How do I create a unique constraint that also allows nulls?

What you're looking for is indeed part of the ANSI standards SQL:92, SQL:1999 and SQL:2003, ie a UNIQUE constraint must disallow duplicate non-NULL values but accept multiple NULL values.

In the Microsoft world of SQL Server however, a single NULL is allowed but multiple NULLs are not...

In SQL Server 2008, you can define a unique filtered index based on a predicate that excludes NULLs:

CREATE UNIQUE NONCLUSTERED INDEX idx_yourcolumn_notnull
ON YourTable(yourcolumn)
WHERE yourcolumn IS NOT NULL;

In earlier versions, you can resort to VIEWS with a NOT NULL predicate to enforce the constraint.

---

SQL Server 2008 +

You can create a unique index that accept multiple NULLs with a WHERE clause. See the answer below.

Prior to SQL Server 2008

You cannot create a UNIQUE constraint and allow NULLs. You need set a default value of NEWID().

Update the existing values to NEWID() where NULL before creating the UNIQUE constraint.

---

SQL Server 2008 And Up

Just filter a unique index:

CREATE UNIQUE NONCLUSTERED INDEX UQ_Party_SamAccountName
ON dbo.Party(SamAccountName)
WHERE SamAccountName IS NOT NULL;

In Lower Versions, A Materialized View Is Still Not Required

For SQL Server 2005 and earlier, you can do it without a view. I just added a unique constraint like you're asking for to one of my tables. Given that I want uniqueness in column SamAccountName, but I want to allow multiple NULLs, I used a materialized column rather than a materialized view:

ALTER TABLE dbo.Party ADD SamAccountNameUnique
   AS (Coalesce(SamAccountName, Convert(varchar(11), PartyID)))
ALTER TABLE dbo.Party ADD CONSTRAINT UQ_Party_SamAccountName
   UNIQUE (SamAccountNameUnique)

You simply have to put something in the computed column that will be guaranteed unique across the whole table when the actual desired unique column is NULL. In this case, PartyID is an identity column and being numeric will never match any SamAccountName, so it worked for me. You can try your own method—be sure you understand the domain of your data so that there is no possibility of intersection with real data. That could be as simple as prepending a differentiator character like this:

Coalesce('n' + SamAccountName, 'p' + Convert(varchar(11), PartyID))

Even if PartyID became non-numeric someday and could coincide with a SamAccountName, now it won't matter.

Note that the presence of an index including the computed column implicitly causes each expression result to be saved to disk with the other data in the table, which DOES take additional disk space.

Note that if you don't want an index, you can still save CPU by making the expression be precalculated to disk by adding the keyword PERSISTED to the end of the column expression definition.

In SQL Server 2008 and up, definitely use the filtered solution instead if you possibly can!

Controversy

Please note that some database professionals will see this as a case of "surrogate NULLs", which definitely have problems (mostly due to issues around trying to determine when something is a real value or a surrogate value for missing data; there can also be issues with the number of non-NULL surrogate values multiplying like crazy).

However, I believe this case is different. The computed column I'm adding will never be used to determine anything. It has no meaning of itself, and encodes no information that isn't already found separately in other, properly defined columns. It should never be selected or used.

So, my story is that this is not a surrogate NULL, and I'm sticking to it! Since we don't actually want the non-NULL value for any purpose other than to trick the UNIQUE index to ignore NULLs, our use case has none of the problems that arise with normal surrogate NULL creation.

All that said, I have no problem with using an indexed view instead—but it brings some issues with it such as the requirement of using SCHEMABINDING. Have fun adding a new column to your base table (you'll at minimum have to drop the index, and then drop the view or alter the view to not be schema bound). See the full (long) list of requirements for creating an indexed view in SQL Server (2005) (also later versions), (2000).

Update

If your column is numeric, there may be the challenge of ensuring that the unique constraint using Coalesce does not result in collisions. In that case, there are some options. One might be to use a negative number, to put the "surrogate NULLs" only in the negative range, and the "real values" only in the positive range. Alternately, the following pattern could be used. In table Issue (where IssueID is the PRIMARY KEY), there may or may not be a TicketID, but if there is one, it must be unique.

ALTER TABLE dbo.Issue ADD TicketUnique
   AS (CASE WHEN TicketID IS NULL THEN IssueID END);
ALTER TABLE dbo.Issue ADD CONSTRAINT UQ_Issue_Ticket_AllowNull
   UNIQUE (TicketID, TicketUnique);

If IssueID 1 has ticket 123, the UNIQUE constraint will be on values (123, NULL). If IssueID 2 has no ticket, it will be on (NULL, 2). Some thought will show that this constraint cannot be duplicated for any row in the table, and still allows multiple NULLs.

---

For people who are using Microsoft SQL Server Manager and want to create a Unique but Nullable index you can create your unique index as you normally would then in your Index Properties for your new index, select "Filter" from the left hand panel, then enter your filter (which is your where clause). It should read something like this:

([YourColumnName] IS NOT NULL)

This works with MSSQL 2012

---

When I applied the unique index below:

CREATE UNIQUE NONCLUSTERED INDEX idx_badgeid_notnull
ON employee(badgeid)
WHERE badgeid IS NOT NULL;

every non null update and insert failed with the error below:

UPDATE failed because the following SET options have incorrect settings: 'ARITHABORT'.

I found this on MSDN

SET ARITHABORT must be ON when you are creating or changing indexes on computed columns or indexed views. If SET ARITHABORT is OFF, CREATE, UPDATE, INSERT, and DELETE statements on tables with indexes on computed columns or indexed views will fail.

So to get this to work correctly I did this

Right click [Database]-->Properties-->Options-->Other Options-->Misscellaneous-->Arithmetic Abort Enabled -->true

I believe it is possible to set this option in code using

ALTER DATABASE "DBNAME" SET ARITHABORT ON

but i have not tested this

---

It can be done in the designer as well

Right click on the Index > Properties to get this window

https://i.stack.imgur.com/nCplb.png

---

Create a view that selects only non-NULL columns and create the UNIQUE INDEX on the view:

CREATE VIEW myview
AS
SELECT  *
FROM    mytable
WHERE   mycolumn IS NOT NULL

CREATE UNIQUE INDEX ux_myview_mycolumn ON myview (mycolumn)

Note that you'll need to perform INSERT's and UPDATE's on the view instead of table.

You may do it with an INSTEAD OF trigger:

CREATE TRIGGER trg_mytable_insert ON mytable
INSTEAD OF INSERT
AS
BEGIN
        INSERT
        INTO    myview
        SELECT  *
        FROM    inserted
END

---

It is possible to create a unique constraint on a Clustered Indexed View

You can create the View like this:

CREATE VIEW dbo.VIEW_OfYourTable WITH SCHEMABINDING AS
SELECT YourUniqueColumnWithNullValues FROM dbo.YourTable
WHERE YourUniqueColumnWithNullValues IS NOT NULL;

and the unique constraint like this:

CREATE UNIQUE CLUSTERED INDEX UIX_VIEW_OFYOURTABLE 
  ON dbo.VIEW_OfYourTable(YourUniqueColumnWithNullValues)

---

In my experience - if you're thinking a column needs to allow NULLs but also needs to be UNIQUE for values where they exist, you may be modelling the data incorrectly. This often suggests you're creating a separate sub-entity within the same table as a different entity. It probably makes more sense to have this entity in a second table.

In the provided example, I would put LibraryCardId in a separate LibraryCards table with a unique not-null foreign key to the People table:

CREATE TABLE People (
  Id INT CONSTRAINT PK_MyTable PRIMARY KEY IDENTITY,
  Name NVARCHAR(250) NOT NULL,
)
CREATE TABLE LibraryCards (    
  LibraryCardId UNIQUEIDENTIFIER CONSTRAINT PK_LibraryCards PRIMARY KEY,
  PersonId INT NOT NULL
  CONSTRAINT UQ_LibraryCardId_PersonId UNIQUE (PersonId),
  FOREIGN KEY (PersonId) REFERENCES People(id)
)

This way you don't need to bother with a column being both unique and nullable. If a person doesn't have a library card, they just won't have a record in the library cards table. Also, if there are additional attributes about the library card (perhaps Expiration Date or something), you now have a logical place to put those fields.

---

Maybe consider an "INSTEAD OF" trigger and do the check yourself? With a non-clustered (non-unique) index on the column to enable the lookup.

---

As stated before, SQL Server doesn't implement the ANSI standard when it comes to UNIQUE CONSTRAINT. There is a ticket on Microsoft Connect for this since 2007. As suggested there and here the best options as of today are to use a filtered index as stated in another answer or a computed column, e.g.:

CREATE TABLE [Orders] (
  [OrderId] INT IDENTITY(1,1) NOT NULL,
  [TrackingId] varchar(11) NULL,
  ...
  [ComputedUniqueTrackingId] AS (
      CASE WHEN [TrackingId] IS NULL
      THEN '#' + cast([OrderId] as varchar(12))
      ELSE [TrackingId_Unique] END
  ),
  CONSTRAINT [UQ_TrackingId] UNIQUE ([ComputedUniqueTrackingId])
)

---

You can create an INSTEAD OF trigger to check for specific conditions and error if they are met. Creating an index can be costly on larger tables.

Here's an example:

CREATE TRIGGER PONY.trg_pony_unique_name ON PONY.tbl_pony
 INSTEAD OF INSERT, UPDATE
 AS
BEGIN
 IF EXISTS(
    SELECT TOP (1) 1 
    FROM inserted i
    GROUP BY i.pony_name
    HAVING COUNT(1) > 1     
    ) 
     OR EXISTS(
    SELECT TOP (1) 1 
    FROM PONY.tbl_pony t
    INNER JOIN inserted i
    ON i.pony_name = t.pony_name
    )
    THROW 911911, 'A pony must have a name as unique as s/he is. --PAS', 16;
 ELSE
    INSERT INTO PONY.tbl_pony (pony_name, stable_id, pet_human_id)
    SELECT pony_name, stable_id, pet_human_id
    FROM inserted
 END

---

You can't do this with a UNIQUE constraint, but you can do this in a trigger.

    CREATE TRIGGER [dbo].[OnInsertMyTableTrigger]
   ON  [dbo].[MyTable]
   INSTEAD OF INSERT
AS 
BEGIN
    SET NOCOUNT ON;

    DECLARE @Column1 INT;
    DECLARE @Column2 INT; -- allow nulls on this column

    SELECT @Column1=Column1, @Column2=Column2 FROM inserted;

    -- Check if an existing record already exists, if not allow the insert.
    IF NOT EXISTS(SELECT * FROM dbo.MyTable WHERE Column1=@Column1 AND Column2=@Column2 @Column2 IS NOT NULL)
    BEGIN
        INSERT INTO dbo.MyTable (Column1, Column2)
            SELECT @Column2, @Column2;
    END
    ELSE
    BEGIN
        RAISERROR('The unique constraint applies on Column1 %d, AND Column2 %d, unless Column2 is NULL.', 16, 1, @Column1, @Column2);
        ROLLBACK TRANSACTION;   
    END

END

---

CREATE UNIQUE NONCLUSTERED INDEX [UIX_COLUMN_NAME]
ON [dbo].[Employee]([Username] ASC) WHERE ([Username] IS NOT NULL) 
WITH (ALLOW_PAGE_LOCKS = ON, ALLOW_ROW_LOCKS = ON, PAD_INDEX = OFF, SORT_IN_TEMPDB = OFF, 
DROP_EXISTING = OFF, IGNORE_DUP_KEY = OFF, STATISTICS_NORECOMPUTE = OFF, ONLINE = OFF, 
MAXDOP = 0) ON [PRIMARY];

---

this code if u make a register form with textBox and use insert and ur textBox is empty and u click on submit button .

CREATE UNIQUE NONCLUSTERED INDEX [IX_tableName_Column]
ON [dbo].[tableName]([columnName] ASC) WHERE [columnName] !=`''`;

---