You can create a table-valued function (TVF) in SQL Server using the CREATE FUNCTION T-SQL syntax.
The syntax is slightly different depending on whether you’re creating an inline table-valued function (ITVF) or a multi-statement table-valued function (MSTVF).
You can create a multi-statement table-valued function (MSTVF) in SQL Server using the T-SQL CREATE FUNCTION syntax.
You can create an inline table-valued function (ITVF) in SQL Server using the T-SQL CREATE FUNCTION syntax.
In SQL Server, it’s possible to use a table-valued function (TVF) to delete data from the underlying tables that the TVF queries.
To delete table data via a TVF, simply use the same T-SQL DELETE syntax that you’d use if deleting a row from the table directly.
In SQL Server, it’s possible to update data via a table-valued function.
What I mean is, you can update data in the underlying tables that the function queries.
For example, if your function returns someone’s first name from a table, you can update their first name by running an UPDATE statement against the function instead of the table.
Note that this only works on inline table-valued functions (ITVFs). As far as I’m aware, it won’t work on multi-statement table-valued functions (MSTVFs).
Also, the columns you update will need to be valid columns in the function’s query.
In SQL Server, it’s possible to insert data via a table-valued function (TVF).
By this, I mean insert rows in the underlying tables that the function queries.
To insert a new row via a TVF, simply use the same T-SQL INSERT syntax that you’d use if inserting data directly into the table.
The SELECT statement is probably the most commonly used statement in SQL Server. Most of the time this statement is run against a view or directly against a table to retrieve rows of tabular data.
But views and tables aren’t the only objects you can run a SELECT statement on. The SELECT statement can also be used on other objects such as rowset functions, OPENXML, and user-defined functions.
This article provides an example of selecting data via a table-valued function.
In SQL Server, a table-valued function (TVF) is a user-defined function that returns a table. This is in contrast to a scalar function, which returns a single value.
You can invoke a table-valued function in the same way that you can query a table. For example, you can use it in a SELECT statement. In some cases, table-valued functions can also be used to update, delete, and insert data.
In SQL Server, the @@SERVERNAME configuration function returns the name of the local server that is running SQL Server.
No argument is required. You can simply use it in a SELECT statement to return the server name.
A widely known limitation of computed columns in SQL Server is that they can’t access data from other tables. That is, your expression can use columns in the same table, but not from other tables.
But this is only half-true. While you can’t reference another table’s column directly within your expression, you can invoke a user-defined function. And therefore, you could create a user-defined function that performs the calculation you need, then simply call that function as your computed column’s expression.
Here’s an example to demonstrate.
Continue reading