Table of Contents |
---|
This example assumes a basic level of understanding the different parts of a PivotTable and how they work. If you are not familiar with PivotTable terminology, we highly recommend that you read THIS ANSWERS POST before continuing. |
This example takes an existing workbook that contains some data and creates a PivotTable. The workbook used in this example is available for download: Download BasicExample.xlsx.
Before writing any PivotTable code, make sure to open the workbook with ExcelApplication
and get references to the data worksheet and a worksheet for the PivotTable. See Adding OfficeWriter to your .NET Application.
ExcelApplication xla = new ExcelApplication(); //Example.xlsx has a worksheet 'Data' with 9 columns and 244 rows of data Workbook wb = xla.Open("\\BasicExample.xlsx"); Worksheet data_ws = wb.Worksheets["Data"]; Worksheet pivot_ws = wb.Worksheets.CreateWorksheet("Pivot"); |
This example places the PivotTable on a separate sheet. It is recommended that each PivotTable be placed on a separate sheet.
Excel does not allow PivoTables to overlap and will use error messages to prevent users from creating overlapping PivotTables. ExcelWriter does not have the ability to render PivotTables, so it cannot detect if two PivotTables will overlap when they are rendered. To avoid this, we encourage you to keep your PivotTables on separate worksheets.
The data source needs to be a continuous block of cells with a header row with column names. The data source can be defined as an Area
or a NamedRange
.
Here is a snapshot of the data for this tutorial, which can also be found in BasicExample.xlsx:
There are 9 columns and 244 rows in the data set, including the row with the header values.
In this case, the data source for the PivotTable will be a dynamically defined area on the data worksheet. Note that the row of column names is included in the area.
Area data_area = data_ws.CreateArea(0, 0, 9, 244); |
To create a PivotTable
, call CreatePivotTable
on the PivotTables
collection. Specify the 0-indexed row and column values for the PivotTable location:
PivotTable pt = pivot_ws.PivotTables.CreatePivotTable(data_area, 0, 0); |
This example shows a PivotTable being created from a fully populated data set. PivotTables can also be created and modified in files that will be populated by |
In Excel, there are a number of options that can be set by going to PivotTable Options. These properties are available through PivotTableSettings
.
There are a couple properties that you should always consider when working with PivotTables with ExcelWriter.
After creating the PivotTable, always set RefreshOnOpen
to true
.
pt.PivotTableSettings.RefreshOnOpen = true; |
ExcelWriter does not have the ability to render a PivotTable, so any modifications made to a PivotTable will not take affect until the output file is opened in Excel and the PivotTable is refreshed. If RefreshOnOpen
is true
, Excel will refresh the PivotTable when the workbook opens, which will re-render the PivotTable.
The other important property to set is ItemsToRetain
. When the PivotTable is created, the values that are available through the row label, column label, or page field drop-down filters are based on the values in the data source of the PivotTable at the time the PivotTable was created.
By default, the PivotTable will retain all the original values in those filters, even if those values are no longer in the data source. Set ItemsToRetain
to None
to make sure the original values are cleared out when the PivotTable is refreshed.
pt.PivotTableSettings.ItemsToRetain.Value = ItemsToRetain.None; |
This is especially important when working with PivotTables that have data markers in the data source. Set |
There are four types of PivotTableField
: DataFields
, RowLabels
, ColumnLabels
, and PageFields
(also called report filters). PivotTableFields
are created from read-only SourceFields
, which are generated based on the PivotTable
data source.
A |
Get a handle on the SourceFields
to use for building PivotTableFields
later.
SourceField prodSubCategory = pivot.SourceFields[0], prodName = pivot.SourceFields[1], sales = pivot.SourceFields[2], dateStart = pivot.SourceFields[3], dateEnd = pivot.SourceFields[4], prodStdCost = pivot.SourceFields[5], listPrice = pivot.SourceFields[6], discount = pivot.SourceFields[7], unitDiscountPrice = pivot.SourceFields[8]; |
To add a PageField
, call CreateField
on the PivotTable.PageFields
collection. You will need to specify the SourceField
that will be used to create the PageField
.
pivot.PageFields.CreateField(dateStart); pivot.PageFields.CreateField(dateEnd); |
Page fields are always placed two rows above the upper left corner of the PivotTable, so if a PivotTable is placed in cell B10, then page fields will be in B8. Any additional page fields are added to B8 and existing page fields are moved up a row (B7, B6 and so on). If there is no space to add the page fields above the PivotTable (i.e. the PivotTable is located in cell A1), then ExcelWriter will automatically move the PivotTable down to accommodate the page fields. |
Similarly to page fields, RowLabels
and ColumnLabels
are created on the PivotTable.RowLabels
and PivotTable.ColumnLabels
collections. As mentioned earlier, only one RowLabel
, ColumnLabel
, or PageField
can be created from a particular SourceField
.
Insert column label and row label code here |
Excel automatically sorts and re-renders a PivotTable any time a change is made. By default, the row label or column label values are sorted alpha-numerically in ascending order.
Since ExcelWriter does not have the ability to render PivotTables or sort the values for a field, the only way to guarantee that the data will be sorted is to set RefreshOnOpen
to true
and set SortOptions.Ordering
on a PivotTablefield
to be Ascending
or Descending
.
This property only affects row and column labels.
Insert code to sort the row label and column label values here |
When Excel refreshes the PivotTable, it will observe the SortOptions
setting for a particular field.
Adding data fields - calculated fields, naming conventions, changing the display name
To create a data field, call CreateField
on the DataFields
collection. Unlike row labels, column labels, and page fields, multiple data fields can be created from the same source field.
CODE TO CREATE A DATA FIELD |
In Excel, a unique name is given to the data field depending on the type of data in the source field (numerical or mixed), whether or not any other data fields were already created from the same source field, and whether the source field name ends in a number or alphabetical character (e.g. Case1 vs. CaseOne).
ExcelWriter uses a consistent naming convention when creating data fields: all data fields follow the format SOURCEFIELDNAME_#, where # is an incremental number starting at 1. To change the name of a data field, use the DisplayName
property.
CODE FOR CHANGING THE DISPLAY NAME |
In Excel, a calculated field is a type of row or column label that is automatically generated when two or more data fields exist in the PivotTable. Calculated fields cannot be modified except to change whether the calculated field is a row label or a column label. Excel handles this by disabling access to the calculate fields' properties. ExcelWriter will return |
And that concludes how to create a basic PivotTable. Here is the full sample code below:
INSERT COMPLETED CODE HERE |