Working with PivotTables

Article
07/30/2014

Applies to: Excel 2010 | Office 2010 | PowerPoint 2010 | Word 2010

This topic discusses the Open XML SDK 2.0pivotTableDefinition class and how it relates to the Open XML File Format SpreadsheetML schema. For more information about the overall structure of the parts and elements that make up a SpreadsheetML document, see Structure of a SpreadsheetML Document.

PivotTable in SpreadsheetML

The following information from the ISO/IEC 29500 specification introduces the PivotTableDefinition (<pivotTableDefinition>) element.

PivotTables display aggregated views of data easily and in an understandable layout. Hundreds or thousands of pieces of underlying information can be aggregated on row & column axes, revealing the meanings behind the data. PivotTable reports are used to organize and summarize your data in different ways. Creating a PivotTable report is about moving pieces of information around to see how they fit together. In a few gestures the pivot rows and columns can be moved into different arrangements and layouts.

A PivotTable object has a row axis area, a column axis area, a values area, and a report filter area. Additionally, PivotTables have a corresponding field list pane displaying all the fields of data which can be placed on one of the PivotTable areas.

The workbook points to (and owns the longevity of) the pivotCacheDefinition part, which in turn points to and owns the pivotCacheRecords part. The workbook also points to and owns the sheet part, which in turn points to and owns a pivotTable part definition, when a PivotTable is on the sheet (there can be multiple PivotTables on a sheet). The pivotTable part points to the appropriate pivotCacheDefinition which it is using. Since multiple PivotTables can use the same cache, the pivotTable part does not own the longevity of the pivotCacheDefinition.

The pivotTable part describes the particulars of the layout of the PivotTable on the sheet. It indicates what fields are on the row axis, the column axis, report filter, and values areas of the PivotTable. It also indicates formatting information about the PivotTable. If conditional formatting has been applied to the PivotTable, that is also expressed in the pivotTable part.

The pivot cache definition contains the definitions of all fields in the PivotTable. If you create a PivotTable based on a regular table, each column in the table becomes a field of the pivot cache definition. The pivot cache contains the field definitions and information about the type of content found in that field. It also maintains a reference to the source data in the cache markup so that the pivot cache can be refreshed along with the cached data in the pivot cache records part.

The data that is displayed in the PivotTable is stored in two locations. The pivot cache records part maintains the actual data for the PivotTable. The table cells in the worksheet also store a cached version of the data, but that is only for display purposes. The pivot cache records part stores data in one of two ways. The unique values for the data area of the PivotTable are cached inline. The repeating items that you normally find on the row and column are referenced. This shared data is actually stored in the pivot cache definition. Each record in the pivot cache record part consists of N values where N is equal to the number of fields defined in the pivot cache definition.

The final step is to create the PivotTable itself. The PivotTable definition part contains the information about which field is present in which place of the PivotTable. You can place a field in four areas: row, column, data or filter. The fields are chosen from the cached fields in the pivot cache definition.

To create a PivotTable that is ready to use when the workbook is opened you also need to create the markup for the table cells. The PivotTable is displayed in the cells of a worksheet and therefore you need to construct them as well. You can also have the user update the PivotTable cells when opening the document.

The following table lists the common Open XML SDK 2.0 classes used when working with the PivotTableDefinition class.

SpreadsheetML Element	Open XML SDK 2.0 Class
pivotField	PivotField
pivotCacheDefinition	PivotCacheDefinition
pivotCacheRecords	PivotCacheRecords

Open XML SDK 2.0 PivotTableDefinition Class

The Open XML SDK 2.0PivotTableDefinition class represents the PivotTable definition (<pivotTableDefinition>) element defined in the Open XML File Format schema for SpreadsheetML documents. Use the PivotTableDefinition class to manipulate individual <pivotTableDefinition> elements in a SpreadsheetML document.

The main function of the PivotTable definition is to store information about which field is displayed on which axis of the PivotTable and in what order. There are many other settings that can be added to the PivotTable definition, but the following explains the basics.

The root element names the PivotTable so that it can be used as a data source. The root element also references the pivot cache by using the ID added to the workbook part, and defines the caption label to display above the data area of the PivotTable. All of these elements are required.

The three main pieces of the pivotTableDefinition are: the location of the table, the display information for the cached fields, and the positioning information of the cached fields. For more information about these and other additional elements that make up the pivotTableDefinition, see the ISO/IEC 29500 specification.

PivotField Class

The PivotTableDefinition element contains the PivotField (<pivotField>) elements. The following information from the ISO/IEC 29500 specification introduces the PivotField (<pivotField>) element.

Represents a single field in the PivotTable. This element contains information about the field, including the collection of items in the field.

First, define the collection of fields that appear on the PivotTable using the pivotFields element. Each field serves as a cache for the data of that field in the data source. You do not need to define the cache. Instead, you can set the item element equal to default and have the user update the table when they open the document. The showAll attribute is used to hide certain elements for that data dimension. After defining which fields are part of the table, the fields are applied to the four areas of the PivotTable.

Pivot Cache Definition Class

The following information from the ISO/IEC 29500 specification introduces the PivotCacheDefinition (<pivotCacheDefinition>) element.

The pivotCacheDefinition part defines each field in the pivotCacheRecords part, including field name and information about the data contained in the field. The pivotCacheDefinition part also defines pivot items that are shared among the pivotTable and pivotRecords parts.

The pivot cache defines the source of the data in the PivotTable, which allows it to be updated, and it defines the list of fields in that data. Be aware that the cache defines all the fields available to the PivotTable, not the ones actually used. The PivotTable definition defines which of the available fields are used by a particular PivotTable.

The data source definition references the data that is displayed in the PivotTable. The PivotTable also maintains the data in the cache-records part to allow the table to be updatable when the data connection is not available. You cannot rely on the cells of the PivotTable to store the data because the data in these cells is transient in nature, it changes when you pivot the table. There are various types of data sources, for example: worksheets, database, OLAP cube, and other PivotTables.

The last part of the cache definition defines the fields of the data source using the cacheField element. The cacheField element is used for two purposes: it defines the data type and formatting of the field, and it is used as a cache for shared strings. The pivot values are stored in the pivot cache records part. When a recurring string is used as a value, the cache record uses a reference into the cacheField collection of shared items.

Pivot Cache Records Class

The following information from the ISO/IEC 29500 specification introduces the PivotCacheRecords (<pivotCacheRecords>) element.

The pivotCacheRecords part contains the underlying data to be aggregated. It is a cache of the source data.

The cache records part can store any number of cached records. Each record has the same number of values defined as there are fields in the cache definition. Each record is defined with the r element. This record contains value items as child elements. You can provide certain typed values, such as Numeric, Boolean, or Date-Time, or you can reference into the shared items.