Skip Headers
Oracle® Fusion Middleware User's Guide for Oracle Business Intelligence Enterprise Edition
11g Release 1 (11.1.1)

Part Number E10544-04
Go to Documentation Home
Home
Go to Book List
Book List
Go to Table of Contents
Contents
Go to Index
Index
Go to Master Index
Master Index
Go to Feedback page
Contact Us

Go to previous page
Previous
Go to next page
Next
PDF · Mobi · ePub

2 Creating Analyses

This chapter describes how to create analyses in Oracle Business Intelligence Enterprise Edition, including how to specify the criteria for analyses, display the results, and examine the SQL statements. It also explains subject areas and columns, the double column feature, and variables.

This chapter includes the following sections:

What Are Analyses?

An analysis is a query against an organization's data that provides answers to business questions. A query contains the underlying SQL statements that are issued to the Oracle BI Server.

Analyses let you explore and interact with information by visually presenting data in tables, graphs, pivot tables, and so on. You can save, organize, and share the results of analyses.

Analyses that you create can be saved in the Oracle BI Presentation Catalog and integrated into any Oracle BI EE dashboard. Analyses can be enhanced through features such as graphs, result layout, calculated items, and drilling.

How Do I Work with Analyses?

To work with analyses, you use one of the following editors:

  • The "Analysis editor" — A more complex editor that allows you to explore and interact with information by visually presenting data in tables, graphs, pivot tables, and so on. For more information, see "What Is the Analysis Editor?"

    You have access to this editor only if you have been granted the Access to Analysis privilege by the administrator.

  • The "BI Composer wizard" — A simple-to-use wizard that allows you to quickly and easily create, edit, or view analyses without the complexities of the Analysis editor. For more information, see Chapter 14, "Using BI Composer to Work with Analyses."

    You have access to this editor only if you have been granted the Access to BI Composer privilege by the administrator.

You specify which editor you want to use on the "My Account dialog: Preferences tab". However, if you have turned on accessibility mode either in the "Sign In page" or the "My Account dialog: Preferences tab", then the BI Composer wizard in accessibility mode will be used as the analysis editor, regardless of your choice.

How Are Analyses Processed?

When an analysis is processed, the request is sent in the form of logical SQL statements to the Oracle BI Server. The BI Server generates queries against the appropriate data sources. When the BI Server receives the query results, they are in a format that is not yet suitable for returning to the user. The BI Server merges the results and applies any additional calculations or filters that pertain to the results. The BI Server then sends the data to Oracle BI Presentation Services to format the results for display.

Embedding an Analysis in a Dashboard

Embedding an analysis in a dashboard causes it to execute automatically and display the most current results every time the dashboard is accessed. This provides access to the most current results. For example, if you are a sales executive whose company captures sales data on a daily basis, then you might want to have the dollar volume of product that was sold today be displayed on the front page of your dashboard.

You can embed saved analyses by using the Dashboard editor. For information about the Dashboard editor, see "Adding Content to Dashboards".

What Are Subject Areas and Columns?

A subject area contains folders, measure columns, attribute columns, hierarchical columns, and hierarchy levels that represent information about the areas of an organization's business or about groups of users with an organization. Subject areas usually have names that correspond to the types of information that they contain, such as Marketing Contracts, Service Requests, and Orders.

A subject area corresponds to the presentation layer in an Oracle BI metadata repository. In a repository, the subject area is the highest-level object in the presentation layer and represents the view of the data that end users see when they create or edit an analyses.

Individuals who design and build metadata repositories (such as a Business Intelligence strategist, metadata provider, or ETL developer) create subject areas using the Oracle BI Administration Tool. Generally, rather than creating one large subject area for their company's data, they create multiple smaller subject areas. This enables them to provide a particular group of users or a particular area of a company with the most important data that they need in one small subject area and the less important data in one or more related subject areas created from the same business model layer. Having these smaller subject areas makes it easier for users to find the data they need. It also makes it easier to maintain the data. For more information, see "Creating Subject Areas" in Oracle Fusion Middleware Metadata Repository Builder's Guide for Oracle Business Intelligence Enterprise Edition.

Note:

The individuals who design and build metadata repositories can specify that a subject area, folder (and its children), or column (both attribute and hierarchical) is to be hidden. A hidden subject area, folder, or column is not visible in the "Subject Areas pane" but is visible elsewhere, such as in an analysis or saved filter contents. (Because the object is still visible elsewhere, hiding a subject area, folder, or column in this way is not a solution for security or access control.)

If the criteria of an existing analysis includes a subject area, folder, or column that is subsequently hidden, the analysis is still accessible but the subject area, folder, or column is no longer visible in the Subject Areas pane of the "Analysis editor: Criteria tab".

Columns contain the individual pieces of data that an analysis returns. Columns usually have names that indicate the types of information that they contain, such as Account or Contact. Together with filters and selection steps, columns determine what data an analysis contains.

When you create a new analysis, filter, or dashboard prompt, you first select the subject area with which you want to work. This is known as the primary subject area and is displayed in the "Subject Areas pane". If, as you work, you find that you need more data, you can add additional subjects areas that are related to the primary subject area that you have chosen. (You can add related subject areas only if they are available for the primary subject area and only if you have permission to access them.)

Typically, when you query a single subject area, all the measure columns that are exposed in that subject area are compatible with all the attribute columns and hierarchical columns that are exposed in the same subject area. However, when you combine columns from multiple subject areas, you must ensure that you do not include combinations of measure columns with attribute columns and hierarchical columns that are incompatible with one another.

For example, a measure column in one subject area might not be associated with the Project attribute column. If measure columns associated with the Project attribute column from another subject area are added to the analysis along with columns that are not associated with Project, then the query might fail to return results, or cause the BI Server error "No fact table exists at the requested level of detail: XXXX."

For an analysis to return data, you must select at least one column to include in the analysis.

What Are the Types of Columns?

Subject areas contain the following types of columns:

  • Attribute Column — Holds a flat list of values that are also known as members. No hierarchical relationship exists between these members, as is the case for members of a hierarchical column. An attribute column was referred to as a presentation column in previous releases (prior to 11g).

    Examples include ProductID or City.

  • Hierarchical Column — Holds data values that are organized using both named levels and parent-child relationships. This column is displayed using a tree-like structure. Individual members are shown in an outline manner. Hierarchies allow you to drill deeper into the data, to view more detailed information. Examples include Time or Geography. The following image shows the Time and Fiscal Time hierarchies in the Subject Areas pane.

    This image is described in the surrounding text.

    A hierarchical column can be one of the following kinds:

    • Level-based hierarchy — Consists of an ordered set of two or more levels. For example, a Time hierarchy might have three levels for Year, Quarter, and Month. Level-based hierarchies can also contain parent-child relationships.

    • Parent-child hierarchy — Consists of values that define the hierarchy in a parent-child relationship and does not contain named levels. For example, an Employee hierarchy might have no levels, but instead have names of employees who are managed by other employees. Employees can have titles, such as Vice President. Vice Presidents might report to other Vice Presidents and different Vice Presidents can be at different depths in the hierarchy.

    In addition to being level-based or parent-child, hierarchical columns can be one of the following:

    • Ragged — A hierarchy in which all the lowest-level members do not have the same depth. For example, a Time hierarchy might have data for the current month at the day level, the previous month's data at the month level, and the previous 5 years' data at the quarter level. This type of hierarchy is also known as an unbalanced hierarchy.

    • Skip-level — A hierarchy in which certain members do not have values for certain higher levels. For example, in the United States, the city of Washington in the District of Columbia does not belong to a state. The expectation is that users can still navigate from the country level (United States) to Washington and below without the need for a state.

  • Measure Column — Holds a simple list of data values. It is a column in an Oracle BI EE repository, usually in a fact table, that can change for each record and can be added up or aggregated in some way. Examples include Revenue or Units Sold.

Throughout this guide, the term "column" on its own generally refers to all three types. Names for specific types of columns are included where necessary.

How Are Columns Indicated Visually?

Each type of column is indicated by its own icon in places such as the Subject Areas pane and Layout pane. You can expand level-based hierarchies and see their levels. Parent-child hierarchies are shown as hierarchical columns that have no levels. Figure 2-1 shows the icons and names of various columns.

Figure 2-1 Column Types With Their Icons

This image is described in the surrounding text.

How Will Measure Columns Created in Previous Releases Be Upgraded?

In previous releases (prior to 11g) of Oracle BI EE, measure columns could easily be treated as attribute columns, which allowed you to move them freely among the edges of views. This release introduces functionality that specifies to not show all the detail when a measure column is moved to an edge but rather to aggregate the measure column to the grain of the edge. During upgrade, all measure columns have the Treat as an Attribute Column box selected in the "Edit Column Formula dialog: Column Formula tab". For more information, see "Upgrading Measure Columns" in Oracle Fusion Middleware Upgrade Guide for Oracle Business Intelligence.

Understanding the Double Column Feature

Oracle BI EE provides a feature called double columns. When a repository is configured for the double column feature, column data includes a display column that has a code column mapped to it (that is, it has double columns). A display column contains the display values for the column data, for example, Excellent, Good, and Poor. A code column contains code values that uniquely identify display values and are consistent across users or locales, for example, the code values 1 (uniquely identifying Excellent), 2 (uniquely identifying Good), and 3 (uniquely identifying Poor).

When Oracle BI EE processes a double column, for example, as part of the criteria of an analysis or as the basis for a filter, it generates and issues SQL statements to the Oracle BI Server that use code values rather than display values, thereby making the filter language independent.

The double column feature lets you:

In addition, if your organization allows the display of code values within Oracle BI EE, you can use code values rather than the display values in some dialogs, such as the "New Filter dialog". This means, for example, when specifying the values to use when applying a filter, you can specify code values rather than display values.

Before you can take advantage of the double column feature, your administrator must configure your repository by mapping code columns to display columns. Check with the administrator to see if your repository is configured for double columns.

What Is the Analysis Editor?

The "Analysis editor" lets you explore and interact with information by visually presenting data in tables, graphs, pivot tables, and so on. You can include the views that you create in an analysis for display in dashboards.

The Analysis editor contains the following tabs:

The tabs of the Analysis editor are organized into various panes and areas. As you work, you can manage these panes to suit your needs. See "Managing Panes in the Analysis Editor Tabs".

You access the Analysis editor when you create (or edit) an analysis. See "Creating New Analyses".

Note:

If you are using Oracle BI EE in accessibility mode, then, when you create (or edit) an analysis, the Analysis editor is replaced by the "BI Composer wizard". For more information on BI Composer, see Chapter 14, "Using BI Composer to Work with Analyses." For more information on accessibility mode, see Appendix C, "Accessibility Features."

Managing Panes in the Analysis Editor Tabs

Each of the tabs in the Analysis editor consists of several panes. For example, the Results tab consists of the "Subject Areas pane", the "Catalog pane", the "Views pane", the "Compound Layout", and the "Selection Steps pane".

As you work, you can manage these panes to give yourself the most efficient work area for the task that you are performing. For example, if you add multiple views to the compound layout on the Results tab, then you might want to hide the other panes in the Results tab to maximize the area in which to see the views that you are adding. You can:

  • Show or hide the panes that are displayed on the left side of the Analysis editor by clicking the sideways triangle button that is displayed between the left and right sides.

  • Expand or collapse an individual pane by clicking the plus or minus button in the title bar of the pane.

  • Resize an expanded pane by dragging its splitter. For example, you can drag the splitter on top of the Catalog pane up to increase the height of the pane.

  • Show or hide a pane by clicking its show/hide toolbar button, if one is available for the pane on the toolbar. For example, you can show or hide the Filters pane in the Criteria tab by clicking the Show/Hide Filters pane button on the toolbar.

What Is the Process for Constructing Analyses?

Note:

If you are using Oracle BI EE in accessibility mode or you have selected Wizard (limited functionality) as the analysis editor, then you use the "BI Composer wizard" rather than the "Analysis editor" to construct analyses. For more information on BI Composer, see Chapter 14, "Using BI Composer to Work with Analyses." For more information on accessibility mode, see Appendix C, "Accessibility Features."

The process for constructing an analysis includes the following steps:

  1. "Creating New Analyses"

  2. "Specifying the Criteria for Analyses"

  3. "Displaying the Results of Analyses"

  4. "Adding Prompts to Analyses"

  5. "Examining the Logical SQL Statements for Analyses"

  6. "Saving Analyses"

Creating New Analyses

This procedure is a step in the process for constructing an analysis. For more information, see "What Is the Process for Constructing Analyses?"

To create an analysis:

  1. In the global header, click New, then Analysis, then one of the following options:

Specifying the Criteria for Analyses

This procedure is a step in the process for constructing an analysis. For more information, see "What Is the Process for Constructing Analyses?"

You use the "Analysis editor: Criteria tab" to specify the criteria for an analysis, including columns, filters, and selection steps. You also can specify:

To specify the criteria for an analysis:

  1. In the Criteria tab, select the columns to be included in the analysis by doing one of the following:

    • Double-clicking the columns in the "Subject Areas pane".

    • Dragging and dropping the columns from the Subject Areas pane to the "Selected Columns pane".

      To select multiple non-contiguous columns, press and hold the Ctrl key, then click each column to include.

    Note:

    In general, the analysis should contain at least one measure column. Measure columns are the key additive measurements of business performance, such as dollar sales per store, or the number of service requests opened and closed each day. An analysis without any measure columns generally is not meaningful, and can cause poor query performance or unexpected results. If you want to build an analysis without any measure columns, then first consult the administrator.

    The selected columns are displayed in a box in the Selected Columns pane. Each column box has two sections. The upper section shows the name of the folder that contains the column, for example, Customers. The lower section shows the name of the column, for example, Customer Name.

    If you want to:

    • Add or remove related subject areas from which to select columns, click the Add/Remove Subject Areas toolbar button in the Subject Areas pane to display the "Add/Remove Subject Areas dialog".

      If you add a related subject area but do not add any columns from that subject area to the analysis, then the subject area is not related to the analysis after you close and re-open the analysis.

    • Refresh the content in the Subject Areas pane, click the Refresh toolbar button on the Subject Areas pane or click the arrow beside the button.

      Clicking the button executes the default behavior of Refresh Display. Clicking the arrow enables you to select Refresh Display or Reload Server Metadata to refresh the subject area's metadata.

      For more information about these options, see the "Subject Areas pane".

  2. Modify the columns or levels of hierarchical columns as needed using the Selected Columns pane:

    • Click the Options button to the right of a column name in the Selected Columns pane to display options to:

      • Specify the sort order for columns. (You cannot specify the sort order for hierarchy levels.)

        For more information, see "Sorting Data in Views".

      • Edit formulas for attribute columns and measure columns, including customizing headings, and specifying the aggregation rule. (You cannot customize headings, specify the aggregation rule, or edit the formulas for hierarchical columns or for hierarchy levels.)

        For more information, see "Editing the Formula for a Column".

      • Edit column properties to control the formatting and interaction of columns and hierarchy levels.

        For more information on applying formatting, see "Applying Formatting to a Column".

      • Add filters for attribute columns and measure columns. (You cannot add filters for hierarchical columns or hierarchy levels.) For more information, see "Creating Column Filters".

      • Delete the columns from the analysis. (You cannot delete hierarchy levels.)

    • Click the Remove all columns from criteria toolbar button in the Selected Columns pane to remove all columns from the analysis.

    • Click the Combine results based on union, intersection, and difference operations toolbar button in the Selected Columns pane to combine the results of two or more analyses into a single result. For more information, see "Combining Columns Using Set Operations".

    • Use the Drag Column buttons in the Selected Columns pane to place the columns in the default order for display in the analysis results.

  3. Add and edit inline filters as needed using the "Filters pane".

    For more information, see "Creating Column Filters".

  4. Create or edit selection steps as needed using the "Selection Steps pane".

    For more information, see "Working with Selections of Data".

  5. Add named filters, calculated items, and groups from the Oracle BI Presentation Catalog as needed using the "Catalog pane".

  6. Use the buttons on the toolbar for the Criteria tab, as needed, to show or hide the Filters pane, to show or hide the Selections Steps pane, and to edit the properties of the analysis, such as the type of message (default or custom) to be displayed if no results are available.

You can now add views to the analysis. See "Displaying the Results of Analyses".

Note:

If you add a column from the Subject Areas pane to the Selected Columns pane after displaying the analysis results, then the column is either included (that is, displayed in the view) or excluded from existing views, depending on the setting of the Display of Columns Added in Criteria tab option in the "Analysis Properties dialog: Data tab":

  • The column is included in existing views as well as in any new views that you add, if the Display in existing and new views option is selected for the Display of Columns Added in the Criteria tab option.

  • The column is excluded from existing views (that is, it is placed in the Excluded drop target of the Layout pane) but included in any new views that you add, if the Exclude from existing views, but display in new views option is selected for the Display of Columns Added in the Criteria tab option.

    For more information on the Excluded drop target, see "Understanding Drop Targets".

Editing the Formula for a Column

You can edit the formulas for attribute columns and measure columns when you specify the criteria for an analysis. This editing affects the column only in the context of the analysis and does not modify the formula of the original column in the subject area. You can also customize table and column headings and specify the aggregation rule for column totals. (This functionality is not available for hierarchical columns.)

A column formula specifies what the column values represent. In its most basic form, such as "Base Facts"."1-Revenue", a column takes the data from the data source as is. You can edit the formula to add functions, conditional expressions, and so on. This editing enables you to present analysis results in a variety of ways. For example, suppose that you want to perform what-if analysis and show what the revenue might be by product if you increased revenue by 10%. You can see this increase by changing the formula for the Revenue column to show the revenue increased by 10%. Figure 2-2 shows an analysis in a pivot table that includes the Revenue column (which shows the revenue in the data source) and the Revenue Increased by 10% column, where the formula for the Revenue column was edited to calculate revenue increased by 10%.

Figure 2-2 Analysis Showing Revenue and Revenue Increased by 10%

This image is described in the surrounding text.

To edit the formula of a column:

  1. In the "Selected Columns pane", click the Options button beside the column whose formula you want to edit and select Edit Formula. The "Edit Column Formula dialog" is displayed.

  2. Use the "Edit Column Formula dialog: Column Formula tab" to perform various tasks such as creating customized headers and creating or editing the formula for the column. You can build a simple mathematical formula using operator and character buttons, such as "Base Facts"."1-Revenue"*1.10.

  3. Optionally, use the "Edit Column Formula dialog: Bins tab" to combine values for the column into sets.

  4. Click OK.

    The column formula is saved with the analysis with which it is used.

Related Topics


"What Are Analyses?"
"Specifying the Criteria for Analyses"

Combining Columns Using Set Operations

After you have selected a subject area for an analysis, you can combine columns from one or more subject areas using Set operations such as Union or Intersect. By combining columns, you create a new column for displaying the data in a different way.

Guidelines for Selecting Columns to Combine

When selecting columns to combine, keep the following guidelines in mind:

  • The number and data types of the columns to combine must be the same. The number of rows that are returned for each column can differ.

  • You can select columns from the same subject area or from a different subject area, but the columns must have some commonality.

  • You can specify one Set operation for one collection of criteria. For example, if you create criteria from the A-Sample Sales subject area, you can apply only one Set operation to those columns. You cannot apply different Set operations to different columns in the collection of criteria.

  • You cannot use hierarchical columns, selection steps, or groups when you combine criteria.

Difference Between Combining Columns Using Set Operations and Adding Columns from Related Subject Areas

Combining columns using Set operations produces different results than adding columns from related subject areas:

  • When you combine columns using Set operations, the analysis results show a single newly combined column governed by a Set operation. For example, see "Example: Combining Columns from One Subject Area".

  • When you add columns from related subject areas to an analysis, the results show each added column individually. For example, if you have the appropriate permissions, then you can create an analysis by selecting one column from a primary subject area and selecting another column from a related subject area.

    Figure 2-3 shows a table with columns from related subject areas. The Revenue measure is stored in the A-Sample Sales subject area and the Quota Amount measure is stored in the B-Sample Quotas Subject Area.

Figure 2-3 Measure Columns from Related Subject Areas

Description of Figure 2-3 follows
Description of "Figure 2-3 Measure Columns from Related Subject Areas"

For information, see "What Are Subject Areas and Columns?"

Example: Combining Columns from One Subject Area

The A-Sample Sales subject area contains the Offices folder, which contains the D1 Office and D2 Department columns. You can combine these two columns and create a new column called Offices & Departments. You can include the Union All Set operation to specify that this new column shows all the values from both columns in a single column in a table.

Combining Columns from One or More Subject Areas

The following procedure describes how to combine columns from one or more subject areas. Some steps include references to the example of creating the Offices & Departments column with the A-Sample Sales subject area.

Note:

Data formatting that has been saved as the systemwide default for a column is not reflected in a combined column. If you want the data formatting in the combined column, then you must reapply it to the combined column. For more information on data formatting, see "Column Properties dialog: Data Format tab".

To combine columns from one or more subject areas:

  1. In the Analysis editor, create an empty analysis that uses a subject area such as A-Sample Sales.

  2. In the Criteria tab, select the columns to include in the analysis. For example, select D1 Office from the Offices folder and 1-Revenue from the Base Facts folder.

  3. In the "Selected Columns pane", click the Combine results based on union, intersection, and difference operations toolbar button to display the Select Subject Area menu.

  4. Select a subject area that contains the columns to combine with the columns that you have previously included. For example, click A-Sample Sales.

    The Set Operations area is displayed in the Selected Columns pane. Note the boxes with dotted line borders that are displayed below the criteria. These boxes indicate the kind of column that you must combine with those that you have previously included. For example, the boxes might include "Add Column (D1 Office)" and "Add Column (1-Revenue)". This text indicates that the columns that you include in each of those boxes is combined with the previously selected D1 Office and 1-Revenue columns using a Set operation to form a new column. When you combine measure columns, no arithmetic operations are performed.

  5. In the Subject Areas pane, select the columns to combine with the originally selected columns. For example, from the Offices folder, select D2 Department and from the Base Facts folder, select 1-Revenue.

    Note that the boxes that previously had dotted line borders now hold the columns that you have just selected. You have now specified the columns to combine.

  6. Click the Union button under the Result Columns link. Select the operation type to use for combining the columns. For example, select the Union All type.

    The result columns are those that are displayed in views after applying the set operation of the derived columns.

  7. Click the Result Columns link. Note that the Selected Columns pane is updated to show the newly combined columns that you have just created. You can work with these columns as you do other columns in this pane. For example, you can rename the first column (that is the single newly combined column) by following these steps:

    1. Click the Options button for the D1 Office column.

    2. Select Column Properties.

    3. Select Column Format.

    4. Ensure that Custom Headings is selected.

    5. In the Column Heading box, enter Offices & Departments.

    6. Click OK.

  8. Click the Results tab to view the columns in a table in the "Compound Layout".

Figure 2-4 shows how the combined Offices & Departments column is displayed in a table.

Figure 2-4 Combined Column Displayed in a Table

Description of Figure 2-4 follows
Description of "Figure 2-4 Combined Column Displayed in a Table"

Viewing Metadata Information from the Subject Areas Pane

You can view metadata information for subject areas, folders, columns, and hierarchy levels. This is helpful, for example, if you need a full description of one of these objects when creating an analysis.

Metadata information is contained in a metadata dictionary. A metadata dictionary is a static set of XML documents. Each XML document describes a metadata object, such as a column, including its properties and relationships with other metadata objects.

Note:

In order for you to view metadata information, the administrator must have performed various configuration tasks, as described in "Providing Access to Metadata Dictionary Information" in Oracle Fusion Middleware System Administrator's Guide for Oracle Business Intelligence Enterprise Edition.

To view metadata information:

  1. In the Subject Areas pane in the Criteria tab or the Results tab, select the subject area, folder, column, or hierarchy level for which you want to view metadata information.

  2. Click View Metadata Dictionary on the toolbar.

    The metadata dictionary information for the object is displayed either in a new browser window or in a tab, depending on the browser setting.

    Note:

    If the metadata dictionary information does not display in the browser, then ensure that the browser setting to access data sources across domains is enabled.

Displaying the Results of Analyses

This procedure is a step in the process for constructing an analysis. For more information, see "What Is the Process for Constructing Analyses?"

You use the "Analysis editor: Results tab" to add different views of the results such as graphs, tickers, and pivot tables. For a list of the views that you can add, see "What Types of Views Are Available?"

When you create an analysis and display the Results tab, you see in the "Compound Layout" a title view and either a table or pivot table view by default, based on the following:

The administrator might have configured different views to display by default in the Results tab.

You can combine views and position them anywhere in the Compound Layout. For example, you can create side-by-side pivot tables that reflect different views of the data, graphs that allow you to explore interrelationships in depth, and filters that limit the results.

Together all of these views are called a compound layout. You can create additional compound layouts, as described in "Creating Additional Compound Layouts".

To display the results of an analysis:

  1. Click the Results tab.

    The results of the analysis are displayed in a table or pivot table.

  2. (Optional) Edit the table or pivot table or add additional views. For more information, see "Adding Views for Display in Dashboards".

Creating Additional Compound Layouts

As you work with an analysis, you can create multiple compound layouts that use different combinations of views. For example, one layout might include a graph and a title, and another layout might include a graph and a narrative view. When you add the analysis to a dashboard page, you can select which layout you want to include on that page.

To create additional compound layouts:

  1. Click the "Analysis editor: Results tab".

  2. Create another compound layout in the "Compound Layout" by clicking one of the following buttons on the toolbar of the Results tab:

    • Create Compound Layout — Creates a new instance of the compound layout.

    • Duplicate Compound Layout — Creates a copy of the current compound layout, with the same views.

Alerting Users to No Data in the Results of Analyses

When the results of an analysis return no data, the following default message is displayed to users:

No Results
The specified criteria didn't result in any data.
This is often caused by applying filters and/or selections that are too restrictive or that contain incorrect values.
Please check your Analysis Filters and try again.
The filters currently being applied are shown below.

Rather than display the default message, you can create a customized message to alert users. This message enables you to provide your own explanation for why the analysis returned no data.

To create a custom message to alert users to no data in the results of an analysis:

  1. Edit the analysis to which you want to add a custom message.

  2. Click the "Analysis editor: Results tab".

  3. Click the Analysis Properties toolbar button. The "Analysis Properties dialog" is displayed.

  4. In the No Results Settings box, select Display Custom Message.

  5. In the Header field, enter the text of the header for the custom message.

  6. In the Message field, enter the explanatory text.

  7. Click OK.

Adding Prompts to Analyses

This procedure is a step in the process for constructing an analysis. For more information, see "What Is the Process for Constructing Analyses?"

You use the "Analysis editor: Prompts tab" to create prompts that allow users to select values to filter an analysis. Prompts allow users to select values that dynamically filter all views within the analysis.

To add a prompt to an analysis:

  1. Click the Prompts tab.

  2. Add the prompt, as described in Chapter 6, "Prompting in Dashboards and Analyses."

Examining the Logical SQL Statements for Analyses

This procedure is an advanced step in the process for constructing an analysis. For information, see "What Is the Process for Constructing Analyses?"

You use the "Analysis editor: Advanced tab" to examine the XML code and logical SQL statement that is generated for an analysis and optionally create a new analysis based on that SQL statement. Generally you need not use the features of this tab, because the vast majority of functionality for working with analyses is provided through the user interface. The following list provides examples of situations in which you might want to examine SQL statements:

Before working with the Advanced tab, keep the following important points in mind:

Note:

You can see the logical SQL statement that an analysis is using by inserting a logical SQL view. You can also enter SQL statements using the "Administration: Issue SQL page".

To edit the XML code or examine the SQL statement that is generated for an analysis:

  1. Click the "Analysis editor: Advanced tab".

  2. Use the links that correspond to the analysis name in the Referencing the Results area at the top of the tab to:

  3. Use the fields in the Analysis XML area to view and modify the XML code, and click Apply XML.

    If you modify the XML code, then you affect the analysis as it is saved in the Oracle BI Presentation Catalog.

    To apply the settings for the Partial Update and the Bypass Oracle BI Presentation Services Cache boxes, you must also click the Apply SQL button at the bottom of the tab.

  4. Use the read-only box in the SQL Issued area to examine the SQL statement that is sent to the Oracle BI Server when the analysis is executed.

    If you want to create an analysis using the SQL statement from the current analysis as the starting point, then click the New Analysis button. Any hierarchical columns, selection steps, groups, or formatting that are specified in the current analysis are removed.

  5. Depending on the content of the analysis, use the available fields in the Advanced SQL Clauses area to change the subject area, add GROUP BY or HAVING clauses, and specify DISTINCT processing and a prefix.

  6. Click Apply SQL to apply your changes.

    Note:

    Use care when clicking the Apply SQL button. When you do, Oracle BI EE creates a new analysis based on the SQL statement that you have added or modified. Therefore, you lose all views, formatting, and so on that you had previously created for the analysis. The XML code is also modified for the new analysis.

Integrating an Analysis with Microsoft Excel's Internet Query Feature

Integrating an analysis with Excel's Internet Query (IQY) feature enables you to run the analysis from within Excel. This feature can be useful for printing and distributing analyses. You can integrate analyses into Excel's IQY feature only with results that are displayed in one or more table or pivot table views. Other view types are not supported.

When the IQY file is opened in Excel, you are prompted to enter your Oracle Business Intelligence user ID and password. (You must have a nonblank password.) Then, the results of the saved analysis are retrieved and placed in an Excel spreadsheet. You can save the Excel spreadsheet so that the data can be refreshed directly from within Excel.

Note:

You might not be able to download results into an Excel spreadsheet if your company is using certain security features. For more information, contact the administrator.

By default, Excel prompts you to enter your user ID and password each time that the query is refreshed. Alternatively, you can save your user ID and password within the spreadsheet. Consult your Excel documentation for details.

To integrate an analysis with Excel's IQY feature:

  1. Save an analysis with one or more table or pivot table views.

  2. In the "Analysis editor: Advanced tab", click the link to generate and download a Web Query file.

  3. Save the file to the desired location.

    After opening the file in Excel, you can modify it, specify additional formatting, enhance it with graphs, and so on. For information, see your Excel documentation.

Saving Analyses

This procedure is a step in the process for constructing an analysis. For more information, see "What Is the Process for Constructing Analyses?"

You can save an analysis to a:

To save an analysis:

  1. In the Analysis editor, click the Save Analysis toolbar button to display the dialog to save the analysis.

  2. If you want to save the analysis to a personal or shared folder:

    1. In the Save In box, select the personal or shared folder in which to save the analysis.

    2. In the Name field, enter a name for the analysis, such as forecasted_sales.

    3. Click OK.

  3. If you want to save the analysis to a Lotus Notes database using a Web Archive file:

    1. In the Save In box, select the folder in which to save the analysis.

    2. In the Name field, enter a name for the analysis, including an .mht file extension, such as forecasted_sales.mht.

    3. Click OK.

    4. Upload the .mht file to the Lotus Notes database.

  4. If you want to save the analysis to a Lotus Notes database using an agent:

    1. In the Save In box, select the folder in which to save the analysis.

    2. In the Name field, enter a name for the analysis, such as forecasted_sales.

    3. Click OK.

    4. Create the agent, scheduling it to be sent by email to a specific address on a Lotus Notes server.

      For more information on creating an agent, see "Creating Agents".

Creating Agents from Analyses

You can create an agent directly from an analysis. When you create an agent using this method, Oracle BI EE does the following:

For more information on agents, see Chapter 8, "Delivering Content."

To create an agent from an analysis:

  1. Locate the analysis from which you want to create an agent using one of the following methods:

  2. Complete the following tabs of the Agent editor:

  3. Save the agent.

Editing Analyses

Note:

If you are using Oracle BI Enterprise Edition in accessibility mode, then you use the "BI Composer wizard" rather than the "Analysis editor" to edit analyses. For more information on BI Composer, see Chapter 14, "Using BI Composer to Work with Analyses." For more information on accessibility mode, see Appendix C, "Accessibility Features."

To edit an analysis:

  1. In the global header, click Catalog to display the "Catalog page".

  2. Navigate to the analysis to edit and click the Edit link. The analysis is displayed in the "Analysis editor".

  3. Make the desired changes.

  4. Save the analysis.

Editing Upgraded Analyses that Contain "Advanced SQL"

In previous releases (prior to 11g), advanced users could edit the SQL statements for analyses using the Advanced tab. In this release, you can continue to use the "Analysis editor: Advanced tab" for this purpose, if the analysis does not contain features such as hierarchical columns, selections, or groups.

For more information, see "Examining the Logical SQL Statements for Analyses". For information on upgrade, see "Upgrading Analyses that Use Advanced SQL" in Oracle Fusion Middleware Upgrade Guide for Oracle Business Intelligence.

About Embedding an Analysis in a Dashboard

Embedding an analysis in a dashboard causes it to execute automatically and display the results within the dashboard. This provides access to current results. For example, if you are a sales executive whose company captures sales data on a daily basis, then you might want to have the dollar volume of product that was sold today be displayed on the front page of your dashboard.

You can embed previously created analyses from the Oracle BI Presentation Catalog by using the Dashboard editor. For information about the Dashboard editor, see "Adding Content to Dashboards".

Working with Direct Database Requests

Users with the appropriate privileges can create and issue a direct database request directly to a physical back-end database. The results of the request can be displayed and manipulated within the Analysis editor, and subsequently incorporated into dashboards and agents. This section includes the following topics:

Setting Privileges for Direct Requests

The following privilege settings in Oracle BI Presentation Services Administration control whether you can modify and execute direct requests:

  • Edit Direct Database Requests

    If this privilege is set for you, then you can create and edit direct database requests.

  • Execute Direct Database Requests

    If this permission is set for you, then you can issue direct requests and see the results.

For information, see "Managing Presentation Services Privileges" in Oracle Fusion Middleware Security Guide for Oracle Business Intelligence Enterprise Edition.

Executing a Direct Database Request

You can send a direct request to the database from Oracle BI EE.

Note:

Oracle BI Server security rules for data are bypassed and cannot be applied when direct database requests are issued from Oracle BI EE.

To execute a direct database request:

  1. In the global header, select New, then Analysis, then Create Direct Database Request.

    The "Analysis editor: Criteria tab" is displayed with options for creating a database request.

  2. Specify the appropriate options on the Criteria tab.

  3. To issue the request and see the results, if any, click the Results tab.

Seeing the Results of Direct Database Requests

After you retrieve columns to create an analysis from a direct database request, you can work with that analysis similarly to how you work with other analyses. The following list outlines differences with analyses whose columns originated from direct database requests:

  • The "Subject Areas pane" does not contain any columns, because you are not working with columns from a repository.

  • You cannot create groups or selection steps for this analysis.

  • You cannot specify conditional formatting for the columns.

Using Variables

You can reference variables in several areas of Oracle BI Enterprise Edition, including in analyses, dashboards, KPIs, actions, agents, and conditions. For example, suppose that you wanted to create an analysis whose title displays the current user's name. You can do this by referencing a variable.

There are four types of variables that you can use:

What Are Session Variables?

A session variable is a variable that is initialized at login time for each user. When a user begins a session, the Oracle BI Server creates a new instance of a session variable and initializes it.

There are as many instances of a session variable as there are active sessions on the Oracle BI Server. Each instance of a session variable could be initialized to a different value.

There are two types of session variables:

  • System — A session variable that the Oracle BI Server and Oracle BI Presentation Services use for specific purposes.

    System session variables have reserved names that cannot be used for other kinds of variables (such as static or dynamic repository variables and non-system session variables).

  • Non-system — A system variable that the administrator creates and names. For example, the administrator might create a SalesRegion non-system variable that initializes the name of a user's sales region.

The administrator creates non-system session variables using the Oracle BI Administration Tool.

For more information, see "About Session Variables" in Oracle Fusion Middleware Metadata Repository Builder's Guide for Oracle Business Intelligence Enterprise Edition.

What Are Repository Variables?

A repository variable is a variable that has a single value at any point in time.

There are two types of repository variables:

  • Static — Repository variables whose value persist and do not change until the administrator decides to change them.

  • Dynamic — Repository variables whose values are refreshed by data returned from queries.

The administrator creates repository variables using the Oracle BI Administration Tool.

For more information, see "About Repository Variables" in Oracle Fusion Middleware Metadata Repository Builder's Guide for Oracle Business Intelligence Enterprise Edition.

What Are Presentation Variables?

A presentation variable is a variable that you can create as part of the process of creating one of the following types of dashboard prompts:

  • Column prompt — A presentation variable created as part of a column prompt is associated with a column, and the values that it can take come from the column values.

    To create a presentation variable as part of a column prompt, in the "New Prompt dialog" (or Edit Prompt dialog), you have to select Presentation Variable in the Set a variable field and then enter a name for the variable in the Variable Name field.

    For information on working with column prompts, see "Creating a Column Prompt".

  • Variable prompt — A presentation variable created as part of a variable prompt is not associated with any column, and you define the values that it can take.

    To create a presentation variable as part of a variable prompt, in the "New Prompt dialog" (or Edit Prompt dialog), you have to select Presentation Variable in the Prompt for field and then enter a name for the variable in the Variable Name field.

    For information on working with variable prompts, see "Creating a Variable Prompt".

The value of a presentation variable is populated by the column or variable prompt with which it was created. That is, each time a user selects a value in the column or variable prompt, the value of the presentation variable is set to the value that the user selects.

What Are Request Variables?

A request variable lets you override the value of a session variable but only for the duration of a database request initiated from a column prompt. You can create a request variable as part of the process of creating a column prompt.

A request variable that is created as part of a column prompt is associated with a column, and the values that it can take come from the column values.

To create a request variable as part of a column prompt, in the "New Prompt dialog" (or Edit Prompt dialog), you have to select Request Variable in the Set a variable field and then enter the name of the session variable to override in the Variable Name field.

For information on working with column prompts, see "Creating a Column Prompt".

The value of a request variable is populated by the column prompt with which it was created. That is, each time a user selects a value in the column prompt, the value of the request variable is set to the value that the user selects. The value, however, is in effect only from the time the user presses the Go button for the prompt until the analysis results are returned to the dashboard.

Where Can I Reference Variables?

You can reference variables in the following areas (but not all types of variables can be referenced in each area):

For the syntax that you use to reference variables, see "What Is the Syntax for Referencing Variables?"

What Is the Syntax for Referencing Variables?

You can reference variables in analyses, dashboards, KPIs, and agents. How you reference a variable depends on the task that you are performing.

For tasks where you are presented with fields in a dialog, you must specify only the type and name of the variable (not the full syntax), for example, referencing a variable in a filter definition.

For other tasks, such as referencing a variable in a title view, you specify the variable syntax. The syntax you use depends on the type of variable as described in Table 2-1.

Note:

In the syntax, if the "at" sign (@) is not followed by a brace ({), then it is treated as an "at" sign.

Table 2-1 Syntax for Referencing Variables

Type of Variable Syntax Example

Session

@{biServer.variables['NQ_SESSION.variablename']}

where variablename is the name of the session variable, for example DISPLAYNAME.

For a list of system session variables that you can use, see "About System Session Variables" in Oracle Fusion Middleware Metadata Repository Builder's Guide for Oracle Business Intelligence Enterprise Edition.

@{biServer.variables['NQ_SESSION.USER']}

Repository

@{biServer.variables.variablename}

or

@{biServer.variables['variablename']}

where variablename is the name of the repository variable, for example, prime_begin.

@{biServer.variables.prime_begin}

or

@{biServer.variables['prime_begin']}

Presentation or request

@{variables.variablename}[format]{defaultvalue}

or

@{scope.variables['variablename']}

where:

  • variablename is the name of the presentation or request variable, for example, MyFavoriteRegion.

  • (optional) format is a format mask dependent on the data type of the variable, for example #,##0, MM/DD/YY hh:mm:ss. (Note that the format is not applied to the default value.)

  • (optional) defaultvalue is a constant or variable reference indicating a value to be used if the variable referenced by variablename is not populated.

  • scope identifies the qualifiers for the variable. You must specify the scope when a variable is used at multiple levels (analyses, dashboard pages, and dashboards) and you want to access a specific value. (If you do not specify the scope, then the order of precedence is analyses, dashboard pages, and dashboards.)

@{variables.MyFavoriteRegion}{EASTERN REGION}

or

@{dashboard.variables['MyFavoriteRegion']}


You also can reference variables in expressions. The guidelines for referencing variables in expressions are described in Table 2-2.

Table 2-2 Guidelines for Referencing Variables in Expressions

Type of Variable Guidelines Example

Session

  • Include the session variable as an argument of the VALUEOF function.

  • Enclose the variable name in double quotes.

  • Precede the session variable by NQ_SESSION and a period.

  • Enclose both the NQ_SESSION portion and the session variable name in parentheses.

"Market"."Region"=VALUEOF(NQ_SESSION."SalesRegion")

Repository

  • Include the repository variable as an argument of the VALUEOF function.

  • Enclose the variable name in double quotes.

  • Refer to a static repository variable by name.

  • Refer to a dynamic repository variable by its fully qualified name.

    If you are using a dynamic repository variable, then the names of the initialization block and the repository variable must be enclosed in double quotes ("), separated by a period, and contained within parentheses. For example, to use the value of a dynamic repository variable named REGION contained in a initialization block named Region Security, use this syntax:

    VALUEOF("Region Security"."REGION")

    For more information, see "About Repository Variables" in Oracle Fusion Middleware Metadata Repository Builder's Guide for Oracle Business Intelligence Enterprise Edition.

CASE WHEN "Hour" >= VALUEOF("prime_begin") AND "Hour" < VALUEOF("prime_end") THEN 'Prime Time' WHEN ... ELSE...END

Presentation

  • Use this syntax:

    @{variablename}{defaultvalue}
    

    where variablename is the name of the presentation variable and defaultvalue (optional) is a constant or variable reference indicating a value to be used if the variable referenced by variablename is not populated.

  • To type-cast (that is, convert) the variable to a string, enclose the entire syntax in single quotes, for example:

    '@{user.displayName}'
    

Note: If the @ sign is not followed by a {, then it is treated as an @ sign.

"Market"."Region"=@{MyFavoriteRegion}{EASTERN REGION}

For the specific areas where you can reference variables, see "Where Can I Reference Variables?".

What Predefined Presentation Variables Are Available?

Table 2-3 contains a list of the predefined presentation variables that you can reference in analyses, dashboards, KPIs, and agents. (Note that the syntax for these predefined presentation variables omits the variables. qualifier, for example, @{session.locale} rather than @{session.variables.locale}.)

Note:

For time zone variables, the time zone for a user must be set to a value other than Default in order for this variable to work. (Users set their preferred time zone in the "My Account dialog: Preferences tab".)

Table 2-3 Predefined Presentation Variables

Scope Presentation Variable Example

system

productVersion

system.productVersion = 11.1.1.3 (Build 090619.0110.000)

system

currentTime

system.currentTime = 6/29/2009 7:35:59 PM

session

locale

session.locale = en-us

session

language

session.language = en

session

rtl

This indicates whether the language selection in the Login page is a right to left language. For example, if the language selection is Hebrew, then this variable returns true.

session.rtl = false

session

timeZone

session.timeZone = (GMT-06:00) Central America

session

timeZone.id

This returns a value that is not localized.

session.timeZone.id = (GMT-06:00) Central America

session

timeZone.name

This returns a localized value.

session.timeZone.name = (GMT-06:00) Central America

session

timeZone.value

This returns a localized value.

session.timeZone.value = (GMT-06:00) Central America

session

loginTime

session.loginTime = 6/29/2009 7:12:01 PM

session

logoutTime

session.logoutTime = 6/29/2009 8:02:01 PM

session

lastAccessTime

session.lastAccessTime = 6/29/2009 7:35:59 PM

session

currentUser

session.currentUser = Administrator

session

currency.name

session.currency.name = $ English - United States

session

currency.symbol

session.currency.symbol = $

session

currency.userPreference

session.currency.userPreference = Global Currency 2

user

id

user.id = Administrator

user

displayName

user.displayName = Administrator

user

homeDirectory

user.homeDirectory = /users/administrator

dashboard

currentPage

dashboard.currentPage = page 1

dashboard

xml

dashboard.xml = the dashboard XML

dashboard

dashboard.currency.name

dashboard.currency.name = Euro

dashboard

dashboard.currency.symbol

dashboard.currency.symbol = $

dashboard

dashboard.currency.userPreference

dashboard.currency.userPreference = Global Currency 1

dashboard

dashboard.path

This returns the path in the catalog.

dashboard.path = /users/administrator/_portal/Sales

dashboard

dashboard.name

dashboard.name = MyDashboard

dashboard

dashboard.caption

This returns the localized name of the dashboard.

dashboard.caption = Sales

dashboard

dashboard.location

This returns the URL for the location.

dashboard.location = Dashboard&PortalPath=/users/administrator/_portal

dashboard

dashboard.description

dashboard.description = Sales by region and district

dashboard

dashboard.author

dashboard.author = Administrator

dashboard.currentPage

dashboard.currentPage.name

dashboard.currentPage.name = Sales page 1

dashboard.currentPage

dashboard.currentPage.path

dashboard.currentPage.path = /users/administrator/_portal/Sales/page 1

dashboard. current Page

dashboard.currentPage.currency.name

dashboard.currentpage.currency.
name = USD

dashboard.current Page

dashboard.currentPage.currency.symbol

dashboard.currentPage.currency.
symbol = USD

dashboard. current Page

dashboard.currentPage.currency.userPreference

dashboard.currentPage.currency.userPreference = Global Currency 2

analysis

report.currency.name

report.currency.name = $ English - United States

analysis

report.currency.symbol

report.currency.symbol = $

analysis

report.currency.userPreference

report.currency.userPreference =
Global Currency 2


Example of Referencing a Variable in a Title View

Suppose that you have created a dashboard prompt called Region, which is based on the Region column and which creates a presentation variable called MyFavoriteRegion as shown in Figure 2-5.

Figure 2-5 Region Prompt

This image is described in the surrounding text.

Suppose also that you have created an analysis that shows Dollars by Region and District. You have also added a filter on the Region column that is set to Is Prompted so that you can prompt the user for a region using this Region prompt.

You can reference the MyFavoriteRegion variable in the Title view so that the user's selection in the Region prompt is displayed in the title of the analysis. To do so, you enter the variable syntax @{variables.MyFavoriteRegion} in the Title field in the Title editor, as shown in Figure 2-6.

Figure 2-6 MyFavoriteRegion Variable Referenced in the Title View

This image is described in the surrounding text.

Figure 2-7 shows the results on a dashboard page when a user selects EASTERN REGION in the Region prompt. Notice that the variable in the title has been updated to the user's selection, EASTERN REGION.

Figure 2-7 Results of Referencing My Favorite Region in the Title View

This image is described in the surrounding text.