (":") operator. In addition, we are using the tuple variation of .Item(), with multiple string indexing option, in the rows axis, where we are cross joining store states and product family. We comply with the rules for the multi-string index approach, in that we stipulate each dimension present in the tuple, in appropriate order, and so we obtain the desired results. (Try changing the order of the strings, or otherwise manipulate them, to see the results, as convenient).

NOTE: For more information on the CrossJoin() function, see my Database Journal article MDX Essentials: Basic Set Functions: The CrossJoin() Function .

The Results pane is populated, and the dataset shown in Illustration 7 appears.

Summary ...

In this lesson, we explored the MDX .Item() function. We discussed the tandem general objectives of the function, to return a member from a tuple, or a tuple from a set, and then we examined the syntax for each of the two variations.

We then undertook illustrative practice examples of business uses for the .Item() function, generating queries that capitalized on its primary features. Finally, we briefly discussed the results datasets we obtained from each of the queries we constructed.

» See All Articles by Columnist William E. Pearson, III

Discuss this article in the MSSQL Server 2000 Analysis Services and MDX Topics Forum.