BLOGS BY TOPIC
BLOGS BY AUTHOR
BLOGS BY YEAR
Using the Pivot operator in SQL Server allows you to create basic pivot tables from your query results. This blog teaches you how to create simple, static pivots, crosstab-style tables and exciting, dynamic pivot tables.
- Pivoting Data in SQL Server
- Using the Pivot Operator in SQL Server
- Dynamic Pivot Tables (this blog)
Posted by Andrew Gould on 09 June 2014
You need a minimum screen resolution of about 700 pixels width to see our blogs. This is because they contain diagrams and tables which would not be viewable easily on a mobile phone or small laptop. Please use a larger tablet, notebook or desktop computer, or change your screen resolution settings.
Dynamic Pivot Tables
Perhaps the biggest limitation of the PIVOT operator is that you must explicitly list each column name that you want to create. Not only is this is immensely tedious when you have a large number of columns, you may not actually know what the column names will be. The solution is to create a dynamic pivot table using dynamic SQL.
Creating a List of Pivot Column Names
Start by creating a simple query to display all of the values that you want to become column headings in the final pivot table. In our example we'll select all of the country names using this statement:
SELECT CountryName
FROM tblCountry
The result of the query is shown below:
Your query may be more complex than the one shown above: the aim is to create a list of all column names for the final pivot table.
Creating a Comma-Separated String
The next step is to create a comma-separated string of the values from the query results. The code below shows how to do this:
--A variable to hold the complete string
DECLARE @Columns VARCHAR(MAX) = ''
--Concatenate each country with a comma
SELECT @Columns += (CountryName + ',')
FROM tblCountry
--Check the result
PRINT @Columns
The result of executing this code is shown below:
The result is a comma-separated list of country names.
Enclosing Column Names in Square Brackets
Each column name needs to be enclosed in a set of square brackets. We can achieve this by using the QUOTENAME function:
--A variable to hold the complete string
DECLARE @Columns VARCHAR(MAX) = ''
--Concatenate each country with a comma
--The QUOTENAME function adds square brackets around each value
SELECT @Columns += (QUOTENAME(CountryName) + ',')
FROM tblCountry
--Check the result
PRINT @Columns
The new result of executing the code looks like this:
Each country name is enclosed in square brackets.
Removing the Trailing Comma
At this point our list of column names contains an extra comma at the end of the string. One way to remove this is shown below:
--A variable to hold the complete string
DECLARE @Columns VARCHAR(MAX) = ''
--Concatenate each country with a comma
--The QUOTENAME function adds square brackets around each value
SELECT @Columns += (QUOTENAME(CountryName) + ',')
FROM tblCountry
--Remove the trailing comma
SET @Columns = LEFT(@Columns, LEN(@Columns) - 1)
--Check the result
PRINT @Columns
The LEN function calculates the number of characters in the string. We subtract 1 from the result and use the LEFT function to return that number of characters from the left of the string.
Concatenating the SQL Statement
The next step is to generate the complete SQL statement that will create the pivot table and store the result in a string variable.
--A variable to hold the SQL string
DECLARE @SQL NVARCHAR(MAX) = ''
--Store the complete string in the variable
--Concatenate the list of column names
SET @SQL =
'WITH BaseData AS
(
SELECT
CountryName
,FilmID
FROM
tblFilm AS f INNER JOIN
tblCountry AS c
ON c.CountryID=f.FilmCountryID
)
SELECT * FROM BaseData
PIVOT
(
COUNT(FilmID)
FOR CountryName IN (' + @Columns + ')
) AS PivotTable'
--Check the result
PRINT @SQL
Printing the result is a good way to check that you've generated a valid SQL statement:
The complete SQL statement.
Executing the SQL Statement
The final step is to execute the dynamic SQL statement using either of the two techniques shown below:
--Execute the statement directly
EXECUTE (@SQL)
--Use a system stored procedure
EXECUTE sp_executesql @SQL
The final result should look like this:
A simple, dynamic pivot table.
From this point you can easily create more complex pivot tables by adding code to the dynamic SQL statement.
- Pivoting Data in SQL Server
- Using the Pivot Operator in SQL Server
- Dynamic Pivot Tables (this blog)
This is absolutely top-notch. Not only knowledgeable, but explained systematically and with a clear pedagogical flavour. Thank you for posting it!
One small addition would be helpful: how do you deal with a situation where the values to be pivoted into column headers come from a non-distinct column and would need to be made distinct before using them as column/field names?
Thanks! Happy to hear that you found it useful. If you need to generate a list of distinct values to create the pivot column headers you could use a Select Distinct statement. For the example in this article you could say:
SELECT DISTINCT
CountryName
FROM
tblCountry
I hope that helps!