Business Objects

Universe tutorial # 1

2009-06-10T09:56:00.011+03:00

Overview:

This is a fast training on how to create and design SAP business objects XI universe. To build a universe you can use the designer tool or the universe builder tool, both of them is a totally independent tool. In this short tutorial I will use the designer tool to build a small universe. I will write another separate post to describe how to use universe builder and what is the deference between universe builder and designer tool.

Before you begin:

I assume that

You have installed business objects client on your machine.
You have an access to business object server.
You have the required permission to create universe and all type of objects.

Login:

Click on start menu.
Go to this path: start --> All Programs --> Business Objects XI Release 2 --> Business Objects Enterprise --> Designer.
Enter system, user name, password and authentication: Enterprise (by default)

Create new universe wizard:

When you login in to the first time the create wizard will start automatically and the welcome screen will displayed.
Click begin and then in the "define universe parameters" screen enter the following
- Universe Name: Ejada
- Connection: create a new connection of use an already exists one.
- In the "create initial classes and objects" screen start by select tables form the left pan and add them by clicking the add button to the right pan. You even can select a specific column. Just expand the + beside table and select column or multiple by holding the CTRL button while you selecting.
- In the "Create measure objects" start selecting your measure column (column that contains numbers and you will not use it in aggregation, like profit, cost, balance etc.) and the function that you want to apply to that measure (count, sum, max, min etc.)
- In the congratulation screen click finish.
- Note: you can just cancel the wizard and do all this steps manually.

Insert new Table:

After finishing the wizard you can add any tables that you want to use in this universe. Let's say that we want to add some lookups tables (tables contains descriptions for some cods stored in the fact table)

Right click anywhere in the designer windows in the area that contains the previously added table.
Select Tables… from the drop down menu.
A list with all tables names in your database will be displayed, just select a table and click on the insert button.
Click refresh button if you don't see the table that you want. This may happen if the tables added while you just open the insert table window.

Joins:

To insert a join you may just drag a column from the first table and drop it on the referenced column in the target table. A black line will join between the 2 columns. Double click on this black line and the join window will open

Table1: primary table with the key column.
Table2: secondary table with the referenced column.
Operation: operation that will be used in join (example: table1.user_id = table2.user_id)
Cardinality: use the detect button to detect the join cardinality or set it manually. (1- many, many-1, many – many)
If you need to specify an outer join just select the box under the table that you want to outer join.
For complex joins use the expression text area below.

Derived table:

To create a derived table, right click in the designer window and select derived tables from the drop down menu.

Derived table name: type the descriptive name for your derived table.
Type the SQL select statement that will define you derived table.
Click on check syntax button and close.
A new table with the specified columns in the select statement will be displayed in the table's area in the designer window.

Class:

To create a new class click on the insert class button from the tool bar or select insert-->class. Then type the class name and description.

Dimension:

Select the class that you want to create the dimension in:

Name: dimension name
Type: character, number, date, long text.
Description: write description for this dimension.
Select: select the column that associated with that dimension. May be also being complex and having some logic.
Where: type the filter that you want to use while displaying this dimension.

Measure:

Name: name for the measure.
Function: function that will be used when aggregating (sum, count, max, min)
Select: data base column or any calculated numeric value.

Condition:

Name: condition name.
Condition: expression that define the condition.

Connection:

Database: oracle, MS SQL, DB2…etc.

Connection string (host, port, service name,…etc)

@prompt:

2009-06-07T15:01:00.001+03:00

Description:

Use this function to prompt user for a value will be used during report run. For example if you want to run a monthly report against a specific month, you have to prompt the user to enter the Month or even select it from the list of value displayed.

Syntax:

@prompt ('prompt text (1)', 'type (2)', 'list of values(3)', 'mono/multi(4)', 'free/constrained(5)')

Parameters:

Param	Description	Mandatory	Values
1	Text used to prompt the user when he trying to run/refresh the report	Yes	N/A
2	Type of the expected value entered by the user and also must be same type of the object that we will use the prompt return value to compare with	Yes	A: alphanumeric C: string D: Date N: Number U: Unicode
3	List of values that will displayed to the user if he click on the Values button when he prompted to enter a value	No	Object defined to retrieve list of values
4	Number of valued allowed	No	Mono: user can select one value only Multi: user can select Multi Values
5	User can edit enter a value by himself or not.	No	Free: user can type his own value Constrained: user must select from list of values.

Where to use this function:

You can use this function anywhere; this is some places that you can use this function

Filter: you can use this function to prompt the user for a value that will use in filtering the retrieved data.
Where clause: when you create an object you can use this function in the where clause.
Select clause: you can use this function even in the select clause see examples.
Derived table: you can use this function in your derived table query.

Tips and Tricks:

You can use the same prompt in more than one place; just make sure that the prompt text exactly the same when you use it with other universe objects.
You can use data conversion function to convert the prompt returned value before using it. For example if the prompt return "Jan09" and you want to compare it with a Date column in database you have to convert it before using it. In our case we will use to_date (@prompt(,,,,,),'MONYY').
If your prompt will return multi value use IN operation instead of equal (=)
Note that not all users having permission to view list of values while they trying to run or refresh the report by default. The administrator should grant this permission to this user.
If you will use the free mode (user can type what he want) you should give a hint in the prompt text message about the data format expected. For example if the user should enter month in format (MON YY) then the prompt message should be descriptive like "Please enter month like Jan 09".

Example # 1 (simple prompt):

We need to create a new condition that will prompt user to enter a valid date to display data related to this date in our report.

Create a new condition with the specification below [Insert --> Condition]

Name: As of Date prompt
Description: Prompt user to enter a specified as of date value
Where:

@select(Dimensions/As of Date) = @prompt ('Enter As of Date:', 'D', ['Dimensions/As of Date'], mono, constrained)

Notes:

Red single quote places. It will give you error message "Parse failed: invalid definition UNV0023"

Example # 2 (Prompt returns multi values):

We need to create a new condition that will prompt user to select cities that he wants to display revenue data

Create a new condition with the specification below [Insert --> Condition]

Name: Cities prompt
Description: Prompt user to select one or multiple cities to filter data according to
Where:

@select(Dimensions/City) IN @prompt ('Select one or more cities from the list:', 'C', ['Dimensions/City'], Multi, constrained)

Notes:

We used Multi key word instead of mono this time this to indicate that the prompt will return one or more values.
Note also we used the IN operator instead of equal. This is because the return value is a collection of values not one single value as the previous example.

Example # 3 (user type Free Prompt):

We need to create a new condition that will prompt user to type the city that he want to display report data for.

Create a new condition with the specification below [Insert --> Condition]

Name: City prompt
Description: Prompt user to select or type a city
Where:

@select(Dimensions/City) = @prompt ('Type a City name:', 'C', ['Dimensions/City'], mono, free)

Notes:

Note that we have used free key word instead of constrained. This will give the user the flexibility to type the city name directly without need to select form list of values.
Note that if the user type Paris while the stored value in @select(Dimensions/City) is PARIS then the query will return no data. To solve this issue you can use the Upper or lower function that will return the given string in upper case format or lower case format.
UPPER(@select(Dimensions/City)) = UPPER(@prompt ('Type a City name:', 'C', ['Dimensions/City'], mono, free))
If you want the user to type multiple cities use this form
@select(Dimensions/City) IN @prompt ('Type Cities name:', 'C', ['Dimensions/City'], Multi,
free)
Please note that user will enter values comma separated without space Like: (Paris,London,Roma,Cairo)

Example # 4 (From To prompt):

We need to create a new condition that will prompt user to enter "From Date" and "To Date" to display data related to this period in our report.

Create a new condition with the specification below [Insert --> Condition]

Name: From To prompt
Description: Prompt user to enter a specified period
Where:

@select(Dimensions/As of Date) Between @prompt ('From Date:', 'D', ['Dimensions/As of Date'], mono, constrained) AND @prompt ('To Date:', 'D', ['Dimensions/As of Date'], mono, constrained)

Example # 5 (Use prompt with dimensions):

We need to create a new dimension that will display customer segment description based on what language user will select.

Prerequisites:

Language dimension Created.
Segment English Description created.
Segment France Description created.
Segment Dutch Description created.
Segment Arabic Description created.

Create a new Dimension with the specification below [Insert --> Dimension]

Name: Customer Segment Description
Description: Customer Segment Description based on the user selected language.
Select:
Decode(@prompt ('Select Language:', 'C', ['Dimensions/Language'], mono, constrained),
'EN', @select(Dimensions/Segment English Description),
'FR', @select(Dimensions/Segment France Description),
'DE', @select(Dimensions/Segment Dutch Description),
'AR', @select(Dimensions/Segment Arabic Description), @select(Dimensions/Segment English Description))

Linking Dimensions across different data providers.

2009-06-06T12:06:00.003+03:00

Some times you need to define more than one data provider in your report. A common example if you trying to build master detail report having actual data divided between 2 data base table. In this case you have to create 2 separate data providers. While you creating the second one, a pop up wizard will prompting you to link the common dimension between 2 data source. If there are common dimensions between the 2 data sources and with same of type, name and same list of values them the linking will done automatically without the need to prompt the user for action.

If you creating a report that reads data from one database data source and the other reads from an excel file, and there is a common dimension that you need to link with. Just open

Data --> View Data --> definition tab then select your dimension from the left pan.

You will notice the link button at the bottom of the page.

When you click on the link button a window with 2 pans will displayed. You can select the dimension that you want to link with from the right pan while you can select the dimension that you want to link to from the left pan. Click the link button in the middle area.

If you want to unlink 2 dimensions then Go to Data --> View Data --> Definition tab then select your dimension from the left pan and click unlink button.

Example # 1:

Universe # 1 (Current Account Universe)
Class	Object Name	Dimension / Measure	Type
Customer Segment	Segment Code	Dimension	Number
Customer Segment	Segment Description	Dimension	Char
Customer Sub Segment	Sub Segment Code	Dimension	Number
Customer Sub Segment	Sub Segment Description	Dimension	Char
Program	Program Code	Dimension	Number
Program	Program Description	Dimension	Char
Regional Information	Region	Dimension	Char
	Area	Dimension	Char
	Branch	Dimension	Char
Time	As of Date	Dimension	Date
	Week	Dimension	number
	Month	Dimension	char
	Quarter	Dimension	char
	Year	Dimension	number
Measures	Balance	Measure	number
Measures	profit	Measure	Number
Universe # 1 (Current Account Universe)
Class	Object Name	Dimension / Measure	Type
Customer Segment	Segment Code	Dimension	Number
Customer Segment	Segment Description	Dimension	Char
Customer Sub Segment	Sub Segment Code	Dimension	Number
Customer Sub Segment	Sub Segment Description	Dimension	Char
Program	Program Code	Dimension	Number
Program	Program Description	Dimension	Char
Regional Information	Region	Dimension	Char
	Area	Dimension	Char
	Branch	Dimension	Char
Time	As of Date	Dimension	Date
	Week	Dimension	number
	Month	Dimension	char
	Quarter	Dimension	char
	Year	Dimension	number
Measures	Credit card balance	Measure	number
Measures	Credit card profit	Measure	Number

Report # 1

	Current Account		Credit Card
Segment	Balance	Profit	Balance	Profit
High Affluent
Mass
Lower Mass

In this example we want to retrieve current account data from current account universe and to retrieve credit card data from credit card universe. Some one may ask why we don't keep both tables in the same universe and make just one query to retrieve date. This may happen because we want to make separate universes for separate products.

Create first data provider:

Universe: Current account

Dimensions: Year
Measures: Balance and profits.
Filter: year = 2009

Create second data provider:

Universe: Credit Cards
Dimensions: Year
Measures: Credit Cards Balance and Credit Cards profits.
Filter: Year = 2009

In this report the linking will done automatically because the year dimension have the same name, type and list of values across the 2 universes.

What if the segment dimension having different name in credit card universe? Customer group for example. In this case when you create the second data provider the linking window will displayed to prompt the user to select the linking dimension.

Example # 2:

Now if you have an excel file with loans data and you want to include loans information in your report.

Excel file Data

Segment	Balance	profit
High Affluent
Mass
Lower Mass

You will need to do as follow:

Open your report.
Create a new data provider based on another data source.
Select excel file.
Complete the wizard.
Go to Data --> View Data --> Definition tab then select your dimension from the left pan and click link button.
Link segment dimension form loans data provider with segment dimension from current account data provider.

Manage Security with business objects designer.

2009-06-03T17:50:00.001+03:00

If you are a universe designer, you will need to specify access levels for groups and users for your universe.

To manage user and group access to your universe do as follow:

Tools à Manage Security à Manage Access restrictions
The window divided into 2 pans
- The left pan displays available restrictions
- The right ban display users and group.

Restrictions:

Object level: you can restrict user or group from access a specific object.
Row level: you can restrict user or group from access specific rows from specific table based on criteria.
Table mapping: display data from another table for user or group.
General restriction: connection, control, SQL

Example:

We want to create the following restrictions:

Restrict user / group from accessing profit object. (object level)
Restrict user / group to display VIP customer's data based on VIP flag column. (row level)
Display data for retail users / groups from retail data table instead of displaying it from the corporate table. (table mapping)
Limit number of rows retrieved by user query (general)

To preview user and group access to your universe do as follow:

Tools à Manage Security àpreview net access restrictions.
Select user or group that you want to preview access for.
Click preview.

Notes:

You can create mixed restriction that mix many types of restrictions (object, row, table mapping and general)
If you apply the same restriction to more that one group then you can specify priority for them.

What is Scope of Analysis?

2009-06-03T15:14:00.005+03:00

Scope of Analysis is one of the most powerful features in business objects. Now you can easily build your custom hierarchies and use them in your report to navigate through data. To make it easy I will consider a small example then go step by step starting from building the universe and end with how to use the drill feature to navigate through data.

Example:

We have one fact table that contains daily sales and number of customers against branches.

Fact Table:

As of Date	Branch CD	Number of Customer	Total Sale
30-Nov-2008	101	10	4245
30-Nov-2008	102	30	543
30-Nov-2008	103	20	6434
30-Nov-2008	104	40	6543
30-Nov-2008	105	28	3264
30-Nov-2008	106	36	657
30-Nov-2008	107	48	677
30-Nov-2008	108	48	675
31-Dec-2008	101	45	546
31-Dec-2008	102	54	765
31-Dec-2008	103	63	658
31-Dec-2008	104	23	768
31-Dec-2008	105	98	876
31-Dec-2008	106	67	786
31-Dec-2008	107	34	453
31-Dec-2008	108	21	564
31-Jan-2009	101	23	546
31-Jan-2009	102	43	545
31-Jan-2009	103	67	654
31-Jan-2009	104	32	654
31-Jan-2009	105	45	675
31-Jan-2009	106	21	765
31-Jan-2009	107	45	675
31-Jan-2009	108	67	546
28-Feb-2009	101	64	657
28-Feb-2009	102	21	865
28-Feb-2009	103	36	324
28-Feb-2009	104	73	143
28-Feb-2009	105	15	432
28-Feb-2009	106	24	324
28-Feb-2009	107	53	324
28-Feb-2009	108	21	532
31-Mar-2009	101	64	321
31-Mar-2009	102	54	432
31-Mar-2009	103	23	543
31-Mar-2009	104	25	543
31-Mar-2009	105	53	546
31-Mar-2009	106	21	564
31-Mar-2009	107	68	544
31-Mar-2009	108	54	654
30-Apr-2009	101	34	546
30-Apr-2009	102	43	654
30-Apr-2009	103	21	567
30-Apr-2009	104	54	764
30-Apr-2009	105	21	675
30-Apr-2009	106	65	765
30-Apr-2009	107	32	765
30-Apr-2009	108	11	745
31-May-2009	101	33	546
31-May-2009	102	45	547
31-May-2009	103	56	564
31-May-2009	104	78	544
31-May-2009	105	98	654
31-May-2009	106	76	546
31-May-2009	107	65	654
31-May-2009	108	54	567

Branch hierarchy Table:

Region	Area	Branch CD	Branch
Central	First Area Riyadh	101	Branch 101
Central	First Area Riyadh	102	Branch 102
Central	Second Area Riyadh	103	Branch 103
Central	Second Area Riyadh	104	Branch 104
West	Jeddah	105	Branch 105
West	Jeddah	106	Branch 106
West	Mekkah	107	Branch 107
West	Mekkah	108	Branch 108

Time hierarchy Table:

As of Date	Month	Quarter	Year
30-Nov-2008	Nov	Q4	2008
31-Dec-2008	Dec	Q4	2008
31-Jan-2009	Jan	Q1	2009
28-Feb-2009	Feb	Q1	2009
31-Mar-2009	Mar	Q1	2009
30-Apr-2009	Apr	Q2	2009
31-May-2009	May	Q2	2009

Build The Universe:

To build the universe just adds the Fact Table and the 2 hierarchy tables. Make the proper joins and create 3 main classes:

Tables:

Fact Table.
Branch Hierarchy table.
Time hierarchy table.

Classes (Universe window should be like this):

Fact Table (class)

----------------------

| -----> As of Date (Dimension)

| -----> Branch Code (Dimension)

| -----> Number of Customers (Measure) [Sum]

| -----> Total Sales (Measure) [Sum]

Time Hierarchy (Class)

----------------------

| -----> As of Date (Dimension)

| -----> Month (Dimension)

| -----> Quarter (Dimension)

| -----> Year (Dimension)

Branch Hierarchy:

-----------------------

| ----->Region (dimension)

| ----->Area (dimension)

| ----->Branch CD (dimension)

| ----->Branch (dimension)

Joins:

First Table	Second Table	Join criteria
Fact Table	Branch Hierarchy table	Branch CD = Branch CD
Fact Table	Time hierarchy table	As of Date = As of Date

Hierarchies:

By default in business objects it consider every dimension created as a standalone hierarchy. By for some cases as our example you want to define a custom hierarchy like time, geographical information and customer segmentation…etc. in our example we need to build 2 custom hierarchies.

Click on hierarchies button from the editing tool bar or go Tools --> Hierarchies

Select the custom hierarchy option and then start building your hierarchy.

You can just add the Time Hierarchy class and Branch Hierarchy class from the left ban to the right ban. Or you can click new button from the right ban. Then type you hierarchy name. after that start adding dimensions from the left ban to your hierarchy tree. When you finish you can start order your hierarchy level by clicking move up and down buttons. Your hierarchy tree should be like this at the end:

Branch Hierarchy:

-----------------------

| --------- > Region

| --------- >Area

| --------- >Branch

Time Hierarchy:

----------------------

| --------- > As of Date

| --------- > Month

| --------- >Quarter

| --------- >Year

Note That if you even store all your data as detailed in one big table you will be able to build your hierarchy based on dimension that you will select to construct your customer hierarchies. For example assume if the fact table looks like this.

As of date	Month	Quarter	Year	Branch	Area	Region	No of Cust	Total Sales

This table is no more fact table and we call it a detail table.

Scope of analysis:

Now after we finish bulding our univers let go and build our report. Assume that we just need to build a very simple report like this

Year	2008		2009
Region	No of Cust	Total Sales	No of Cust	Total Sales
center	342	23432	234	23432
west	234	23424	424	324

Create a new report based on our universe that we build earlier. Select region, year, number of customer and total sales then run the report. After that click on the drill icon (Analysis --> Drill) and then right click on any of the selected dimension (region, year) you will find that there is no option to drill by.

Now edit data provider (Data --> Edit data provider) and click view SQL button for the tool bar. You will find that the data selected are just the four columns that we have selected. Now to find what Scope of Analysis do click on the Scope of Analysis button and select both time and branch hierarchies. Please note that you can expand the hierarchies build before and select a specific dimensions. Let's assume that you want to analysis your figures by quarter, and year only and you are not interested in monthly figures or daily figures. Then all what you need is to put a check mark in the check box beside year and quarter in the time hierarchy tree when you open the scope of analysis window.

Now click again on view SQL button and note that all dimension with a check make beside in scope of analysis added to your SQL statement even if you didn't select them in the query builder window (remember that we have selected 2 dimensions [region and year])

Run the report after that.

Drill:

After running the report click on drill icon (analysis à drill) and then right click on region you will find that you can do the following:

Drill down: will go to the second level of the hierarchy. In our example it will display the area information and note that there is a drill down filter (combo box) displayed in the drill down filter tool bar. If you right click again on are and select drill down you will go to the second level (branch) and so on. If you filter on a specific region from the drill down menu it will just display the selected region date.
Drill up: will go a level up. If you were in area and click on drill up it will return back to region.
Drill by: a menu will displayed contains all hierarchy that you have selected in scope of analysis. The menu should be like this if you right click on region dimension
Drill by:
Area
Branch
----------
Quarter

Drill through: to select another dimension that you didn't select in scope of analysis. In our example it will be month because we select year and quarter only from the time hierarchy tree.

Scope of analysis: if you want to update hierarchies that you want to analysis your data according to.
Set as drill filter: make this column a drill filter that you can filter your report data from it.

Universe

2008-11-22T15:54:00.000+03:00

A universe is a data abstraction layer which is built with the BusinessObjects Designer tool. The universe dose not contains any data itself, it only acts as an interface. This layer is where you define your "Business Objects" [Dimensions, Measures, Details and Filters] according to your business requirements. You will need also to define a valid database connection to the database schema that you want to use to build your universe. Then you can select tables you want or create new derived tables from the existing ones. You can also create joins between tables or let the designer tool automatically detect any reference join defined in the database schema by a reference constrain. The good point is, if the schema for the data base is changed later you will not need to go through all old reports that you build to change them. Just when you change them once in universe it will be reflects in all reports.

This is a simple way to understande the roll of the universe
Database --> Schema (Tables + Columns) --> Universe (Objects [Dimentions, measures, detailes and filters]) --> report.

Convert Balance

2008-09-01T11:35:00.000+03:00

Introduction:

While you build your report you may face a case that you want to convert balance or money amount from currency to another. This case comes from that fact that when you want to store balances you store the amount in a decimal type column in your database and you store the currency in another column. This gives you the flexibility to use any convert balance function later to display your balance in your report currency. So we accept that you may display your balances in report with currency (report currency) that differs from the currency stored in the database (database currency).

Report Level Convert:

Business objects presents 4 functions that you can use to convert your balance. But there is two restrictions to use these functions.

Functions:

EuroConvertFrom(balance_amount ; from_currency_ISO ; Decimals _Blaces)

2) EuroConvertFrom(balance_amount; from_currency_ISO ; Decimals _Blaces)

3) EuroFromRoundError(balance_amount; from_currency_ISO ; Decimals _Blaces)

4) EuroToRoundError(balance_amount; from_currency_ISO ; Decimals _Blaces)

Restrictions:

1) You can convert from or to Euro only.

2) You have to select one of the following currencies (Reason: those European nation currencies have a fixed exchange rate with Euro)

Currencies table:

(code):country(currency)

BEF: Belgian (franc)

DEM: German (mark)

GRD: Greek (drachma)

ESP: Spanish (peseta)

FRF: French (franc)

IEP: Irish (punt)

ITL: Italian (lira)

LUF: Luxembourg (franc)

NLG: Dutch (guilder)

ATS: Austrian (schilling)

PTS: Portugese (escudo)

FIM: Finnish (mark)

Universe Level Convert:

You have to write a database procedure and use it whenever you need to convert balance. You can write a tailored procedure or a general use one. Here is a small pseudo code that you can use:

Convert_balance (p_balance, p_from_currency, p_to_currency, p_as_of_date) return converted _balance

{

Select exchange_rate

Into v_exchange_rate

From daily_exchange_rate_table a

Where a.as_of_date = p_As_of_date

And a.from_Currency = p_from_currency

And a.to_currency = p_to_currency

Return p_balance * v_exchange_rate;

}

@Aggregate_Aware

2008-08-25T17:51:00.000+03:00

Function Name: @Aggregate_Aware
Description:

This function used to increase the performance by directing the report query to select objects form the aggregated table instead of the detailed table.

It may be also used to make some objects incompatible with some others. The best example for such case is the stand alone table that you add in your universe for a specific report. For sure you don’t want the user select other incompatible objects from other incompatible tables.

Example # 1: (performance)

[UNIVERSE TABLES]

Assume that you have universe with 2 tables:

BANK_TRANSACTION_DAILY (

AS_OF_DATE,

TRANSACTION_ID,

PARTY_ID,

ACCOUNT_NUMBER,

BRANCH,

TRANSACTION_AMOUNT

)

This table contains detailed daily information for the transaction. The second table

BANK_TRANSACTION_MONTHLY (

MONTH,

BRANCH,

NO_OF_TRANSACTION,

TRANSACTION_AMOUNT

)

This is an aggregated table that aggregates this data monthly.

[UNIVERSE OBJECTS]

Dimensions:

Date: @Aggregate_Aware (BANK_TRANSACTION_MONTHLY. AS_OF_DATE, BANK_TRANSACTION_DAILY. MONTH)

Transaction ID: BANK_TRANSACTION_DAILY. TRANSACTION_ID

Party ID: BANK_TRANSACTION_DAILY. PARTY_ID

Account Number: BANK_TRANSACTION_DAILY. ACCOUNT_NUMBER

Branch: @Aggregate_Aware (BANK_TRANSACTION_MONTHLY. BRANCH, BANK_TRANSACTION_DAILY. BRANCH)

Measures:

Number of Transactions: BANK_TRANSACTION_MONTHLY. NO_OF_TRANSACTION

Transaction Amount: @Aggregate_Aware (BANK_TRANSACTION_MONTHLY. TRANSACTION_AMOUNT

, BANK_TRANSACTION_DAILY. TRANSACTION_AMOUNT)

Note:

1 When we use the ‘@Aggregate_Aware’ function we consider putting the faster column (aggregated) first. So we put the monthly column then the daily one.

2 We can use more than one aggregated table. In our example we can add a yearly level aggregated table and so when we use the ‘@Aggregate_Aware’ the yearly column will come first cause it is the faster one among (most aggregate) all the three tables.
Ex: Transaction Amount: @Aggregate_Aware (BANK_TRANSACTION_YEARLY. TRANSACTION_AMOUNT, BANK_TRANSACTION_MONTHLY. TRANSACTION_AMOUNT, BANK_TRANSACTION_DAILY. TRANSACTION_AMOUNT)

[REPORTS]

Report # 1:

Objective: display Transaction amount per branch monthly wise (Month, Branch, and Transaction Amount)

Screen:

Month	Branch	Transaction Amount
Jan	Bran#1	10
Jan	Bran#2	15
Feb	Bran#1	20
Feb	Bran#2	10

Query Panel:

Objects: (Date, Branch and Transaction Amount)

Filter : N/A

Description: Now we have selected 3 objects. The business objects engine has two ways to calculate data in this report. The first one is to go to the daily table and to sum the transaction amount grouped by the branch and month. The second is simply to go to the aggregate table (BANK_TRANSACTION_MONTHLY) in our case and just retrieve the required data directly without need to sum. Note that number of rows in BANK_TRANSACTION_MONTHLY less than number of rows in the daily one and so there will be a great increase in query performance if it runs against the monthly (aggregated) table.

Report # 2:

Objective: display Transaction amount per Customer monthly wise (Month, Branch, and Transaction Amount)

Screen:

*Month*	*Customer*	*Transaction Amount*
Jan	Cust#1	10
Jan	Cust#2	15
Feb	Cust#1	20
Feb	Cust#2	10

Query Panel:

Objectsà (Date, Party ID and Transaction Amount)

Filter à N/A

Description: Now look carefully for this report. The date information can be retrieved from the two tables (Monthly and daily) this is the same also for the Transaction Amount object. But what about the Party ID object. This information is only available in the detailed table (Daily one) and there is no other place to retrieve this information from. So business object query engine will go to the bad choice because this is the only way to calculate data in this report.

[NOTES]

1 @Aggregate_Aware function tells business objects that it can calculate the column value from many tables.

2 Business objects query engine decide based on the objects chosen in the report witch table to use to retrieve data from (aggregated table or the detailed one) based on columns passed as parameters to @Aggregate_Aware function.

3 @Aggregate_Aware function is basically used to increase performance by target the report’s query to the fastest table when it is visible.

4 Use Toolsà aggregate navigation [Designer Application] to set the incompatible objects with your tables and incompatible tables with your objects.

Example # 2: (incompatible)

If you have in your universe a standalone table that used for some specific and tailored report, and you don’t want the end user to miss up with objects from the stand alone table and from the other tables in the universe. Just do as follow:

1 Define all stand alone table use the @Aggregate_Aware function Ex: Revenueà @Aggregate_Aware(STANDALONE_TABLE.REVENUE)

2 Use Toolsà aggregate navigation [Designer Application] to set the incompatible objects with your tables and incompatible tables with your objects.

3 While you build your report if you choose and object form the stand alone table in the query panel and then tried to select other incompatible objects from another table you will receive an incompatible objects error message and it will force you to use objects from the standalone table only or form other universe tables only.

END: For any information or question related for this topic just leave a comment or send me e-mail tahaqp@gmail.com