26 Using ADF Pivot Table Components

26.1.1 Pivot Table Elements and Terminology

The following list of pivot table terms uses Figure 26-1 as a Sales Pivot Table sample in its descriptions of terms:

• Edges: The axes in pivot tables, such as:

  • Row edge: The vertical axis to the left of the rows in the body of the pivot table. In the sample, the row edge contains two layers, Year and Products, and each row in the pivot table represents a year and a product.

  • Column edge: The horizontal axis that appears above the columns in the body of the pivot table. In the sample, the column edge contains three layers, Measures, Channels, and Geography, and each column in the pivot table represents a measure value (Sales or Units), a channel indicator (All Channels), and a geographic location (World or Boston).

• Layers: Nested attributes that appear in a single edge. In the sample, the following three layers appear in the column edge: Measures, Channels, and Geography. The following two layers appear in the row edge: Years and Products.

• Header cell: The labels that identify the data displayed in a row or column. Row headers appear on the row edge and column headers appear on the column edge.

• Data cell: The cells within the pivot table that contain data values, not header information. In the sample, the first data cell contains a value of 20,000.000.

• QDR (Qualified Data Reference): An argument that maps a fully qualified data reference to a row, a column, or an individual cell. For example, in the sample Sales Pivot Table, the QDR for the first cell in the table must provide the following information:

  • Year=2007

  • Product=Tents

  • Measure=Sales

  • Channel=All Channels

  • Geography=World

26.6.1 How to Set the Overall Size of a Pivot Table

The default size of a pivot table is a width of 300 pixels and a height of 300 pixels. Instead of entering pixels for width and height, you have the option of specifying a percentage value for width, height, or both. This percentage value refers to the portion of the page that you want the pivot table to use.

To customize the default settings of a pivot table:

1. In the visual editor, display the page that contains the pivot table.

2. Click Source to display the XML code on the JSPX page.

3. Enter the following code for the inlineStyle attribute of the pivotTable tag, where value1 is an integer that represents either the number of pixels or the percentage of the page for the width of the pivot table and value2 is an integer that represents either the number of pixels or the percentage of the page for the height of the pivot table: inlineStyle="width:value1;height:value2"

Example 26-2 shows the setting of the inlineStyle attribute that specifies the width of the table as 50 percent of the page size and the height of the table as 400 pixels.

<dvt:pivotTable
.
.
.
   inlineStyle="width:50%;height:400px">
</dvt:pivotTable>

26.6.2 How to Resize Rows, Columns, and Layers

The pivot table autosizes rows, columns, and layers during design. At runtime, you can change the size of rows, columns or layers by dragging the row, column, or layer separator to a new location.

To resize rows, columns, and layers at runtime:

1. If you want to resize a row, do the following:

   1. Position the cursor in the row header on the separator between the row you want to resize and the next row.

   2. When the cursor changes to a double-sided arrow, click and drag to the desired location.

2. If you want to resize a column, do the following:

   1. Position the cursor in the column header on the separator between the column you want to resize and the next column.

   2. When the cursor changes to a double-sided arrow, click and drag the column separator to the desired location.

3. If you want to resize a layer, do the following:

   1. Position the cursor in the layer header on the separator between the layer you want to resize and the next layer.

   2. When the cursor changes to a double-sided arrow, click and drag to the desired location.

26.6.3 What You May Need to Know About Resizing Rows, Columns and Layers

When you resize rows, columns, or layers, the new sizes remain until you change the attributes of the row or column edge through a pivot operation. After a pivot operation, the new sizes are cleared and the pivot table rows, columns, and layers return to their original sizes.

If you do not perform a pivot operation, then the new sizes remain for the life of the session. However, you cannot save these sizes through MDS (Monitor and Display Station) customization.

26.8.1 How to Create a CellFormat Object for a Data Cell

To specify customization of the content of a data cell, you must code a method expression that returns an instance of oracle.dss.adf.view.faces.bi.component.pivotTable.CellFormat.

To create an instance of a CellFormat object for a data cell:

1. Construct an oracle.adf.view.faces.bi.component.pivotTable.DataCellContext object for the data cells that you want to format. The DataCellContext method requires the following parameters in its constructor:

   • model: The name of the dataModel used by the pivot table.

   • row: An integer that specifies the zero-based row that contains the data cell on which you are operating.

   • column: An integer that specifies the zero-based column that contains the data cell that you want to format.

   • qdr: The QDR that is a fully qualified reference for the data cell that you want to format.

   • value: A java.lang.Object that contains the value in the data cell that you want to format.

2. Pass the DataCellContext to a method expression for the dataFormat attribute of the pivot table.

3. In the method expression, write code that specifies the kind of formatting you want to apply to the data cells of the pivot table. This method expression must return a CellFormat object.

26.8.2 Constructing a CellFormat Object

An instance of a CellFormat object lets you specify the following arguments:

• Converter: An instance of javax.faces.convert.Converter, which is used to perform number, date, or text formatting of a raw value in a cell.

• CSS style: Used to change the CSS style of a cell. For example, you might use this argument to change the background color of a cell.

• CSS text style: Used to change the CSS style of the text in a cell. For example, you might use this argument to set text to bold.

• New raw value: Used to change the cell's underlying value that was returned from the data model. For example, you might choose to change the abbreviated names of states to longer names. In this case, the abbreviation NY might be changed to New York.

26.8.3 Changing Format and Text Styles

Figure 26-7 shows a pivot table with sales totals generated for products and for product categories. In the rows that contain totals, this pivot table displays bold text (which is a text style change) against a shaded background (which is a style change). These changes show in both the row header cells and the data cells for the pivot table. The row headers for totals contain the text "Sales Total".

The pivot table also shows stoplight and conditional formatting of data cells. For more information, see Section 26.8.4, "Creating Stoplight and Conditional Formatting in a Pivot Table".

Figure 26-7 Sales Data Per Product Category

Example 26-4 shows sample code that produces the required custom formats. The example includes the code for method expressions for both the dataFormat attribute and the headerFormat attribute of the dvt:pivotTable tag.

public CellFormat getDataFormat(DataCellContext cxt)
{
    CellFormat cellFormat = new CellFormat(null, null, null);
    QDR qdr = cxt.getQDR();
    //Obtain a reference to the product category column.
    Object productCateg = qdr.getDimMember("ProductCategory");
     //Obtain a reference to the product column.
    Object product = qdr.getDimMember("ProductId");

     if (productCateg != null && productCateg.toString().equals("Sales Total")) 
      {
       cellFormat.setTextStyle("font-weight:bold")
       cellFormat.setStyle("background-color:#C0C0C0");
      }
     else if (product != null && product.toString().equals("Sales Total")
      {
       cellFormat.setTextStyle("font-weight:bold");
       cellFormat.setStyle("background-color:#C0C0C0");
      }
    return cellFormat;
}


public CellFormat getHeaderFormat(HeaderCellContext cxt) 
{
  if (cxt.getValue() != null) 
   {
    String header = cxt.getValue().toString();
    if (header.equals("Sales Total")) 
     {
        return new CellFormat(null, "background-color:#C0C0C0", 
                              "font-weight:bold");
      }
    }
    return null;
  }

26.8.4 Creating Stoplight and Conditional Formatting in a Pivot Table

Stoplight and conditional formatting of the cells in a pivot table are examples of customizing the cell content. For this kind of customization, an application might prompt a user for a high value and a low value to be associated with the stoplight formatting. Generally three colors are used as follow:

• Values equal to and above the high value are colored green to indicate they have no issues.

• Values above the low value but below the high value are colored yellow to warn that they are below the high standard.

• Values at or below the low value are colored red to indicate that they fall below the minimum acceptable level.

Figure 26-7 shows data cells with stoplight formatting for minimum, acceptable, and below standards sales for States.

Example 26-5 shows code that performs stoplight formatting in a pivot table that does not display totals. If you want to do stoplight formatting for a pivot table that displays totals, then you might want to combine the code from Example 26-4 (which addresses rows with totals) with the code for stoplight and conditional formatting.

public CellFormat getDataFormat(DataCellContext cxt)
{
   //Use low and high values provided by the application.
  double low = m_rangeValues.getMinimum().doubleValue() * 100;
  double high = m_rangeValues.getMaximum().doubleValue() * 100;

  CellFormat cellFormat = new CellFormat(null, null, null);

 // Create stoplight format
 if (isStoplightingEnabled()) 
 {
   String color = null;
   Object value = cxt.getValue();
   if (value != null && value instanceof Number) 
   {
       double dVal = ((Number)value).doubleValue();
       if (dVal < low) 
       {
         color = "background-color:" + ColorUtils.colorToHTML(m_belowColor) + ";";
       }
       else if (dVal > low && dVal < high) 
       {
         color = "background-color:" + ColorUtils.colorToHTML(m_goodColor) + ";";
       }
       else if (dVal > high) 
       {
         color = "background-color:" + ColorUtils.colorToHTML(m_aboveColor) + ";";
       }
    }
    cellFormat.setStyle(color);
  }
    return cellFormat;
}

26.9.1 Specifying Custom Images for Data Cells

With data cell stamping you can use the dvt:dataCell tag to specify a custom image for a data cell using af:image, af:icon, or af:commandImageLink as a child tag. Example 26-7 shows sample code for using an af:commandImageLink as a custom image in a pivot table data cell.

<dvt:pivotTable var="cellData" varStatus="cellStatus">
  
  <!-- This is the default data cell that will be used for all data attributes-->
  <dvt:dataCell>
    <af:commandImageLink text="Go"
                     icon="/images/go.gif"
                     actionListener="#{pivotTableBean.imageLinkClick}"/>    
    <af:outputText  value="#{cellData.dataValue}"   />
  </dvt:dataCell>
</dvt:pivotTable>

Actions associated with the image are handled through a registered listener, actionListener. In a bean class you specify the method to be called when the image link is clicked, for example:

public void imageLinkClick (javax.faces.event.ActionEvent.action)