How to drop a unique constraint from table column?

问题

I have a table 'users' with 'login' column defined as:

[login] VARCHAR(50) UNIQUE NOT NULL

Now I want to remove this unique constraint/index using SQL script. I found its name UQ_users_7D78A4E7 in my local database but I suppose it has a different name on another database.

What is the best way to drop this unique constraint? Or at least any...

Thanks.

回答1:

SKINDER, your code does not use column name. Correct script is:

declare @table_name nvarchar(256)  
declare @col_name nvarchar(256)  
declare @Command  nvarchar(1000)  

set @table_name = N'users'
set @col_name = N'login'

select @Command = 'ALTER TABLE ' + @table_name + ' drop constraint ' + d.name
    from sys.tables t 
    join sys.indexes d on d.object_id = t.object_id  and d.type=2 and d.is_unique=1
    join sys.index_columns ic on d.index_id=ic.index_id and ic.object_id=t.object_id
    join sys.columns c on ic.column_id = c.column_id  and c.object_id=t.object_id
    where t.name = @table_name and c.name=@col_name

print @Command

--execute (@Command)

回答2:

ALTER TABLE users
DROP CONSTRAINT 'constraints_name'

回答3:

This works mostly.

drop index IX_dbo_YourTableName__YourColumnName on dbo.YourTableName
GO

回答4:

To drop a UNIQUE constraint, you don’t need the name of the constraint, just the list of columns that are included in the constraint.

The syntax would be:

ALTER TABLE table_name DROP UNIQUE (column1, column2, . . . )

回答5:

Use this SQL command to drop a unique constraint:

ALTER TABLE tbl_name
DROP INDEX column_name

回答6:

You can use following script :

Declare @Cons_Name NVARCHAR(100)
Declare @Str NVARCHAR(500)

SELECT @Cons_Name=name
FROM sys.objects
WHERE type='UQ' AND OBJECT_NAME(parent_object_id) = N'TableName';

---- Delete the unique constraint.
SET @Str='ALTER TABLE TableName DROP CONSTRAINT ' + @Cons_Name;
Exec (@Str)
GO

回答7:

This statement works for me

  ALTER TABLE table_name DROP UNIQUE (column_name);

回答8:

I have stopped on the script like below (as I have only one non-clustered unique index in this table):

declare @table_name nvarchar(256)  
declare @col_name nvarchar(256)  
declare @Command  nvarchar(1000)  

set @table_name = N'users'
set @col_name = N'login'

select @Command = 'ALTER TABLE ' + @table_name + ' drop constraint ' + d.name
    from sys.tables t join sys.indexes d on d.object_id = t.object_id  
    where t.name = @table_name and d.type=2 and d.is_unique=1

--print @Command

execute (@Command)

Has anyone comments if this solution is acceptable? Any pros and cons?

Thanks.

回答9:

Expand to database name >> expand to table >> expand to keys >> copy the name of key then execute the below command:

ALTER TABLE Test DROP UQ__test__3213E83EB607700F;

Here UQ__test__3213E83EB607700F is the name of unique key which was created on a particular column on test table.

回答10:

I would like to refer a previous question, Because I have faced same problem and solved by this solution. First of all a constraint is always built with a Hash value in it's name. So problem is this HASH is varies in different Machine or Database. For example DF__Companies__IsGlo__6AB17FE4 here 6AB17FE4 is the hash value(8 bit). So I am referring a single script which will be fruitful to all

DECLARE @Command NVARCHAR(MAX)
     declare @table_name nvarchar(256)
     declare @col_name nvarchar(256)
     set @table_name = N'ProcedureAlerts'
     set @col_name = N'EmailSent'

     select @Command ='Alter Table dbo.ProcedureAlerts Drop Constraint [' + ( select d.name
     from 
         sys.tables t
         join sys.default_constraints d on d.parent_object_id = t.object_id
         join sys.columns c on c.object_id = t.object_id
                               and c.column_id = d.parent_column_id
     where 
         t.name = @table_name
         and c.name = @col_name) + ']'

    --print @Command
    exec sp_executesql @Command

It will drop your default constraint. However if you want to create it again you can simply try this

ALTER TABLE [dbo].[ProcedureAlerts] ADD DEFAULT((0)) FOR [EmailSent]

Finally, just simply run a DROP command to drop the column.

回答11:

I had the same problem. I'm using DB2. What I have done is a bit not too professional solution, but it works in every DBMS:

Add a column with the same definition without the unique contraint.
Copy the values from the original column to the new
Drop the original column (so DBMS will remove the constraint as well no matter what its name was)
And finally rename the new one to the original
And a reorg at the end (only in DB2)

ALTER TABLE USERS ADD COLUMN LOGIN_OLD VARCHAR(50) NOT NULL DEFAULT '';
UPDATE USERS SET LOGIN_OLD=LOGIN;
ALTER TABLE USERS DROP COLUMN LOGIN;
ALTER TABLE USERS RENAME COLUMN LOGIN_OLD TO LOGIN;

CALL SYSPROC.ADMIN_CMD('REORG TABLE USERS');

The syntax of the ALTER commands may be different in other DBMS

回答12:

If you know the name of your constraint then you can directly use the command like

alter table users drop constraint constraint_name;

If you don't know the constraint name, you can get the constraint by using this command

select constraint_name,constraint_type from user_constraints where table_name = 'YOUR TABLE NAME';

来源：https://stackoverflow.com/questions/5499574/how-to-drop-a-unique-constraint-from-table-column

标签

sql

database

sql-server-2005

unique-constraint

non-clustered-index