This chapter describes the views that are available in Oracle Business Intelligence Enterprise Edition to present the results of analyses in meaningful, intuitive ways. It explains how to add views for display in dashboards, how to modify values (that is, perform write back) in views, and how to set up master-detail linking of views. It also describes interactions, drilling, sorting, and printing in views.
This chapter includes the following sections:
Views use the presentation capabilities of Oracle BI Enterprise Edition to help you look at results of analyses in meaningful, intuitive ways. You can add a variety of views to the results, such as graphs and pivot tables that allow drilling down to more detailed information, explanatory text, a list of filters that were used to limit the results, and more.
A title view, which displays the name of the saved analysis.
A table or pivot table view, which displays the results of the analysis, depending on the types of columns that the analysis contains:
If the analysis contains only attribute columns, only measure columns, or a combination of both, then a table is the default view.
If the analysis contains at least one hierarchical column, then a pivot table is the default view.
You can customize or delete the existing views for an analysis, add other views, and combine and position views anywhere in the pane.
Preparing multiple views of results can help you identify trends and relationships in data. If you are customizing results for display on a dashboard, then you can preview how the combination and position of views looks when viewed on a dashboard.
You can then save the analysis with the collection of views.
Table 3-1 describes the types of views that are available.
Your privileges control whether you can access all views or specific views only.
Displays a title, a subtitle, a logo, a link to a custom online help page, and timestamps to the results.
Displays results in a visual representation of data organized by rows and columns. A table provides a summary view of data and enables users to see different views of data by dragging and dropping rows and columns.
Displays results in a pivot table, which provides a summary view of data in cross-tab format and enables users to see different views of data by dragging and dropping rows and columns. Pivot tables structure data similarly to standard tables that contain column groups, but can display multiple levels of both row and column headings. Unlike regular tables, each data cell in a pivot table contains a unique value. By organizing data in this way, a pivot table is more efficient than a row-based table. Pivot tables are ideal for displaying a large quantity of data, for browsing data hierarchically, and for trend analysis.
Displays numeric information visually, which makes it easier to understand large quantities of data. Graphs often reveal patterns and trends that text-based displays cannot. However, when precise values are needed, graphs should be supplemented with other data displays, such as tables.
A graph is displayed on a background, called the graph canvas.
For the types and subtypes of graphs that are available, see Table 3-2.
Displays results as a three-dimensional graph that represents target and actual values using volume, level, and color. Typically, funnel graphs are used to graphically represent data that changes over different periods or stages. For example, funnel graphs are often used to represent the volume of sales over a quarter.
Funnel graphs are well suited for showing actual compared to targets for data where the target is known to decrease (or increase) significantly per stage, such as a sales pipeline.
In funnel graphs, the thresholds indicate a percentage of the target value, and colors provide visual information for each stage. You can click one of the colored areas to drill down to more detailed information.
For the types of funnel graphs that are available, see Table 3-3.
Shows a single data value. Due to its compact size, a gauge is often more effective than a graph for displaying a single data value
Gauges identify problems in data. A gauge usually plots one data point with an indication of whether that point falls in an acceptable or unacceptable range. Thus, gauges are useful for showing performance against goals.
Depending on the data in the analysis, a gauge view might consist of multiple gauges in a gauge set. For example, if you create a gauge view to show the sales data for the last twelve months, the gauge view consists of twelve gauges, one for each month. If you create one to show the total sales in the US, then the gauge view consists of one gauge.
A gauge or gauge set is displayed on a background, called the gauge canvas.
For the types of gauges that are available, see Table 3-4.
Displays multidimensional data shown as a set of cells in a grid, where each cell represents a subset of data using a particular graph type. Data can be represented with graphs, microcharts, and numbers.
The trellis view has two subtypes: Simple Trellis and Advanced Trellis.
Simple trellis views are ideal for displaying multiple graphs that enable comparison of like to like. Advanced trellis views are ideal for displaying spark graphs that show a trend.
A simple trellis displays a single inner graph type, for example a grid of multiple Bar graphs. The inner graphs always use a common axis; that is to say, the graphs have a synchronized scale.
An advanced trellis displays a different inner graph type for each measure. For example, a mixture of Spark Line graphs and Spark Bar graphs, alongside numbers. In this example, the Spark Line graph might show Revenue over time, and the Spark Bar graph might show Units Sold. A measure column displaying numbers might be placed adjacent to the Spark Line graphs, showing the Revenue measure as a total value for a year.
In an advanced trellis, each measure column operates independently for drilling, axis scaling, and so on.
For definitions of microchart, and spark graph, see "What Terms are Used with Trellis Views?"
Displays results overlain on a map. Depending on the data, the results can be overlain on top of a map as formats such as images, color fill areas, bar and pie graphs, and variably sized markers.
Displays the filters in effect for an analysis. Filters, like selection steps, allow you to constrain an analysis to obtain results that answer a particular question. Filters are applied before the query is aggregated.
Displays the selection steps in effect for an analysis. Selection steps, like filters, allow you to obtain results that answer particular questions. Selection steps are applied after the query is aggregated.
Adds a column selector in the results. A column selector is a set of drop-down lists that contain pre-selected columns. Users can dynamically select columns and change the data that is displayed in the views of the analysis.
Adds a view selector in the results. A view selector is a drop-down list from which users can select a specific view of the results from among the saved views.
Adds a legend to the results, which enables you to document the meaning of special formatting used in results, such as the meaning of custom colors applied to gauges.
Displays the results as one or more paragraphs of text. You can type in a sentence with placeholders for each column in the results, and specify how rows should be separated.
Displays the results as a ticker or marquee, similar in style to the stock tickers that run across many financial and news sites on the Internet. You can control what information is presented and how it scrolls across the page.
Adds static text in the results. You can use HTML to add banners, tickers, ActiveX objects, Java applets, links, instructions, descriptions, graphics, and so on, in the results.
Displays the SQL statement that is generated for an analysis. This view is useful for trainers and administrators, and is usually not included in results for typical users.
You cannot modify this view, except to format its container or to delete it.
Displays a Create Segment link in the results. Users can click this link to create a segment in their Oracle's Siebel Marketing operational application, based on the results data.
This view is for users of the Oracle's Siebel Marketing Version 7.7 (or higher) operational application. For information, see Oracle Marketing Segmentation Guide.
Create Target List
Displays a Create Target List link in the results. Users can click this link to create a target list, based on the results data, in their Oracle's Siebel operational application. This target list can be used as the foundation for initiating and logging call activity information for this defined set of targets.
This view is for users of Oracle's Siebel Life Sciences operational application integrated with Oracle's Siebel Life Sciences Analytics applications
Table 3-2 shows the types of graphs that are available and describes their uses. It also shows the available styles for each type. All graph types except for scatter, radar, and microchart can be 2-dimensional (2D) or 3-dimensional (3D). Not all types of graphs are appropriate for all types of data.
Shows quantities associated with categories. Bar graphs show quantities as bar lengths and categories as bars or groups of bars.
Bar graphs are useful for comparing differences among like items; for example, competing product sales, same product sales over different time periods, or same product sales over different markets.
Can be used to compare measure columns by showing bars in a horizontal or vertical direction.
Shows quantities over time or by category.
Line graphs are useful for showing trends over time.
Can be used to plot multiple measure columns.
Shows the trend of the contribution of each value over time or by category.
It is a line graph for which the regions between lines are filled in. Regions stack, adding up to the total value for each time period or category.
Shows data sets as percentages of a whole.
Pie graphs are useful for comparing parts of a whole, such as sales by region or by district.
Plots two sets of data with different ranges, one set as bars, and one set as lines overlaid on the bars.
Line bar graphs are useful for showing trend relationships between data sets.
Time Series Line
Plots time series data. It scales the horizontal axis based on the time that has elapsed between data points.
Is a form of bar graph and line graph that displays criteria in descending order. In this graph type, the line shows a cumulative total of the percentages.
Pareto graphs are useful for identifying significant elements, such as best and worst or most and least.
Displays x-y values as discrete points, scattered within an x-y grid. It plots data points based on two independent variables. This enables you to plot large numbers of data points and observe the clustering of data points.
Scatter graphs are useful for observing relationships and trends in large data sets.
Is a variation of a scatter graph that displays data elements as circles (bubbles). It shows three variables in two dimensions. One value is represented by the location of the circle on the horizontal axis. Another value is represented by the location of the circle on the vertical axis. The third value is represented by the radius of the circle.
Bubble graphs are useful for plotting data with three variables, and for displaying financial data over a period of time.
Plots the same information as a bar graph, but instead displays data radiating from the center of the graph. Each data element has its own value axis.
Radar graphs are useful for examining overlap and distribution.
A tiny graph (of similar size to a piece of nearby text) that displays only in the context of the trellis view and that is ideal for showing trend information.
A microchart graph type is useful within an advanced trellis, where data is displayed as a mixture of spark graphs and numbers.
A microchart does not have axes or legends. Like larger graphs, a microchart's measure values are rendered as relatively sized bars (or lines, or area). Each measure name is displayed in its column header. Further details of the measure appear as tooltip text when you hover the mouse over a data cell.
Table 3-3 lists and describes the types of funnel graphs that are available. The style of each type can be either solid or gradient fill.
Uses a standard shape with equal stage widths.
Uses a standard shape with unequal stage widths
Uses a standard shape with equal stage widths. It is similar to standard funnel except that the target values of all the stages before the last is calculated based on the last stage target value and a constant called target factor.
The list of graph types that are available for use in trellis views varies by trellis view subtype: Simple Trellis or Advanced Trellis.
The following types of graphs can be used in simple trellises:
Bar (subtype Vertical)
Bar (subtype Horizontal)
The following visualization choices are available in advanced trellises:
Microchart, of the following subtypes:
For details on each graph type, see Table 3-2.
Table 3-4 lists and describes the types of gauges that are available. All gauges are 2-dimensional (2-D) except for bulb gauges which can be either 2-dimensional or 3-dimensional (3-D). Their sizes can be small, medium, large, or a custom size.
Shows data using a dial arc with one or more indicators that point to to where the data falls within predefined limits.
Shows data using a horizontal bar that changes color to indicate whether the data is within predefined limits. The inner rectangle of the horizontal bar shows the current level of data against the ranges marked on an outer rectangle.
Shows data using a vertical bar that changes color to indicate whether the data is within predefined limits. The inner rectangle of the vertical bar shows the current level of data against the ranges marked on an outer rectangle.
Shows data using a circle that changes color to indicate whether the data is within predefined limits.
Bulb gauges are useful when you must know what the status is, and not any indication of a specific value, or relative information about other threshold ranges.
For information on how the behavior of charts (known as graphs in this release) that were created in previous releases (prior to 11g) has changed, see "Upgrading Charts" in Oracle Fusion Middleware Upgrade Guide for Oracle Business Intelligence.
To add a view to the results of an analysis:
The view is displayed in a container on the "Compound Layout". (Note that defaults are used to generate this view, such as the default view type, default formatting, and so on.)
To format the container for the view, click the Format Container toolbar button on the view container to display a formatting dialog.
For information, see Chapter 7, "Formatting Analyses, Views, and Dashboard Pages.".
To edit the view, click the Edit View toolbar button on the view container.
For more information, see "Editing Views".
Save the view. For information, see "Saving Views".
If you select a graph type or gauge type that is incompatible with the results, then no results are shown.
Do not use a pie graph to visualize data that includes negative values. Either filter the analysis to ensure that all values are greater than 0 or use a different graph type.
As you design graphs, funnel graphs, and gauges, note that there are differences in the display of a graph or gauge legend, depending on the whether the graph or gauge is displayed on a desktop or on a mobile device:
On a desktop (where the graph or gauge format is Flash), the legend list is scrollable and the legend text does not wrap.
On a mobile device (where the graph or gauge format is PNG), the legend list is not scrollable and the legend text can wrap.
Each view type, except for a Logical SQL view, has its own editor in which you perform your edits. (You cannot edit a Logical SQL view. You can only format its container or delete it.)
Each view editor contains unique functionality for that view type but might also contain functionality that is the same across view types.
To edit a view:
Open the analysis that contains the view to edit.
Click the "Analysis editor: Results tab".
Click the Edit View button for the view (or, alternatively, select the view in the "Views pane" and click the Edit View toolbar button).
One of the following editors is displayed:
Using the editor for the view, make the appropriate edits.
For additional guidance, see:
Save the view. For information, see "Saving Views".
Two common views for analyzing and displaying data are the table and pivot table. Figure 3-1 shows a sample table and pivot table on a dashboard page.
These views share the same type of editor and much of the same functionality such as dragging and dropping, sorting, drilling, and conditional formatting. The following list provides some differences between the two types:
Default View — When you create an analysis and display the results, the default view depends on the data in the analysis:
Table — If the analysis contains only attribute columns, only measure columns, or a combination of both, then a table is the default view.
Pivot Table — If the analysis contains at least one hierarchical column, then a pivot table is the default view.
Drop Targets — Drop targets, which you use to modify column layout, differ slightly between the two views. Both views have the <view-type> Prompts, Sections, and Excluded drop targets. Figure 3-1 shows that when multiple columns are placed in the <view-type> Prompts target, or page edge, each column displays its values in an individual drop-down list.
Table — Tables have columns and measures in the same drop target, and they do not have rows.
Pivot Table — Pivot tables can have rows, columns, and measures as separate drop targets.
Drag and Drop Measure Columns — Measure columns behave slightly differently when you drag and drop them on tables and pivot tables.
Table — In tables, you can drag and drop measure columns in the middle of the table and they act as columns.
Pivot Table — In pivot tables, you can drag and drop measure columns in the middle and their labels can be in many locations.
Column Names and Headings — You can turn off the display of the column headings in both tables and pivot tables. For column names, however, the following differences apply:
Table — In tables, you always see the names of columns.
Pivot Table — In pivot tables, you can turn off the display of the column names.
Properties — The properties of the two views differ slightly, in areas such as paging controls and green bar formatting.
You can use the editors for the table and pivot table to customize the look and functionality of the view. The editors share much of the same functionality.
To set the properties of the table or pivot table, click the Table View Properties or the Pivot Table View Properties button on the toolbar. The "Table Properties dialog" or the "Pivot Table Properties dialog" is displayed, where you can:
Specify formatting preferences for tables and pivot tables.
Set up for users to modify values in columns in tables, which is referred to as "write back."
Use the buttons on the toolbar of the editor to perform functions such as the following:
To modify formatting, see "Applying Formatting to Tables and Pivot Tables".
To add a group or calculated item to the view, see "Working with Groups and Calculated Items".
To affect the contents of the view, use the following panes:
To specify the sort order for a column in the view, see "Sorting Data in Views".
To drill in the data in the view, see "Drilling in Views".
Right-click to perform basic view interactions. For information, see "Right-Click Menu for Tables, Pivot Tables, and Trellises".
Enable the write-back capability for a table, which enables a user to update or write to the back-end database, if your user name has the appropriate permissions. For example, a user can enter sales targets for the current quarter in a Sales dashboard. See "Modifying Values and Performing Write Back" for information about the write-back capability.
Green bar styling shows alternating rows or columns in a light green color. Cosmetic formatting affects the overall appearance of the view and enables you to change the default green bar color. Figure 3-2 shows a table on a dashboard page with green bar styling.
To add green bar styling and cosmetic formatting to a table or pivot table:
On the toolbar of the editor, click the View Properties button.
To add green bar styling, click the green bar styling box.
To change the way the styling is applied, select a style from the list.
To add cosmetic formatting, click the alternate formatting button.
The "Edit Format dialog" is displayed.
To change the default green bar color, choose a new background color for the cell format. Make any other appropriate changes, and click OK.
In the Properties dialog, make any other appropriate changes, and click OK.
For more information about cosmetic formatting, see "What is Cosmetic Formatting?"
You can use graphs of various types for analyzing and displaying data. Figure 3-3 shows a sample bar graph on a dashboard page.
To edit a graph view:
To set the properties of the graph, click the Edit Graph Properties button on the toolbar. The "Graph Properties dialog" is displayed, where you can set properties:
Related to the graph canvas, such as legend location
That control the appearance of the graph, such as the style
For axis limits and tick marks.
That control the display of titles and labels for the graph
Use the buttons on the toolbar of the editor to perform functions such as printing a view, exporting a view, and so on.
To affect the contents of the view, use the following panes:
To drill in the data in the view, see "Drilling in Views".
You can format the visual appearance of graphs based on two settings:
The position of the graph elements (such as lines or bars in a line-bar graph or slices in a pie graph).
Conditions applied to columns. See "Graph Formatting Based on Columns"
Positional formatting enables you to customize the appearance of a graph based on the position of graph elements; that is, the numeric sequence in which graph elements (for example, bars) are displayed in a group. A group is determined by the attribute columns that are displayed in the Group By drop target area. (For information on drop target areas, see "Understanding Drop Targets".
For example, suppose that you have a bar graph whose criteria include the Region, Dollars, and Units columns. Dollars are displayed first, then Units in the Measures drop target area. Region is displayed in the Group By drop target. In this graph, Dollars is in position 1 and Units is in position 2 for each regional grouping as shown in Figure 3-4.
You can format the visual appearance of a graph based on position in terms of its color, line width, and line symbols.
Conditional formatting enables you to customize the appearance of a graph based on conditions applied to columns. The formatting is applied to the column values that meet the condition.
You can specify a color in which to display graph data based upon a specific column value, or range of column values that meet the condition specified for the column.
Conditionally changing the color of a graph based on specific column values.
A user wants to create a bar graph to compare sales between two beverages, Lemonade and Cola. When creating a bar graph the user specifies two conditions, one where the bar representing Lemonade sales is yellow, and another where the bar representing Cola sales is blue.
Conditionally changing the color of a graph based on a range of column values.
A sales manager wants to create a bar graph to compare sales for all reps across two sales bands.When creating a bar graph the sales manager specifies two conditions, one where the bar is red for all sales reps with sales less than $250,000, and another where the bar is green for all sales reps with sales greater than $250,000.
The way that you specify conditional formatting for graphs is different from that used for other views. For information, see "Applying Conditional Formatting to Tables, Pivot Tables, and Trellises".
To format the appearance of a graph:
Click Edit Graph Properties on the toolbar of the graph editor.
The "Graph Properties dialog" is displayed.
Click the "Graph Properties dialog: Style tab".
Click the Style and Conditional Formatting button.
The "Style and Conditional Formatting dialog" is displayed.
Click "Style and Conditional Formatting dialog: Style Formatting tab" to format the appearance of a graph based on position of the graph elements.
To add a custom formatted position:
Select the tab for the graph element (for example, bar) to which you want to add a custom formatted position.
Click the Add new position button. A new position entry is displayed in the Custom Formatted Positions table.
Specify the formatting. For example, to select the color to be applied to the position, click the down arrow next to the Color box to access the Color Selector dialog. (Note that the formatting options depend on the element.)
Click the "Style and Conditional Formatting dialog: Conditional Formatting tab" to format the appearance of a graph based on a condition that is applied to columns.
To add a condition to a column:
Click Add Condition Format and select the column to which you want to apply a condition.
The "New Condition dialog" is displayed.
Select the operator and enter a column value, or a range of column values for this condition.
Click OK when you have finished.
The new condition is displayed in the Conditional Formatting tab.
To select the color to be applied to column values when the condition is met, click the down arrow next to the Color box to access the Color Selector dialog.
The following rules apply for building and using conditions in graphs:
Conditions can be created only from columns that are being used by the graph.
When format conditions conflict with each other, conflicting conditions are prioritized in the following order:
Conditional formatting on attributes
Conditional formatting on measures
Style formatting based on the positions of graph elements
When a user drills on a graph that has conditional formatting applied, the following rules apply:
A conditional format based on measures is not carried to the next level. (It does not make sense to carry the conditional format to a different level; for example if, in a geographic hierarchy, from Region to City.
A conditional format based on attributes is carried to the next graph if it has not been drilled on.
For example, if you had the conditional format "Lemonade = Blue" and only drill on years, then "Lemonade = Blue" stays in place.
Table 3-5 lists the graph exceptions that apply to conditional formatting based on columns.
A trellis view is a kind of graph view that displays a grid of multiple graphs, one in each data cell.
Figure 3-5 shows a sample trellis view.
A trellis view falls into one of two subtypes:
Simple Trellis. A simple trellis displays a core inner graph multiplied across row sets and column sets, rendering many small multiples that are ideal for comparing and contrasting. Figure 3-5 is a simple trellis.
Advanced Trellis. An advanced trellis displays a grid of small spark graphs that are ideal for monitoring trends and spotting patterns in a data set.
To edit a trellis view:
To set the properties of the trellis, click the Trellis Properties button on the toolbar. The "Trellis Properties dialog" is displayed, where you can set properties:
Related to the grid canvas, such as legend location (simple trellis views only)
Related to graph size for the visualizations included in the trellis
That control whether the trellis listens to master-detail events
That control the appearance of the trellis's grid and its visualizations, such as various style choices and the way that legends are displayed
That control the type of scale and the appearance of scale markers for each of the trellis's visualizations (simple trellis views only)
That control the display of titles and labels (simple trellis views only)
Use the buttons on the toolbar of the editor to perform functions such as printing a view, exporting a view, and so on.
To affect the contents of the view, use the following panes:
To specify the sort order for a column in the view, see "Sorting Data in Views".
Right-click to perform basic view interactions. For information, see "Right-Click Menu for Tables, Pivot Tables, and Trellises".
Table 3-6 defines the terms that are used with trellis views.
In the context of Oracle BI EE, a visualization is the choice of graph that appears within a data cell in a trellis view. There are many visualizations from which to choose when creating a trellis view, including bar graphs, scatter graphs, and spark graphs.
A nested graph, inside the grid of a trellis graph. Each inner graph has its own dimensionality as specified in the Visualization area of the Layout pane.
The outer edges are the parts of a trellis view that border the inner graphs. These include the column and row headers, the section headers, and so on.
A trellis view that displays inner visualizations that are all the same type, such as all scatter graphs. The inner visualizations all use a common axis, also known as a synchronized scale.
A trellis view that can display multiple visualization types within its grid, for example, Spark Line graphs, Spark Bar graphs, and numbers. Each visualization type displays a different measure.
You can think of an advanced trellis as a pivot table, except that for each measure you add to the pivot table, you can optionally associate a dimension and render that dimension as a spark graph visualization.
(Applicable to Simple Trellis subtype only) A synchronized scale means that all the visualizations within the trellis are viewed on the same scale, that is, they share a common axis. Having a common axis makes all graph markers easy to compare across rows and columns.
A tiny graph displayed in a grid along with other tiny graphs and numbers, comprising the data cell contents of an advanced trellis view. In Oracle BI EE, a microchart is always a spark graph.
An embedded mini-graph that, in conjunction with other mini-graphs and numbers, illustrates a single trend. Spark graphs are also known as sparks.
Sparks do not include axes or labels; they get their context from the content that surrounds them. Each type of spark graph has only one measure, which is hidden; the scale is relative to itself only.
A spark graph can be of the graph subtype Spark Line, Spark Bar, or Spark Area.
The trellis view, also referred to as a trellis graph, is the same as a pivot table—with one major exception: the data cells within the trellis contain graphs. Whereas a stand-alone graph type such as a single bar graph or a single scatter graph works on its own, the trellis graph works only by displaying a grid of nested graphs, known as inner graphs. So a bar-graph trellis view is actually comprised of multiple bar graphs.
In a general sense within the field of business analytics, a visualization is a visual representation of data, displayed in graphs, gauges, maps, and so on. In the trellis editor, within the Layout pane, there is a drop target called Visualizations; this is where you drag the columns for the inner graphs that will comprise the trellis you are building.
The visualizations that make up a simple trellis view are all graphs, and they can be many types of the existing stand-alone graphs described in "What Types of Graphs Are Available?" The visualizations that make up an advanced trellis are always microcharts, of the graph subtypes Spark Line, Spark Bar, or Spark Area. Ideally, the microcharts within an advanced trellis are placed alongside numbers (representing the same measure). For more information, see "What Are Microcharts?"
For the most part, a trellis view behaves like a pivot table, and the main difference between a trellis and a pivot table is the way the data cells appear.
In the row and column label cells of a trellis, you can:
Right-click to do things like create groups and calculated items
Right-click to hide or move measure labels
Right-click to sort data
Drag to reposition rows and columns
In the data cells of a trellis, you can hover the mouse pointer to show related contextual information.
Numeric data cells in a trellis behave the same as numeric data cells in a pivot table.
The ways in which the behavior of a trellis view differs from the behavior of a pivot table are the following:
Graph data cells: There is no right-click functionality for the data cells in simple trellises, nor drilling in trellis graph data cells (left-click functionality).
Microchart data cells: When you hover the cursor over the data cells in spark graphs, you are shown contextual information (such as first, last, minimum, and maximum values) that otherwise is not displayed as it would be in a pivot table view.
Figure 3-6 shows an advanced trellis twice, the second time with a tooltip displayed as the cursor hovers over a microchart data cell. Microcharts do not show axis labels, as regular charts do, so it is not immediately obvious what the inner graph dimensionality is. Use a microchart's tooltips to understand the data being represented inside the graph—sample individual values, as well as the overall dimensionality.
A microchart is, as its name implies, a tiny chart. A microchart is displayed in trellis views of the Advanced Trellis subtype. A microchart is unique as a graph type in that it can never be an independent, stand-alone graph. It can be used only within the context of a trellis view, and its meaning comes only from the fact that it is one of many small multiples.
A microchart can be one of several graph subtypes, including Spark Line, Spark Bar, and Spark Area.
Figure 3-7 shows a sample trellis composed of the Spark Line subtype, and numbers.
Spark graphs, also called sparks, are unique from line graphs, bar graphs, and the other stand-alone graphs available in Oracle BI EE. Spark graphs are embedded mini-graphs that illustrate a single trend. Simple in their appearance, they do not include axes or labels and they get their context from the content that surrounds them. Each type of spark graph has only one measure, which is hidden; the scale is relative to itself only.
In Oracle BI EE, the visual appearance of sparks emphasizes trends, and within those trends, the highest and lowest values. Due to the condensed manner in which sparks display trends, it is possible for many trends to be compared (along with numeric values) on a single page.
Notice how the advanced trellis in Figure 3-7 visualizes the same measure twice (Revenue), setting the first instance to "Number," and associating a time dimension with the second instance (months, in Revenue Trend), while displaying the second instance as a spark. This advanced trellis shows the total value for the year, and next to it, how the total value trended over the year.
While sparks are useful for certain types of analysis, such as high-level observation of trends and spotting of patterns, it is important to note that they do not illustrate the same specificity as their larger, fuller-featured counterparts.
Sparks are unique mainly because they are small and because they can display a lot of information in a very constricted space. In Oracle BI EE, a spark graph is also different from a stand-alone graph in that, as a microchart, it can only be used within trellis views.
You can modify the size of your sparks in the same way that you do with full-featured graphs, in the Graph Properties dialog.
Figure 3-8 shows a sample of a spark graph with markers for highest and lowest values in revenue data over the course of a year, for three years.
For more information about creating trellis views that include sparks, see "Design Considerations for Trellis Views and Microcharts".
When you create a trellis view, the first thing you do is choose between two subtypes: Simple Trellis and Advanced Trellis. The Simple Trellis subtype displays a single type of inner visualization, for example, all bar graphs. The inner visualization always uses a common axis, so that all inner graphs are viewed on the same scale. (This concept of a common axis is also referred to as a synchronized scale.) Having a common axis makes all graph markers easy to compare across rows and columns.
The Advanced Trellis subtype allows for display of multiple visualization types within its grid. An advanced trellis illustrating sales trends might show a grid containing numbers in the cells of one column (revenue, for example), while another column alongside the numbers column displays Spark Line graphs in its cells, and those Spark Line graphs visualize the same measure as represented by the numbers (again revenue, but over a time period). Next to that column, a different microchart might be displayed, such as a column of Spark Bar graphs that visualize a different measure, such as unit totals.
Each measure that is visualized is assigned a different inner graph type. Each cell of the grid is scaled independently.
Think of an advanced trellis as a pivot table with spark graphs inside its data cells. But, for each measure you add, you can optionally associate a dimension and render it as a microchart visualization. This makes an advanced trellis very different from a simple trellis. In a simple trellis, all of the measures are rendered in the same visualization, along with additional dimensions.
The following are some ideas to consider when designing content displayed in trellis views:
For comparisons, choose the Simple Trellis subtype.
For trend analysis, choose the Advanced Trellis subtype.
The inner graphs that make up a trellis should be readable and not too dense, so a trellis view is not especially useful for displaying multiple series or multiple groups. If you cannot easily target a data point with your mouse (to be shown a tooltip), then it is likely that the inner graph is too dense to be readable.
When using the Simple Trellis subtype, note the following:
Designing a simple trellis is like designing a pivot table, except that the total number of cells that can be rendered is much less for a trellis.
The main difference between designing a simple trellis and designing a pivot table is that for a trellis, one or two of the dimensions can be associated with the visualization; so, that many less dimensions must be added to the outer edge.
It is best to design the trellis with a small number of outer-edge dimensions. The entire graph series should be visible at once (for easy comparison of like to like) with no need to scroll. If you must show additional dimensionality, consider adding the dimensions to the graph prompt.
When determining which data to show in column headers and which to show in row headers, the column headers should show one or two dimensions (each dimension with a small number of members). Most often, the dimension shown in column headers is time. Place the remaining dimensions in the row headers or in graph prompts.
When using the Advanced Trellis subtype, note the following:
The key use case for an advanced trellis is to show trend graphs alongside numeric values, in a compressed form. So a typical advanced trellis contains a combination of spark graphs alongside number representations of the same measure.
Ideally, place no dimensions in the column headers, just place the measures here.
The dimensionality typically associated with a spark graph is time. As there are no visible labels in a spark graph, it is important that the data visualized is intrinsically ordered. For example, a spark graph visualizing regions would be meaningless, because the ordering of the regions (which would be the specific bars, in a Spark Bar graph) is unintuitive.
Just as when designing pivot tables, you generally display time on the horizontal axis, with the other dimensions displayed on the vertical axis. The eye then scans from left to right to see how the dimensionality changes over time.
Hierarchical columns do not work well with the Simple Trellis subtype, because when a hierarchical column is displayed on the outer edge, parents and children (such as Year and Quarter) will by default be shown using a common axis scale. However, because Year and Quarter have different magnitudes, the markers in child graphs may be extremely small and hard to read against the parent scale. (Hierarchical columns do work well with the Advanced Trellis subtype, however, because each data cell is a different scale.)
You can use gauge views of various types for analyzing and displaying data. Figure 3-9 shows a sample gauge view on a dashboard page.
To edit a gauge view:
To set the properties of the gauge view, click the Edit Gauge Properties on the toolbar. The "Gauge Properties dialog" is displayed, where you can set properties:
Related to the gauge canvas, such as, legend location.
That control the appearance of the gauge, such as the width and height of the gauge.
For gauge limits and tick marks.
That control the display of titles, footers, and labels for the gauge or gauges in a gauge set.
Use the buttons on the toolbar of the editor to perform functions such as printing a view, exporting a view, and so on.
To affect the contents of the view, use the following panes:
To drill in the data in the view, see "Drilling in Views".
You can create a map view for presenting business data to users. Figure 3-10 shows a sample map view on a dashboard page.
As with views such as tables, pivot tables, and graphs, map views allow users to display data on maps in several different formats and to interact with the data. When data is visualized on a map, relationships among data values that might not have been obvious previously can be displayed in a much more intuitive manner. For example, a map view can show a map of a city with the zip codes color-coded by sales performance, while an image marker displays the average discount given per order.
Changes to data are reflected in the map views, and users can perform the following tasks:
Display data with a spatial attribute on a map:
Display color-coded regions as thematic maps.
Show point-based information, such as sales locations within territories.
Zoom, pan, and drill to additional map detail.
Use a map view to interact with other dashboard elements:
Dashboard prompts drive map content.
Interaction with the map view updates other dashboard content, including driving master-detail interactions with other views.
Show additional related contextual information by hovering the mouse pointer over map areas.
Complete additional tasks using actions.
Before you as the content designer can create map views for analyses, the administrator must set configuration options and manage metadata, as described in "Configuring Mapping and Spatial Information" in Oracle Fusion Middleware System Administrator's Guide for Oracle Business Intelligence Enterprise Edition.
Map views offer the following functions:
You can display multiple attribute columns and measure columns on a map view. Map views do not support hierarchical columns.
You can format a map view, including with colors, bar graphs, pie graphs, variably sized bubbles, images, lines, or colored shapes that provide the ability for binning and other formatting options.
You can drill in information windows. When you hover over a value, you see a popup with information such as the geographic region (city, state, or country), the measures in the form of column names and column values and formatting metrics, and non-geographic data such as year. You can click the value to display the actionable links for drilling in the information window, then click the appropriate link.
Table 3-7 defines the terms that are used with map views.
An entity with spatial and non-spatial attributes such as cities, rivers, and highways. A feature can be represented by a polygon, a line, or a point.
Defines rendering properties for a feature. For example, if the feature is a polygon that shows a county, then the format can define the fill color for the county or can define a pie graph to be drawn over the county. Formats are tied to a particular geographic level such as continent, country, region, state, or city.
Any collection of features and formats that have a common set of attributes and a location. For example, a layer that shows US states can include color coding of states by sales, and a pie graph that shows sales per brand for that state. In addition to the US states layer, you can use a layer that displays stores within a state as individual points, with popup notes that show sales for each store.
Layers are displayed on a background or template map. When you zoom in and out on the map, various layers are hidden or displayed. Some layers must be enabled for data, so you can display it in the map. Other layers, such as one that shows roads, is not related to data.
A map that consists of a background or template map and a stack of layers that are rendered on top of each other in a window. A map has an associated coordinate system that all layers in the map must share. The map can be an image file, the object representation of an image file, or a URL that refers to an image file.
A map view contains the following components:
Main content — The main content is the background or template map, which provides the background geographic data and zoom levels. The main content can be any image that can be configured using MapViewer, for example, the floor maps of office buildings and the appearance and presence of items such as countries, cities, and roads.
Layers — One or more interactive or custom layers overlay the main content. For information, see "What Are Layers?"
Toolbar — Provides buttons that allow you to manipulate map contents directly. The map view itself has a toolbar. The content designer specifies whether to display the toolbar for the map view on a dashboard page. On a dashboard page, the toolbar is displayed directly over the map and contains only the Pan, Zoom Out, and Zoom In buttons. The toolbar is visible by default and its display is controlled by the "Map Properties dialog: Tools tab".
Each button represents a mutually exclusive mode for manipulating the map contents. The default mode is Pan. When you click a button to activate a mode, that button becomes highlighted and the mouse pointer changes to match the image on the button.
The toolbar in the Map editor contains additional options for modifying the map view.
Zoom controls — Adjusts the detail of the geographic data that is shown in the map view. For example, zooming in from a country might show state and city details.
On the Administration page, the administrator specifies which zoom levels each layer is visible for. You can have multiple zoom levels for one layer, and you can have a single zoom level associated with multiple layers. When you zoom, you change the map information at that zoom level, but you do not affect the display of Oracle BI data at that level. You affect the display of Oracle BI data by drilling.
The zoom controls consists of pan buttons and a zoom slider that render in the upper left-hand corner of the map view. When the zoom control is zoomed-out all the way, the zoom level is set to 0 and the entire map view is displayed.
The "Map Properties dialog: Tools tab" lets you determine the visibility of the zoom control. By default, the map is initially zoomed into the highest zoom level that fits the entire contents of the top-most layer. For example, if the highest ordered layer contains data only in the state of California, then the map zooms to the highest zoom level that still shows all of California.
Pan buttons — Consists of buttons (with arrows that point north, south, east, west, northeast, northwest, southeast, and southwest) at the top of the zoom control. You can use these buttons to move the entire map in specific directions. When you click the pan center button, the map returns to the zoom level and the center that the map was displayed with initially.
Zoom slider — Consists of a slider with a thumb for large scale zooming and buttons for zooming a single level. You can use the plus button to zoom in and the minus button to zoom out one level at a time. When the thumb is at the bottom of the slider, the zoom level is zero.
Scale tool — Also known as Distance Indicator. Provides a key to distance interpretation of the map and consists of two horizontal bars that display in the lower left-hand corner of the map view below the information panel and above the copyright. The top bar represents miles (mi) and the bottom bar represents kilometers (km). Labels are displayed above the miles bar and below the kilometers bar in the format: [distance] [unit of measure]. The length and distance values of the bars change as the zoom level changes and as the map is panned.
Legend — The legend is a semi-transparent area in the upper right-hand corner of the map view that you can display and hide. The legend shows the information that relates to the current zoom level. The legend provides a read-only visual key for symbols, layers, and formatting on the map and displays the following:
All visible formats that are applied to the map. If a format is turned off, then the corresponding legend item is hidden also. If a format is turned on but zoomed out of view, then it is not displayed in the legend. The legend displays text such as "No formats defined for current zoom level" if you have no formats defined at the zoom level that is currently displayed.
Scrollbars, if needed.
Use the Expand Map Legend and Collapse Map Legend buttons in the upper right-hand corner to control the display of the legend.
Overview map — Consists of a miniature view of the main map that is shown in the lower right-hand corner of the main map. This overview map provides regional context and lets you change the viewable area of the main map without having to use the pan buttons.
The following items are part of the overview map:
Reticule: Displays as a small window that you can move across a miniature view of the main map. The position of the reticule in the miniature map determines the viewable area of the main map. As you move the reticule, the main map is updated automatically. You can also pan in the overview map without using the reticule.
The overview map is automatically hidden if the reticule cannot be shown. This hiding generally happens when the scale difference between successive zoom levels is too small to show the miniature view in the overview map.
Show/Hide button: Displays in the upper left-hand corner when the overview map is displayed. When you click the Show/Hide button, the overview map becomes invisible and only the button can be seen in the lower right-hand corner of the main map.
Interactive panel — Accompanies the view in a dashboard and in the "Compound Layout". The top section of the interactive panel enables you to create and edit BI data formats in the Analysis editor. If a format has editable thresholds, then a slider is displayed in the Map editor that enables you to edit thresholds by dragging the slider. The interactive panel enables you to rearrange formats within a geographic layer. For example, if the States layer has three formats, then you can select the order in which the formats are displayed.
Dashboard users can control the visibility of formats (by turning them on or off) and can adjust format thresholds if the content designer has allowed them to.
The lower section of the panel includes the Feature Layer area, where you can select non-BI layers to add to the map. A non-BI layer is one that has not been associated with a BI column. Typical examples include highways, parks, or automated teller machines. You cannot apply formats to non-BI layers.
Layers are collections of geometries that share common attributes. Layers can be either predefined or custom:
A predefined layer is one whose geometry is defined in a spatial table in an Oracle Database. The administrator makes predefined layers available using the Administration pages, as described in "Administering Maps" in Oracle Fusion Middleware System Administrator's Guide for Oracle Business Intelligence Enterprise Edition.
A custom point layer is one that you define while editing a map view, as described in this section.
Layers can be of different types. A polygon layer represents regions, such as states. An example is a New England layer for the United States that consists of Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island, and Vermont.
A point layer represents specific points on a map, based on a coordinate system. For example, a point layer might identify the locations of warehouses on a map. The layer can use a different image for the type of inventory (electronics, housewares, garden supplies) in a set of warehouses to differentiate them from each other.
You can create a custom point layer that applies points to a map using longitude and latitude coordinates. For example, suppose that your company is headquartered in New York City. You can create a custom point layer that displays your company icon over New York City, and that icon is always displayed whenever a map that includes New York City is shown. The custom point layer is always on top of the other layers and is not affected by zoom levels. The custom point layer is used only by the current map view in the current analysis; it is not used by other map views, even for the same analysis.
You can select layers to be visible or hidden on a map, but you cannot modify predefined layers. You also create formats to apply to the layers, such as colored regions, bubbles, points, lines, or bar or pie graphs. Not all formats are available for all layer types. For example, point layers cannot have color fill formats.
When the administrator has configured maps, you as the content designer can create analyses whose data is displayed in map views.
You create a map view after selecting columns to display in that view. If the administrator has specified multiple background maps, then the map view is initially displayed using the first background map that was specified by the administrator that includes at least one layer that has been associated with a column that you selected. When editing the map view, you can select a different background map. You can apply layers to the background map and formats to the layers.
To create a map view for an analysis:
Create an analysis by selecting the attribute and measure columns to display on a map view. You generally select attribute columns that are related to geography, such as Cities, States, or Regions.
Select only those geography columns that the administrator has mapped to layers using the Administration pages, as described in "Administering Maps" in Oracle Fusion Middleware System Administrator's Guide for Oracle Business Intelligence Enterprise Edition.
Select the Results tab.
On the toolbar, click the New View button, then select Map.
If you see a message that states that a background map is required to render a map, then no metadata exists to populate the map. Contact the administrator about configuring the metadata for maps using the Administration page.
Otherwise you see a map that uses the background map that was specified on the Administration page.
Click the Edit View button to display the view in the "Results tab: Map editor".
You see a default map view in which the measures have been divided into reasonable bins and color coding has been applied. You can adjust the map view and layers as you like. If you do not see any default format created, then you must create a format.
Specify map view properties as appropriate using the "Map Properties dialog".
Apply formats to the layers, as described in "Applying Formats to Layers".
A map view uses columns of Oracle BI EE data. Each column has a set of properties that define its characteristics, such as for formatting and interaction. Any formatting that has been applied to a column is not applied to the map, except for the settings for interaction. Any formatting that originates from thresholds that the map contains is applied.
You can apply various kinds of formats to map views. The following list describes the options for defining formats to apply to BI layers. You cannot apply formats to non-BI layers.
Color Fill — Displays the "Color Fill (Layer) dialog", which you use to render areas in fill colors that indicate that an area meets a particular condition.
Color fill formats apply to regions or polygons. For example, a color fill format might identify a range of colors to represent the population in the states of a region or the popularity of a product in the states of a region. A map view can have multiple color formats visible at different zoom levels. For example, a color fill format for the layer at zoom levels 1-5 might represent the population of a state, and the county median income for the layer at zoom levels 6-10. You can also specify different colors to identify a range of data values.
Bar Graph — Displays the "Bar Graph (Layer) dialog", which you use to render a series of data as a bar graph within an area. Graph formats can show statistics related to a given region such as states or counties. For example, a graph format can display the sales values for several products in a state.
Even though you can create multiple graph formats for a particular layer, such creation is not recommend as the formats might overlap on the layer and the displayed results might be undesirable.
Pie Graph — Displays the "Pie Graph (Layer) dialog", which you use to render a series of data as a pie graph within an area.
Shape — Displays the "Variable Shape (Layer) dialog", which you use to render a measure column that is associated with an area by drawing markers or shapes within the region. You can also specify different colors for the shape to identify a range of data values.
Bubble — Displays the "Bubble (Layer) dialog", which you use to render a bubble within an area, similar to the shape format.
Image — Displays the "Image (Layer) dialog", which you use to render an image within an area, similar to the shape format. You can specify different images to identify a range of data values. You select images that have been specified by the administrator.
Line — Displays the "Line (Layer) dialog", which you use to render a line on a map.
You can include lines on maps to display paths such as highways, railway lines, and shipping routes. You can specify the width of lines and you can use the Map Wrap-Around feature on the "Map Properties dialog: Canvas tab" to allow lines to be unbroken, such as when showing an airline flight path from San Francisco to Tokyo.
Custom Point — Displays the Format Custom Point (Layer) dialog, which you use to render a point format, such as a bubble, image, or shape in a layer. Custom points are displayed at all zoom levels and on top of all other map formatting. When you create a Custom Point format, you select columns to specify the latitude and longitude.
The visibility of a format depends on the following:
The zoom level on the map and the "zoom range" of the format. For example, a Color Fill format for States is visible when state boundaries are visible and it is turned on, but is no longer visible when the map is zoomed out to the Continent level.
The data point limit. Formats are generally visible when they are zoomed into view and are turned on, but they might not be displayed if a particular layer has exceeded its maximum number of data points.
Custom point formats are unique in that they are displayed on the map always, for all zoom levels.
Format data is displayed in the legend only when the format is both turned on and zoomed into view. A format is turned on when the box beside its name is selected in the Map Formats area.
The map cannot display multiple non-point formats at a time (at a single zoom level) but can display multiple point formats simultaneously, if they do not share the same latitude and longitude location. If multiple graph formats are specified at the same geographic layer, then they render on top of each other.
The following formats apply to geographic areas such as polygons: Color Fill, Bubble, Pie Graph, and Bar Graph.
The following formats are based on a single latitude and longitude location (a point): Bubble, Variable Shape, Image, and Custom Point.
The line format is displayed only when a line geometry is present. Line formats are the only format that you can create for line geometries.
When you define formats, you can specify that different formats apply to different measure columns in a layer.
To apply a format to each layer as appropriate, use one of the following ways:
Click the Create a new Map Format link, if no layers are specified in the Map Formats list.
Click the Add New Map Formats button, either in the Map Formats title bar or beside a layer name.
For more information, see "Showing or Hiding Formats on a Map View".
You can use options on the "Map Properties dialog: Interaction tab" to specify the view port (the initial center of the map and zoom level) when a map view is first loaded or refreshed in a browser. This specification allows you to affect the initial display of the map for users.
Many of the same functions are available for map views, whether you are editing them or displaying them on a dashboard page. Some functions are available only when editing map views, as described in the following list:
When you edit a map view and work with layers, you create formats for the BI layers. You can display a dialog for adding and editing formats. You can also specify the order of formats for each layer. The order affects the visibility of the formats for the layer. Formats that are displayed higher in the list overlay formats that are lower in the list. When you display a map view, you can turn on and off the display of formats, but you cannot add or remove them from the map view.
As the content designer editing a map view, you can control the display of layers, but only of those layers that the administrator has associated with this background map. The administrator must ensure that BI layers are ordered higher than non-BI layers. If a non-BI layer is ordered higher than any BI layers, then the non-BI layer is displayed on top of the lower BI layers on the map, which prevents the BI layers from being interactive.
When you edit a map view, you decide whether users of dashboard pages can modify the thresholds of Color Fill formats using sliders. When you display a map view, you can modify the thresholds of Color Fill formats (if that ability was provided). If you close and re-open the map view, then you see the thresholds at the same state in which you left them. To save the thresholds when you leave the dashboard, first save the customizations of the dashboard.
When you edit or display a map view, legend contents display all active and visible formats.
For more information on the functions that are available when you display a map, see "Working with Map Views on Dashboard Pages".
You use a narrative view to provide information such as context, explanatory text, or extended descriptions along with column values. Figure 3-11 shows a sample of a narrative view on the left, alongside a table with green bar styling.
In a narrative view, you can include values from attribute columns, hierarchical columns, and measure columns. For a hierarchical column, you can use selection steps to display hierarchy levels with the hierarchical column. For example, create a step to select members based on hierarchy and add members of the specified level. You cannot drill in narrative views.
To edit a narrative view:
In the Narrative box, specify the columns to include in the view. To do so, use an at sign (@), optionally followed by a number. For example, include @2 to indicate the second column per the order of the column criteria.
Complete other fields as needed.
A column selector is a set of drop-down lists that contain pre-selected columns. Users can dynamically select columns and change the data that is displayed in the views of the analysis. Figure 3-12 shows a sample of a column selector view.
One drop-down list can be attached to each column in the analysis, and multiple columns can be attached to each drop-down list. You can create drop-down lists for attribute columns and measure columns. Updates that you make in the column selector view affect all the data views in the analysis.
You add columns to drop-down lists from the Subject Areas pane. When you add columns in this way, they are not added to the Criteria tab for the analysis. Instead, when you display the Criteria tab, you see that the column is now referred to as a "Column Group" with the default column for the list specified also. The default column is the one on which you created the drop-down list.
To edit a column selector view:
Select Include Selector for each column for which you want to include a drop-down list of columns. Ensure that the column is highlighted in the editor.
To add a new column to a drop-down list, double-click the column to add in the "Subject Areas pane".
Complete the other fields as appropriate.
A view selector view enables users to select a specific view of the results from among the saved views for an analysis. When placed on a dashboard, the view selector is displayed as a list from which users can choose the view that they want to display below the selector. Figure 3-13 shows a sample of a view selector view in which a narrative view was the most recently selected view.
Generally, you would include views in the view selector that are not being displayed in the Compound Layout view. For example, you might create a table, graph, gauge, and view selector view for an analysis, but include only the table and view selector view on the Compound Layout view. When the analysis is displayed on a dashboard page, users can select the graph or gauge view from the view selector view.
To edit a view selector view:
You use a legend view to document the meaning of special formatting used in an analysis, such as the meaning of custom colors that are applied to gauges. Figure 3-14 shows a sample of a legend view in which the captions describe the meanings of the colored data.
To edit a legend view:
In the Captions box on the toolbar, select the position for the caption relative to the legend symbol.
In the Legend items per row box on the toolbar, select the number of legend items to display in a row.
For each legend item:
In the Caption box, enter the meaning of the legend symbol.
In the Sample Text box, enter the text that is to be displayed within the legend symbol.
To add additional legend items, click Add Caption.
Complete other fields as needed.
Each editor for a data view contains the "Layout pane", except for the map editor. The Layout pane is displayed slightly differently for each view type, such as graphs and pivot tables. The Layout pane shows how the data in a view is laid out using drop targets. For more information on drop targets, see "Understanding Drop Targets".
You use the Layout pane to modify the way that data is arranged in the view. Specifically, you can:
Modify the data in the view, as described in "Adding and Rearranging Columns in Views"
Set properties, as described in "Setting Properties for View Bodies and Drop Targets"
Add totals, as described in "Adding Totals to Tables and Pivot Tables"
Display running sums and relative values, as described in "Displaying Running Sums and Relative Values for Measure Columns in Pivot Tables"
Define section sliders in graphs and gauges, as described in "Defining Section Sliders in Graphs and Gauges"
Add legends to graphs by selecting the Show In Legend box in the Layout pane
In the Layout pane, the columns in a data view are displayed in drop targets. Drop targets indicate where a column can be inserted, moved to, or dropped. They represent a valid position for a column.
You use drop targets to modify the way data is arranged in a data view by dragging and dropping columns to different targets within the view.
The Layout pane for trellis views is significantly different from the Layout panes that appear for other data views. For exceptions to the information provided here about drop targets, see "Drop Target Guidelines for Trellises".
Each data view contains the following drop targets:
<view-type> Prompts — Provides an interactive result set that enables users to select the data that they want to view. The values from the columns that are displayed in this drop target are used as the initial criteria. In a view, these values are displayed in a drop-down list for selection, which is often referred to as the "page edge."
Sections — Populates the areas that divide the view into sections. If you select the Display as Slider option in this drop target, then the values of the columns that are dropped in the Sections drop target are displayed as a section slider rather than as unique views.
<view-type> area — Simulates the plot area or the body of the view itself and assists you in seeing what the view looks like.
For tables, this area contains the Columns and Measures drop target, which contains all the columns in the view. In tables, all measures are treated as columns.
For all other data views, this area contains the following drop targets:
Measures drop target — Populates the part of a view that contains summary data. Depending on the type of view, this area might include a single Measures drop target (for example, for pivot tables) or might contain sub- drops targets (for example, the Bars (Y1-Axis) and the Lines (Y2-Axis) for line bar graphs). You drag and drop measure columns to these drop targets.
Other drop targets — Used to summarize the columns in the measure drop target or targets. You generally drag and drop attribute and hierarchical columns to these drop targets.
The other drop targets that are displayed in a view depend on the type of view as described in Table 3-8.
|View||Drop Target or Targets|
Includes these drop targets:
Rows — Shows the columns that are displayed in the gauges.
Bar, line, area, line-bar, time series line, pareto, scatter, or bubble graph
Includes these sub-drop targets within the main drop target (Bars, Line, Area, Bars & Lines, Line, Points or Bubbles). Target names differ depending on graph type. For example, Group By is Bubbles for bubble graphs and Points for scatter graphs.
Radar Sections — Shows column values as points on each line along a radius of the circle.
Includes these drop targets:
Stage — Shows each column value as a stage in the funnel.
Includes these drop targets:
Includes these sub-drop targets within the main drop target Visualization for simple trellis views:
Includes these sub-drop targets within the main drop target Visualization for advanced trellis views:
Excluded — Excludes columns from the view results but leaves the columns as part of the analysis. See "Columns in the Excluded Drop Target" for more information.
In addition, each <view-type> area, except for the Table area and the Trellis area, contains the Measure Labels element. The Measure Labels element represents the labels for all the Measures columns in the drop targets in the Measures area. You can modify how measure labels are shown in a view by dragging it from one drop target and dropping it in another. For example, in a vertical bar graph, you can show each measure label in a different color by dragging and dropping the Measure Labels element to the Vary Color By drop target.
In pivot tables, you can also edit the format of measure labels or hide them.
A column in the Excluded drop target is not included in the view results but still remains as part of the analysis. A column can be placed in the Excluded drop target after views have been created for the analysis in various ways. A general rule is that a column is placed in the Excluded drop target for a view if it is not added explicitly to one or all views.
Some of the ways in which a column is placed in the Excluded drop target are as follows:
You select the Exclude Column option from the right-click menu in a table or pivot table view.
You select a column from the Subject Areas pane to add to views in an analysis as described in the following scenarios:
Criteria tab — If you add a column from the Subject Areas pane to the Selected Columns pane after displaying the analysis results and the Exclude from existing views, but display in new views option is selected for the Display of Columns Added in the Criteria tab property in the "Analysis Properties dialog: Data tab"), then the column is placed in the Excluded drop target of the existing views' Layout pane but included in any new views that you add.
Results tab — The behavior might differ depending on whether you add columns to a view editor or to the Compound Layout:
View Editor — If you add a column from the Subject Areas pane to a view, then you place the column in that view. The column is placed in the Excluded drop target for all other views in the analysis.
Compound Layout — If you double-click a column in the Subject Areas pane, then you place the column in a default drop target of the Layout pane for all existing views in the current Compound Layout. The column is placed in the Excluded drop target for all other views in the analysis.
If you drag and drop a column from the Subject Areas pane to a tabular view, then you place the column in the drop target for that view. The column is placed in the Excluded drop target for all other views in the analysis.
If you want a column that is in the Excluded drop target to be displayed in a view, then you can easily move it. Simply display the Layout pane for the view, and drag and drop the column from the Excluded drop target to the desired one.
Excluding columns differs from removing columns. You can use the Remove Column option from the More Options button in the Layout pane for a view to remove a column entirely from the analysis.
In a pivot table or graph that includes columns in the Excluded drop target, an aggregation rule is applied to aggregate measures in the data body of the view to a single value. For example, suppose that you have the following columns in the Selected Columns pane:
Region City Dollars East NY 1000 East Boston 500
If the City column is placed in the Excluded drop target, then the view typically displays the following:
The aggregation rule is applied to aggregate 1,000 and 500 into 1,500. In a pivot table or graph, the aggregation rule that is specified in the "Edit Column Formula dialog" applies. For a pivot table, you can select a specific aggregation rule using the More Options menu in the Layout pane.
Suppose that you wanted the table, pivot table, or graph to show the following values:
East 1000 East 500
To achieve this aggregation, include both Region and City columns in the view layout, but hide the City column using its "Column Properties dialog: Column Format tab".
The following restrictions and guidelines apply to dragging columns from one drop target and dropping them in another in graphs and funnel graphs:
A bubble graph requires at least three measures. Plot one measure on the horizontal axis, another measure on the vertical axis, and a third measure on the bubble size axis.
In a bubble, line-bar, scatter, or funnel graph, you cannot drag and drop measure labels.
A pareto graph can have only one measure.
If you drop another measure on the Measures drop target, then the measures are swapped; that is, the existing measure is replaced by the newly dropped measure and is moved automatically to the Excluded drop target.
A time series line graph requires a single date or date-time data column to be selected on the horizontal axis. It has a single vertical axis, but supports multiple data series.
A scatter graph requires at least two measures. For example, you can plot one measure column on the horizontal axis and another measure column on the vertical axis. These measures are plotted for values on the Group By axis.
A funnel graph uses two measures but only one is required. If you do not select a second measure, then the first measure is used for the second measure. If you have selected two measures and then select a new measure, then the new measure replaces the measure currently in the Actual Measures drop target.
The following guidelines apply to working with drop targets in trellises:
Expanding Drop Targets in Trellises — The Layout Pane of the trellis editor is notably different in that it is vertical rather than horizontal. When drop targets contain more data than can be shown within this more narrow Layout Pane, the drop target containers expand on hover. That is, when you hover the mouse over an overfilled drop target, you see the complete selection of measures and columns, and you can move and reorder even the measures and columns that were not visible before the hover. The chosen item appears with a slight transparency as you drag it.
Placement of Measures — The following are considerations to keep in mind as you work with measure columns in trellis views:
Measures can be reordered within the measures drop targets by dragging left or right.
In simple trellis views, measure columns can be placed only on the measure edge of Color By or Group By.
In some simple trellis visualizations (scatter graphs, bubble graphs, and line-bar graphs), you can swap measures. If you drag an existing measure from one axis and drop it in the target for another axis, the two measures' positions in the view are swapped.
Scatter graphs, bubble graphs, and line-bar graphs allow a single measure per axis, as follows:
Scatter graph drop targets: Horizontal Axis, Vertical Axis
Bubble graph drop targets: Horizontal Axis, Vertical Axis, Size
Line-bar graph drop targets: Bar Axis, Line Axis
In advanced trellis views, measures comprise the innermost column headers of the trellis.
When moving measures from the Color By drop target to or from the Group By drop target:
Dragging a single measure moves all the measures along with it. (This is known as sticky behavior.)
Exception: Upgraded analyses automatically select the Treat as an Attribute Column box for measures (in the "Edit Column Formula dialog: Column Formula tab"), so this sticky measure behavior only happens if you deselect the Treat as an Attribute Column box for existing measures.
Dragging a new measure into the view moves all existing measures to wherever you place the new measure.
To place a measure on the non-measure edge of a visualization, or in the Rows target or Columns target, you must first convert the measure to an attribute column. You do this in the "Edit Column Formula dialog: Column Formula tab".
Attribute columns can be dragged out of the Measures drop target without causing the drop target or the measures inside it to move with the attributes.
Using the view editor and the Layout pane, you can easily modify and change the order of columns in the following ways:
Drag and drop columns in tables and pivot tables to the desired positions in the editor using the handles and drop targets. For example, if you have two columns in the Rows section of a pivot table, reverse the order of the columns by dragging and dropping the first column after the second one. In a table, you can drag and drop columns, but you cannot stack columns, as you can in a pivot table.
You can also drag and drop columns in this way in the Compound Layout.
Drag and drop columns in the Layout pane. A target is active and ready for the "drop" when it appears highlighted. When you hover the mouse pointer over a column in the Layout pane, the cursor state changes from the pointer to a move cursor when you can "grab" the column and move it over a drop target. For example, you can easily move a column in a pivot table from the Rows drop target to the Sections target to create a unique pivot table for each value in the column. For details of dragging and dropping columns in the Layout pane of the trellis editor, see "Drop Target Guidelines for Trellises".
For more information on drop targets, see the "Layout pane".
Add a column to a graph or gauge. To do so, drag the column from the Subject Areas tab to the appropriate location in the view editor or to a drop target in the Layout pane.
Remove columns in the Layout pane. For example, you can easily remove a column in a view by selecting Remove Column from the More Options button. Removing columns differs from excluding columns, which is described in "Understanding Drop Targets".
This list provides only a partial summary of where dragging and dropping can occur. You can drag and drop columns and catalog objects whenever you see the "Subject Areas pane" and "Catalog pane". You can also drag and drop items in views on dashboards. See "Columns in the Excluded Drop Target" for more information.
In the Layout pane, you can click the Properties button to display a dialog in which you specify properties for the view body (such as Pivot Table) or for a drop target (such as Sections). For example, you can specify whether to display headings, set background color, and insert page breaks.
To modify view body or drop target properties:
In the view editor, display the Layout pane and click the Properties button beside the view body (for tables and pivot tables) or drop target name.
The Properties dialog for the view body or drop target is displayed.
Complete the appropriate fields and click OK.
In the Layout pane, you can add totals for columns in tables and pivot tables. You can position the totals at various locations in the view. You can add totals for columns that are displayed on the various edges. For each measure, the total uses the aggregation rule for that measure. The default aggregation rule for a measure column is specified in the Oracle BI repository, or by the original author of the analysis.
If you specify a total in the Rows or Columns drop target of a pivot table, then the totals that are displayed are the result of the columns that are specified in the Measures drop target. Total values are not displayed on the Columns or Rows edges of the pivot table but rather in the data in the center of the pivot table.
To add totals to a view:
Display the "Layout pane" for the view.
To add grand totals to the entire view:
For a table, in the Columns and Measures drop target, click the Totals button, then click the location such as Before.
For a pivot table, in the Rows or the Columns drop target, click the Totals button, then click the location such as After.
To turn on and off the totals that apply to all the values in the drop target, click the Totals button beside the drop target name, such as Sections. Then select the location for the total, such as Before the data items. A totals area is added to the view.
In previous releases (prior to 11g), you had the ability to create report-based totals in table views. Because report-based totals are handled slightly differently in this release, see "Upgrading Report-Based Totals" in Oracle Fusion Middleware Upgrade Guide for Oracle Business Intelligence for information.
You can use the Layout pane to display running sums or the relative value of measure columns in pivot tables, as described in the following sections.
In a pivot table, you can display numeric measures as running sums, where each consecutive cell for the measure displays the total of all previous cells for that measure. This option is a display feature only that has no effect on actual pivot table results.
Typically, running sums are displayed for duplicated attribute columns or for measure columns for which the option to show data as a percentage of the column has been selected, with the last value being 100 percent. Running sums apply to all totals. The running sum for each level of detail is computed separately.
Column headings are not affected when the running sum option is selected. You can format the column heading if you want it to indicate that the running sum option is in effect.
The following usage rules are in effect for running sums:
A running sum is incompatible with the SQL RSUM function (the effect would be a running sum of the running sum).
All running sums are reset with each new section. A running sum does not reset at a break within a section or continued across sections.
If a measure does not display in a single column or in a single row, then the measure is summed left to right and then top to bottom. (The lower right cell contains the grand total.) A running sum does not reset with each row or column.
You cannot specify rolling minimums, maximums, and averages in the Layout pane. You can include these if administrators create formulas for them in the metadata repository.
To display a measure as a running sum:
In the Layout pane for the pivot table, in the Measures area, click the More Options button for the row or column to be summed and select Display as Running Sum.
In a pivot table, you can dynamically convert a stored or calculated measure into a percent or an index. This shows the relative value of the item, compared to the total, without the need to explicitly create a calculated item for it. You can view the measure as a percentage between 0.00 and 100.00, or as an index between 0 and 1.
For example, if you are using a pivot table to examine sales by product, then you can duplicate the sales measure and view it as a percentage of the total. This enables you to see the actual sales, and the percentage of sales, that each product accounts for.
To show an item as a relative value:
In the Layout pane for the pivot table, click the More Options button for the item you want to show as a relative value.
The following step is optional. When you duplicate the measure column in the pivot table, you can see both the total for the measure and its relative value. This eliminates the need to add the column twice on the Criteria tab to see the total and its relative value in the pivot table.
To duplicate the measure column, select Duplicate Layer.
The measure is displayed a second time in the pivot table, with the same name.
Click the More Options button, select Show Data As, select Percent of or Index of, then select the appropriate option.
The Show Data As option is available only for items that are stored or calculated measures.
To rename the new column, click the More Options button, select Format Headings and in the Edit Format dialog, enter a value in the Caption field.
A section slider displays members of one or more attribute or hierarchical columns as values on a rectangular bar and provides mechanisms to select a value. You use a section slider to limit the data that is shown in a graph or gauge.
Figure 3-15 shows an example of a section slider. It displays the members of the Product Type column. Using this slider, for example, you can limit the data that is shown in a graph or gauge to a particular product type, such as Cell Phones.
A section slider consists of the following components:
Slider bar — Displays the members of one or more attribute or hierarchical columns as values along a rectangular bar.
The administrator configures the maximum number of values that can be displayed on the slider bar. However, you can set a maximum lower than the system maximum by setting the Maximum number of section slider values field in the "Section Properties dialog".
Slider thumb — Indicates the current value of the section slider. The thumb in Figure 3-15 indicates that the current value is Cell Phones. You can drag the thumb to the desired value.
Decrease button — Moves the slider thumb to the value to the left of the current value. For example, in Figure 3-15, clicking the decrease button (the second button from the left on the slider) moves the slider thumb to the value Camera.
Increase button — Moves the slider thumb to the right of the current value. For example, in Figure 3-15, clicking the increase page button (the right-most button on the slider) moves the slider thumb to the value Fixed.
Play button — Sequentially moves the thumb through the slider values. For example, in Figure 3-15, clicking the play button (the left-most button on the slider) moves the slider thumb to the value Fixed and then to Install, and so on. After being clicked, the play button changes to a pause button to allow you to stop on a particular value.
To define a section slider in a graph or a gauge:
Open the analysis that contains the graph or gauge in which you want to define a section slider.
Click the "Analysis editor: Results tab".
Click the Edit View toolbar button in the graph, funnel, or gauge view.
For each column to be displayed on the section slider, in the "Layout pane" of the Graph or Gauge editor, drag and drop the column to the Sections drop target.
Select Display as Slider.
Click Done to close the editor and return to the Results tab.
Click the Save Analysis toolbar button to save your changes.
Users of a dashboard page or an analysis might have the ability to modify the data that they see in a table. This ability is often referred to as "write back." Users can update a value in the view that is written back to the record in the data source or is used in other views in the analysis.
For example, you can create an analysis that has Sales Quota Amount defined as a write back field, Sales Amount as a field from the data warehouse, and Percentage of Quota as a calculated field (Sales Amount/Sales Quota Amount). When viewing the analysis a user can modify the Sales Quota Amount and the Percentage of Quota field recalculates appropriately.
The ability to modify values affects the following types of users:
Administrators of the repository configure the columns in the repository to allow for write back.
Administrators set privileges for write back and prepare a write-back template, and administrators and content designers configure for write back from table views in dashboards and analyses.
End users modify records in table views.
The following list describes the steps in the process of write back:
As the content designer, you work with the administrator of the repository to assess the reporting needs in the organization and make a list of write-back columns needed and the analyses in which they should be displayed.
Hierarchical columns do not support the write-back capability but attribute columns, measure columns, and double columns do support the write-back capability. For double columns, you can write back to the display column. No automatic translation of the code column is provided.
The administrator of the repository configures the Oracle BI Server and the columns that should be available for write back, as described in Oracle Fusion Middleware Metadata Repository Builder's Guide for Oracle Business Intelligence Enterprise Edition.
The administrator performs the following tasks:
Prepares a file of SQL code to use as the "template" for write back.
Grants appropriate privileges for setting up and using write back. For example, the administrator gives you the privileges to set up views in which users can modify values.
For information on the write-back template and on privileges, see "Configuring for Write Back in Analyses and Dashboards" in Oracle Fusion Middleware System Administrator's Guide for Oracle Business Intelligence Enterprise Edition.
As the content designer, you enable one or more columns for write back, as described in "Adding the Write-Back Capability to a Column".
As the content designer, you enable a table view for write back, as described in "Adding the Write-Back Capability to a Table View".
As the content designer, you provide users with information on working with write back, as described in "Modifying Data in a Table View in a Dashboard Page or Analysis".
Users modify the values in views as appropriate.
To enable the write-back capability for a column, administrators and content designers use the Write Back tab on the Column Properties dialog. Correct privileges must be set, and the attribute, measure, or double column must be set as writeable in the repository. Ensure that the column relates to the write-back template that the administrator prepared. All the table views that include that column can display it as available for write back.
To enable the write-back capability for a column:
Open the analysis for editing in the Criteria tab.
Click the Options button beside a column on which you want to enable write back and select Column Properties.
The "Column Properties dialog" is displayed.
Click the Write Back tab.
If the column has been enabled for write-back in the repository, then the Enable Write Back box is available.
Click the Enable Write Back box.
Additional options for write-back are displayed.
In Text Field Width box, specify the width of the column.
Click OK to close the Column Properties dialog.
To enable the write-back capability for a table view, use the Write Back tab of the Table Properties dialog.
To enable the write-back capability for a table view:
In the Analysis editor, open the table view for editing.
Click the Table View Properties toolbar button.
The "Table Properties dialog" is displayed.
Click the Write Back tab.
Ensure that the Enable Write Back box is selected.
Select the Template Name box, then type the name of the template to employ for writing back values.
Check with the administrator for the appropriate template name.
If you enable write back in a view and accept the system defaults, then a user has two modes in which to work with values:
View mode: In this mode, a user simply views values. The user lacks the ability to modify values, before clicking the Update button to enter Edit mode.
Edit mode: In this mode, a user actually modifies values. To enter Edit mode, the user clicks the Update button on the view, when it is available. When in Edit mode, the user enters data in columns and can click the following buttons:
Revert: Changes back to the original values any modifications that the user has made but has not yet written back to the data source. Keeps the user in Edit mode.
Apply: Writes back to the data source any changes that the user has made and refreshes the view to display the most current data based on those changes. Keeps the user in Edit mode.
Done: Writes back to the data source any changes that the user has made, refreshes the view to display the most current data based on those changes, and returns the user to View mode.
You can modify the labels for these buttons using the options in the Write Back tab of the Properties dialog for the view. You can also use the Toggle Table Mode box to indicate that users do not toggle between View and Edit mode. Instead, they remain always in Edit mode. When toggling is disabled, the Update button is not displayed on the view.
You can specify the types of interactions that are to occur when users:
Click in a data view. See "Click Interactions in Views".
Right-click in a table, pivot table view, or trellis at runtime. See "Right-Click Interactions in Views"
You can specify the types of interactions that are to occur when users click (that is, left-click) in a data view. The following types of interactions are available:
None — Specifies that nothing happens when users click the column heading or a value. This option turns off drilling for attribute columns; it does not turn off drilling for hierarchical columns.
Drill — Enables users to drill down to more detailed information. See "Drilling in Views".
Action Links — Enables users to click a hot spot in a data view and then select an action link to execute an action, such as navigating to a saved analysis or invoking an Enterprise Java Bean (EJB). See Chapter 10, "Working with Actions."
You create interactions at the criteria level using the "Column Properties dialog: Interaction tab". Interactions that you create take effect in all data views (that is, tables, pivot tables, graphs, funnel graphs, trellises, gauges, and maps).
In previous releases (prior to 11g) of Oracle BI EE, you could set up interactions at the view level. In this release, you create all left-click interactions at the criteria level. For more information, see "Upgrading Interactions in Views" in Oracle Fusion Middleware Upgrade Guide for Oracle Business Intelligence.
You can specify the types of interactions that are to occur for an analysis when users right-click in a table, pivot table view, or trellis at runtime. The following right-click interactions are available:
Drill (when not a primary interaction)
Create/Edit/Delete Calculated Items
Display/Hide Running Sum
You enable right-click interactions for an analysis using the "Analysis Properties dialog: Interactions tab". Right-click interactions that you enable take effect in all data views except graphs, gauges and maps.
Right-click interactions are not available for graphs in trellis views (Simple Trellis and Advanced Trellis), but you can right-click on text in the trellis view in the same manner as in a pivot table. This includes data cells in an advanced trellis that have Visualization set to Text.
If the administrator has configured columns for drilling in the subject area, then you can allow users to drill in data in tables, pivot tables, graphs, trellises, gauges, and maps. Drilling is a way to navigate through data in views quickly and easily. This section provides the following information on drilling:
You can drill in attribute columns and hierarchical columns. For information on drilling in columns, see "Drilling in Results".
As the content designer, you specify whether users can drill in particular columns in views on dashboards. You control whether drilling is allowed in particular columns by specifying options in the "Column Properties dialog: Interaction tab".
If drilling is not a primary interaction of a particular column (as set in the "Column Properties dialog: Interaction tab"), then you can allowing drilling as a right-click interaction in table and pivot table views at runtime. To do so, you select the Drill (when not a primary interaction) option in the "Analysis Properties dialog: Interactions tab".
Drilling in columns affects their filters and selection steps as described in the following list. For more information, see Chapter 5, "Filtering and Selecting Data for Analyses."
Hierarchical columns: No steps are added to the selection when you expand or collapse members in a hierarchical column. That is, the expanding and collapsing does not change the selection of data for the column.
For example, suppose that you create a pivot table in which you select 2008 as the only member in the Time dimension, and you arrange the data so that this one Time member is the column header in the pivot table. You can expand to show quarters in 2008 and then the months in the last quarter. At this point the pivot table has child members for 2008, Q1 2008, Q2 2008, Q3 2008, Q4 2008, October 2008, November 2008, and December 2008. If you display the Selection Steps pane, however, you see that the selection for the Time dimension still contains only the 2008 member.
Expanding and collapsing in a hierarchical column affects only that particular view. No other views are affected.
Attribute columns: You can drill down from the row heading or column heading or from a member in an attribute column. Drilling on a heading adds the lower level to the view. Drilling on a member adds the lower level and affects both filters and selection steps:
Drilling on a member adds a filter for the current member, thereby limiting the results. For example, if you drill on the Game Station member in a table that includes the P1 Product column, you add the E1 Sales Rep Name column, which adds a filter that specifies that P1 Product equals Game Station.
Drilling on a member adds the lower-level column to the analysis and updates the column in the Selection Steps pane without providing a step update during design.
See the following topics that are related to drilling:
For information on drilling in maps, see "Drilling in Map Views".
For information on configuring for drilling in views in dashboards, see "Adding Content to Dashboards".
For information on configuring hierarchical columns for drilling, see Oracle Fusion Middleware Metadata Repository Builder's Guide for Oracle Business Intelligence Enterprise Edition.
When you create an analysis, you can specify sorting for the results that are displayed in graphs, pivot tables, tables, and trellises.
Some sort functionality is common to graphs, pivot tables, tables, and trellises, although you cannot sort values interactively on a graph.
Graphs, pivot tables, tables, and trellises reflect the sorting that you specify in the Selected Columns pane of the Criteria tab for attribute columns and hierarchical columns. Use the Sorting menu options that are available from the Options button for a column. These sorting options provide the ability to sort by a single column or by multiple columns, each in either ascending or descending order. By creating sorts for multiple columns, you can specify multiple level sorts such as second-level and third-level sorts of the data. These sort levels are indicated with numbers and a sort icon for the column in the Criteria tab.
The options that you specify on the Criteria tab serve as the initial sorting state of the analysis. You can modify the sort later when the analysis is used in a table, pivot table, or trellis.
When you sort items in a hierarchical column, you always sort within the parent. That is, children are never sorted outside of their parent. The children are displayed below the parent in the proper sort order; the parent is not sorted within its children.
If you include totals in a view, then those values are not sorted. They remain in the location in which you placed them, regardless of sorting.
If you sort an attribute column or hierarchical column that includes a group, then that group is always displayed at the bottom of the list. If there are multiple groups, then the groups are listed in the order in which their corresponding steps are specified in the Selection Steps pane.
Sorting allows you to quickly sort rows and columns of a view, either from lowest to highest data values, or from highest to lowest. You can also return the order to the order in the data source by clearing all sorts. You can specify alphanumeric sorts on the row and column edges of pivot table views, table views, and trellis views.
You make sorting specifications in various ways, including those in the following list:
In the Selected Columns pane of the Criteria tab, click the Options button beside the column and select the appropriate sort option from the menu. For more information on the sorting options menu, see "Sorting Options Menu".
Right-click in a pivot table, table, or trellis, select Sort or Sort Column (whichever is available), and then select the appropriate sorting option from the menu. For more information on the sorting options menu, see "Sorting Options Menu".
Hover the mouse pointer over the area to the right of the column name in the header of a pivot table, table, or trellis and click either the upward-facing triangle (Sort Ascending) or the downward-facing triangle (Sort Descending) for that column. These sort specifications override those that you make with the right-click menu.
If you see a shaded-in sort button in the column header or the row header, then you know that the column contains a primary sort. You can add a second-level or third-level sort by hovering over another innermost column header or row header and clicking the appropriate sort button or clicking the right-mouse button.
Any sorting options that you specify in a view override those that were made in the Selected Columns pane.
You can use various methods to clear sorts:
For sorts that have been applied in the Selected Columns pane, click Clear All Sorts in All Columns. The sort specifications that you made in the Selected Columns pane are removed. Sorts that you specified in a view remain.
For sorts that have been applied directly in a pivot table, table, or trellis:
Click a sort button in an unsorted column to remove the primary sort from the column to which it now applies and apply it to the column whose button you just clicked.
Select the Clear All Sorts in View option from the right-mouse menu.
Suppose that you have a pivot table with Time Hierarchy on the column edge and H2 Offices on the row edge; both are hierarchical columns. 1-Revenue in dollars is the measure that is shown in the pivot table. You can sort H2 Offices by the Revenue for a particular year as highest revenue to lowest revenue, or lowest to highest, or no sorting at all.
Figure 3-16 shows an example of sorting H2 Offices by Revenue for year 2008, in ascending order. The Offices Total parent is always displayed above its children. The values for other years might not be sorted numerically, because only year 2008 was selected.
Figure 3-17 shows an example of sorting the Time Hierarchy by the Revenue for Offices Total. The years are re-arranged in the order 2009 and 2008. In 2009, Offices Total had its lowest value; and in 2008, it had its highest value.
Suppose that you have a pivot table with the Time Hierarchy on the column edge and both the Products Hierarchy and H2 Offices on the row edge, with Products outside H2 Offices. 1-Revenue in dollars is the measure that is shown in the pivot table. You can sort the row edge by the Revenue for 2008 in descending order. The H2 Offices are sorted as in Example 1, but the Products are not sorted, because they are in the outermost layer. See Figure 3-18.
Suppose that you have a pivot table with the Time Hierarchy on the column edge and H2 Offices on the row edge. Both 1-Revenue and 2-Billed Quantity are the measures that are shown in the pivot table. If you try to sort the years by the leaf-node for GenMind Corp using the upward facing triangle (Sort Ascending), then the first measure on the column edge (reading left to right) is chosen as the one on which to sort. If Revenue is the first measure on the edge, then the pivot table is sorted as shown in Figure 3-19.
If the measures are on the row edge and you sort by a year, then the top-most measure is used for sorting.
If you sort using the Sort right-click interaction, then options that allow you to select which measure to use in the sort are displayed.
In previous releases (prior to 11g) of Oracle BI EE, you could create sorts for columns in views. If you upgrade from a previous release (prior to 11g), then you might notice that the views are sorting the columns slightly differently. For information, see "Upgrading Sorts" in Oracle Fusion Middleware Upgrade Guide for Oracle Business Intelligence.
You can save a view that you are working with at any time. To save a view, you must save the analysis.
To save a view:
If you want to save a view:
In a new analysis or in an existing analysis, click the Save Analysis toolbar button in the "Analysis editor".
In an existing analysis that you must save as another name, click the Save As toolbar button in the "Analysis editor".
If this is the first time that you are saving the analysis, or you are saving an existing analysis as another name, then the "Save As dialog" is displayed where you specify the information for saving the analysis.
You can rename a view or a compound layout.
To rename a view or compound layout:
If you want to rename:
A view, on the toolbar of the view's editor, click the Rename View button.
A compound layout, in the "Analysis editor: Results tab" toolbar, click the Rename Compound Layout button.
The "Rename View dialog" is displayed.
Enter the new name for the view or compound layout.
You can use your mouse to rearrange a view (for example, title, legend, table, or graph) within a compound layout.
To rearrange a view:
Place the cursor just inside the top edge of the view to be rearranged.
Click and hold the left mouse button (if configured) on the view.
The view is displayed as a transparent movable object.
Use the mouse to drag the view to the required position.
A view can be rearranged as follows:
At the boundary of a view.
Each view has four boundaries (upper, lower, left or right), and can be arranged alongside the boundaries of other views in an analysis. You can drag and drop a view to display it next to the boundary of another view.
At the outer boundary of the compound layout.
Each compound layout contains one or more views arranged in either a square or a rectangle. The compound layout has four boundaries (upper, lower, left and right). You can drag and drop a view to the outer boundary of an compound layout. This action displays the view across the entire length or breadth of the compound layout.
Release the mouse button.
The view is rearranged in the compound layout in the position that is indicated by the blue bar.
Repeat the previous steps to rearrange further views.
You can print views using HTML or Adobe PDF (Portable Document Format). Adobe Reader 6.0 or greater is required to print using Adobe PDF.
You can also specify PDF and print options, including adding headers and footers. For more information on printing, see "Changing Print Options for Views"
The HTML method of printing relies on the print handling capabilities of your browser. If you do not get the results you want, then choose PDF to open and then print the analysis.
You can print:
A single view from the view's editor
An assemblage of views that are displayed in the "Compound Layout"
To print a single view or an assemblage of views:
If you want to print:
A single view, on the toolbar of the view's editor, click the Print this analysis button.
The assemblage of views displayed in the Compound Layout, in the "Analysis editor: Results tab" toolbar, click the Print this analysis button.
Select Printable HTML or Printable PDF.
For HTML, a new window shows the view or views to print.
From the File menu, select Print.
For PDF, an Adobe PDF window shows the view or views to print.
Use the options that are available in the Adobe PDF window to save or print the file.
To change print options:
You can preview views to see how they are displayed on a dashboard. You can preview:
A single view from the view's editor
An assemblage of views displayed in the "Compound Layout"
To preview a view:
If you want to preview:
A single view, on the toolbar of the view's editor, click the Show how results will look on a Dashboard button.
The assemblage of views that are displayed in the Compound Layout, in the "Analysis editor: Results tab" toolbar, click the Show how results will look on a Dashboard button.
If the Show how results will look on a Dashboard button is not available, then the administrator has suppressed its display.
The dashboard preview is displayed in a new window. Any prompts are displayed and applied in the preview.
You can remove a view from:
A compound layout.
Removing a view from a compound layout removes it only from the compound layout, not from the analysis.
Removing a view from an analysis removes the view from the analysis and from any compound layout to which it had been added.
To remove a view:
When you are working with views that show results data, such as the table and pivot table, you can refresh the results of the current analysis. Refreshing results is useful if you make a change such as adding a filter that limits results and you want to see the effects of your change.
To refresh the results of the current analysis:
In the "Analysis editor: Results tab" toolbar, click the Refresh the results of the current analysis button.
Delivery content is automatically formatted based on the destinations that you specify in the "Agent editor: Destinations tab". You can manually control which view or views are sent to a particular destination by including them in a compound layout. For example, you might want to create a view that contains mostly text and add that view to a compound layout, if you plan to deliver it to a mobile phone. In previous releases (prior to 11g), this functionality was referred to as "specialized views."
To control which compound layout is sent to a particular destination:
For the appropriate analysis, create and save the view to use for delivered content.
On the toolbar of the "Analysis editor: Results tab", click the Create Compound Layout button.
Add the view that you created to the newly created compound layout.
On the toolbar of the "Analysis editor: Results tab", click the Edit Analysis Properties button.
In the View for Text Delivery box, select the compound layout to which you added the view.
Master-detail linking of views enables you to establish a relationship between two or more views such that one view, called the master view, drives data changes in one or more other views, called detail views.
For example, suppose you have the following two views:
A table that shows Dollars by Region
A graph that shows Dollars by Brand with Region on a section slider
Using the master-detail linking functionality, you can link the two views so that when you click a particular Region in the table, the Region on the section slider of the graph as well as the data in the graph changes to reflect the Region that was clicked on the table. For example in Figure 3-20, clicking CENTRAL REGION in the table, positions the thumb in the section slider on CENTRAL REGION and updates the data in the graph to the data for the CENTRAL REGION.
Master-detail linking is not supported when you click the plot area of a graph.
A master view drives data changes in one or more detail views. A view becomes a master when you set up the interaction of a column in the view to send master-detail events on designated channels. This column is known as the master column.
The master column is the column whose values when clicked send a master-detail event, which passes information to update the data in a detail view.
The following types of views can be master views:
Trellis (only the outer edges, not the inner visualizations)
A master view can be in the same analysis as the detail view or in a different analysis. A master view can update data in one or more detail views.
Any type of column — hierarchical, attribute, or measure — can be a master column. However, the master column cannot be displayed on the page edge or the section slider in the master view. It must be displayed in the body of the view.
While hierarchical columns are supported as master columns, if a user clicks a hierarchy level, then a change takes effect in the detail view only if the level is expanded there as well.
When a master-detail event is raised, it passes the definition of the cell (or item) in the master view that was clicked. The specific information it passes depends on the type of column:
For an attribute column, it passes the clicked value and the contextual information to the left and above the axis member.
For example, in Figure 3-21, if the attribute column District is the master column, and you click CHICAGO DISTRICT (highlighted in blue), then the master-detail event information passed is Region = CENTRAL REGION and District = CHICAGO DISTRICT.
For a measure column, it passes all the contextual information for the data value.
For example, in Figure 3-22, if the measure column Dollars is the master column, and you click 4724882.00 (highlighted in blue), then the master-detail event information passed is Region = CENTRAL REGION, District = CHICAGO DISTRICT, and Year = 2000.
The master-detail event information defines the position of a page edge or a section slider in a detail view.
A detail view is a view that listens for and responds to master-detail events sent by a master view on a specified channel. A view becomes a detail view, when you set up the view to listen to master-detail events.
A detail view includes one or more columns whose values are changed directly by the information passed by a master-detail event. These columns are known as detail columns. A detail column must:
Be on the page edge or the section slider of the detail view. Whether a detail column is displayed on the page edge or a section slider, depends on the type of view:
For pivot tables and tables, it must be on the page edge
For graphs and gauges, it can be on the page edge, or on the section slider
If a detail column is on a section slider and a page edge is present, the detail column is not updated by the information in the master-detail event. For example, if the information in the event is Region=CENTRAL and Year=2009, and the page edge of the detail view has Region and the section slider has Years, then only the Region is updated.
Match a column in the master view
When a detail view contains multiple detail columns, it is listening for and responding to master-detail events on multiple channels.
The following types of views can be detail views:
Trellis (only the outer edges, not the inner visualizations)
A detail view:
Can listen for master-detail events from multiple master views
Can be in the same analysis as the master view or in a different analysis
Cannot act as a master to another view
When a master view and a detail view are linked and a user clicks a cell (or item) in the master column, a master-detail event is generated on the designated channel and the master-detail event information is passed to the detail view.
The detail view (which is listening on the designated channel for master-detail events) receives the event and the master-detail event information. It reviews the master-detail event information to determine which column in this information matches the detail column. When it finds a match, it takes the column values in the master-detail event information, updates the detail column, and then refreshes the entire detail view.
If a master column is on a section slider, then any detail view that is not visible initially in the browser is not updated and refreshed until the view becomes visible when you scroll down. For more information on section sliders, see "Defining Section Sliders in Graphs and Gauges".
For an example of how master-detail linking of views works, see "Example of a Table and Graph Linked in a Master-Detail Relationship".
To link views in master-detail relationships, you must define the master view and the detail view.
For more information on master-detail linking, see "What Is Master-Detail Linking of Views?"
To define a view as a master view:
Create or edit the analysis whose view (or views) you want to define as the master view.
Click the "Column Properties dialog: Interaction tab".
In the Primary Interaction box in the Value area, select Send Master-Detail Events. The Specify channel field is displayed.
In the Specify channel field, enter a name for the channel on which the master view sends master-detail events. You can use any appropriate name, for example, Sales Analysis Channel, Channel 1, or channel 1.
Add the view to use as a master view. Note that any view that you add that includes the master column can be used as a master view.
To define a view as a detail view:
Create or edit the analysis whose view you want to define as the detail view.
Create the view that is to be the detail view, for example, a graph.
Edit the view.
Click the view's Property button on the toolbar of the view's editor. For example for a graph, click the Graph Properties toolbar button in the Graph Editor. The view's property dialog is displayed, for example, for a graph, the "Graph Properties dialog".
In the "Graph Properties dialog: General tab" (for graph or funnel graph), the "Gauge Properties dialog: General tab" (for a gauge), the "Table Properties dialog: Style tab" (for a table), or the "Pivot Table Properties dialog" (for a pivot table, select the Listen to Master-Detail Events box.
In the Event Channels field, enter the name of the channel on which the detail view is to listen for master-detail events. The name must match (including case) the name of the channel on which the master view sends master-detail events. (For more information on the channel in the master view, see "Defining the Master View".
In the "Layout pane", drag the column (or columns) to be the detail column (or columns) to either the Prompts drop target or the Section drop target as follows:
If the detail view is a pivot table or table, then drag the column or columns to the Prompts drop target.
If the detail view is a graph or a gauge, then drag the column or columns to the Prompts drop target or to the Sections drop target, whichever you prefer, unless the graph or gauge has a column on the Prompts drop target. Then, you must drag the column or columns to the Prompts drop target.
You must drag and drop all detail columns to the same drop target. You cannot have some on the Prompts drop target and some on the Section drop target.
For more information on drop targets, see "Understanding Drop Targets".
If you dragged and dropped one or more detail columns to the Section drop target, then select the Display as Slider box.
In this example, a table, Regional Sales Master View, is linked to a graph, Regional Sales View Detail View, in a master-detail relationship.
The analysis for the master view contains the Region and Dollars columns. The Dollars column is set up as the master column as shown in Figure 3-23. The view that was added to be the master view is a table.
The analysis for the detail view contains the Region, Brand, and Dollars columns. The view that was added to be the detail view is a horizontal bar graph and is set up to listen for master-detail events on Channel1, as shown in Figure 3-24.
The Region column is set up to be the detail column. This was done by dragging and dropping the Region column to the Graph Prompts drop target in the Layout pane as shown in Figure 3-25. (Notice that the Region column is then displayed on the page edge in the table.)
When a user clicks on the cell that contains the dollars for the Eastern region in Regional Sales Master View, the view sends a master-detail event on Channel1 and passes the master-detail event information of Region = EASTERN REGION.
Regional Sales Detail View, which is listening on Channel1, receives and reviews the master-detail event information. It determines that the Region column matches a column on its page edge, updates the page edge to EASTERN REGION, and refreshes the entire view as shown in Figure 3-26.