Close

Automation-regel in Jira wanneer pull request wordt samengevoegd

Posted by: AJ Welch

Changing the data structure of a column in SQL Server from NULL to NOT NULL, thereby disallowing non-null values in that column, is generally performed using the relatively simple ALTER TABLE syntax to appropriately change the column in question.

In this tutorial we’ll examine the important safety precautions necessary when altering existing data in a column, prior to actually issuing any ALTER commands that would potentially cause harm to the table itself.


Verder gaan dan agile


Before any changes are made to your table, it’s important to briefly go over what data can (and cannot) be specified within an existing column that you wish to alter to NOT NULL, ensuring that no row is allowed to have a NULL value in that column.

Most critically, all existing NULL values within the column must be updated to a non-null value before the ALTER command can be successfully used and the column made NOT NULL. Any attempt to set the column to NOT NULL while actual NULL data remains in the column will result in an error and no change will occur.

Unnullifying existing column data

To ensure that there are no NULL values in our column, we’ll use a basic UPDATE command, applicable explicitly to rows where the value is currently NULL. For example, we have a basic table of client data with nameemail, and phone. Currently a few of the records have a NULL phone value, which we don’t want to allow:

clientsID     name              email                                         phone
1             Neville Estes     Quisque@diamProin.com                         1-843-863-2697
2             Flynn Fry         velit@senectus.net
3             Wyatt Schmidt     nibh.dolor@sit.co.uk                          1-950-895-1847
4             Oleg Hill         lacinia.vitae.sodales@acrisusMorbi.edu        1-173-344-1578
5             Randall Bullock   eu@lacus.net
6             Lamar White       ut.sem@risus.com                              1-421-757-4907
7             Fuller Hill       Vivamus.nisi@tempor.ca                        1-178-437-8281
8             Ulysses Boyle     sem.mollis.dui@Integeraliquamadipiscing.net   1-535-515-1494
9             Paki Palmer       nec@euismod.org
10            Kamal Buchanan    sapien.gravida@tellusnonmagna.co.uk           1-325-847-4838

Therefore, we can insert a default value for all the phone values that are currently NULL with the following statement:

UPDATE
  clients
SET
  phone = '0-000-000-0000'
WHERE
  phone IS NULL;

Now our NULL values have all been replaced with the value we consider the default0-000-000-0000:

clientsID   name            email                                       phone
1           Neville Estes   Quisque@diamProin.com                       1-843-863-2697
2           Flynn Fry       velit@senectus.net                          0-000-000-0000
3           Wyatt Schmidt   nibh.dolor@sit.co.uk                        1-950-895-1847
4           Oleg Hill       lacinia.vitae.sodales@acrisusMorbi.edu      1-173-344-1578
5           Randall Bullock eu@lacus.net                                0-000-000-0000
6           Lamar White     ut.sem@risus.com                            1-421-757-4907
7           Fuller Hill     Vivamus.nisi@tempor.ca                      1-178-437-8281
8           Ulysses Boyle   sem.mollis.dui@Integeraliquamadipiscing.net 1-535-515-1494
9           Paki Palmer     nec@euismod.org                             0-000-000-0000
10          Kamal Buchanan  sapien.gravida@tellusnonmagna.co.uk         1-325-847-4838

Verder gaan dan agile


Now that there are no NULL values any longer, we can issue our ALTER statement to update the column so all future additions do not allow NULL values. Since we’re altering the phone column in this example, the statement will look something like this:

ALTER TABLE
  clients
ALTER COLUMN
  phone
    NVARCHAR(20) NOT NULL;

Verify altered nullability

Once the alteration to your column is made, it is a good practice to verify the column no longer allows any NULL values by running a simple INSERT test and trying to insert a new record with the NULL value in the altered column:

INSERT INTO
  clients(name, email, phone)
VALUES
  ('John Doe', 'jdoe@domain.com', NULL);

If all went according to plan, SQL Server will issue an error stating that the column doesn’t allow NULL values:

Cannot insert the value NULL into column 'phone', table 'library.dbo.clients'; column does not allow nulls. INSERT fails. [SQL State=23000, DB Errorcode=515]

Volgend onderwerp