Primary Key:
Primary Key enforces uniqueness of the column on which they are defined. Primary Key creates a clustered index on the column. Primary Key does not allow Nulls.
Create table with Primary Key:
CREATE TABLE Authors (
AuthorID INT NOT NULL PRIMARY KEY,
Name VARCHAR(100) NOT NULL
)
GO
Alter table with Primary Key:
ALTER TABLE Authors
ADD CONSTRAINT pk_authors PRIMARY KEY (AuthorID)
GO
Unique Key:
Unique Key enforces uniqueness of the column on which they are defined. Unique Key creates a non-clustered index on the column. Unique Key allows only one NULL Value.
Alter table to add unique constraint to column:
ALTER TABLE Authors ADD CONSTRAINT IX_Authors_Name UNIQUE(Name)
GO
Reference : Pinal Dave (http://blog.SQLAuthority.com), BOL
pinaldave
Can u please tell me, what is the maximum number of unique key constraints that can be defined in a table.
Also I have a query regarding joins in SQL-Server 2005.
In a single select query, what is a number of max joins that we can have?
Read about Maximum Capacity Specifications in SQL Server help file
Hi Pinal,
In your Unique Key definition section you wrote, “Unique Key does not allow Nulls.” But i think Unique Keys allows one null value in the column on which they are defined.
primary key is unique identifiying rows in s trable Ok…
i agree with that but whats the problem with Primary Key means if you are not difine as a NOT NULL it will also taking null values as like unique key .
So you should difine Not Null while difinig key as a Primary key.
First check it out Reply me don’t forget oK……….
According to My Knowledge
Primary Key not Allow Null Values , But UNique kay may be null
to
respected sir,
i have created ta table and now i want to add a constraint
primary key in SQL SERVER 2005
but its generating error
code:
1) create table sample(sno varchar(20))
2) ALTER TABLE sample
ADD CONSTRAINT pk_sno PRIMARY KEY (sno)
error details :
Msg 8111, Level 16, State 1, Line 1
Cannot define PRIMARY KEY constraint on nullable column in table ’sample’.
Msg 1750, Level 16, State 0, Line 1
Could not create constraint. See previous errors.
KINDLY SEND ME A SOLUTION
hi ankhenatraj
Please refer the dataschema of your database and check if your column ’sno’ of sample table has set to ‘NULL’. If so change that to ‘NOT NULL’ as primary key cannot be defined for a nullable column.
Or else you can generate a unique constartint for that column ’sno’ as unique constratints can be defined for the nullable columns.
hi
Can u please tell me, I have two questions
1.what is difference between Primary key & Unique Key Constraints?
2.Which Constraints act as a primary key like not null and unique key Constraints?
Hi,
I want to drop the primary key on one table but i cannot know which constraint is there. Is there a way to drop the primary key without specifying constraint.
The basic idea of doing this is :
I have one table with 4 columns e.g.
1. SrNo
2. NodeID
3. EnrollmentNo
4. FingerNo
Now the two columns already have Primary key NodeID and EnrollmentNo now i want to add another primary key FingerNo as well.
Thanks,
Prashant Hirapara
@ Prashant
I am not sure if you want to use GUI to do the same. Running queries is always good… but this can be done very easily using Enterprise Manager (SQL Server 2000 ) or SSMS ( SQL Server 2005)
1. SQL Server 2000
open Enterprise manager -> expand server -> expand database -> expand tables -> on the right side-> right click table name -> design table ->
now you can see all the columns in that tables also you can see some primary key on one or more than one column, if you want to add one more column in the primary key, just do the following,
1. First remove allow null check on that column.
2. Press the control button and select all the column you want to be in your primary ( include previous primary key columns as well)
3. Now right click the select columns, select Set Primary key.
4. On the top, save the table.
Just to check again, refresh database open table – design and see if you can see the primary key includes your new column.
SQL Server 2005.
Expand Server- Database – tables – keys – right click key name and click modify, do the same process you did in SQL Server 2000, first you will uncheck allow null on that column, and then you will press control button and you will select all the columns you want to have in primary keys and then right click on those selected column, select set primary key.
at the top, click save and check once again if the changes has taken place.
One more thing, if you are using Varchar(max), text, ntext,nvarchar(max), image datatypes, then I dont think you will be able to set primary key, if you have any varchar(max) columns then try to give a fix values to those columns, like varchar(100).
Now your question remains the same, which one is primary key, what is the name of the primary key,
IN SQL Server 2005, you can easily detect what is your primary key( under database – tables- keys) , because the prange color of primary key is different from foriegn keys which is grey color, and there can be only one primary key on one table.
In SQL Server 2000 it is difficult to see as primary key and a foreign key.
Try running these script, which will give you all primary key on all tables in the database including column information.
SELECT A.TABLE_NAME, A.CONSTRAINT_NAME, B.COLUMN_NAME
FROM INFORMATION_SCHEMA.TABLE_CONSTRAINTS A, INFORMATION_SCHEMA.CONSTRAINT_COLUMN_USAGE B
WHERE CONSTRAINT_TYPE = ‘PRIMARY KEY’ AND A.CONSTRAINT_NAME = B.CONSTRAINT_NAME
ORDER BY A.TABLE_NAME
Sometimes you will see the primary constraint name repeats, meaning that primary key is a composite key.
So now you have Primary Key constraint names and I told you procedure , you can do like that or what ever procedure you were trying to do earlier, I guess you were trying to drop primary key and then create a new primary key on the table. You can do this as well.
Make sure no users are connected to that database, because when you create a primary key, by default a clustered index is created, clustered actually physically stores the data, meaning the tables on which you are trying to create primary key will be un available for users.
Hope this helps.
Imran.
Hi Imran,
Thanks for your prompt reply and i am sorry i have not mentioned that i am using SQL 2000 and i want to execute my things through query.
Now your above query gives me the list of all the primary key constraint along with the column in which primary key is used. but it does not allow me to delete any of the key as it says that the information is view and cannot be update or delete.
At last i have come to this query which is not working
:-
———————————————————————–
Alter Table tblFPTemplates Drop Constraint
(
Select Name
From sysobjects
Where xtype=’PK’ And parent_obj = (Select id From sysobjects Where xtype=’U’ And Name=’tblFPTemplates’))
———————————————————————–
Can u please modify the above query to work it out for me. Again thanks a lot for your input.
Thanks and Regards,
Prashant Hirapara
[...] Question from Prasant: [...]
Hi,
Can u please tell me,
PRIMARY KEY is equivalent to UNIQUE KEY + NOT NULL ????
Can u please tell me how to update table with primary key on diffrent column and unique key on diffrent column? is that based on primary or unique key?
[...] Both primary key and unique key enforces uniqueness of the column on which they are defined. But by default primary key creates a clustered index on the column, where are unique creates a nonclustered index by default. Another major difference is that, primary key doesn’t allow NULLs, but unique key allows one NULL only. (Read More Here) [...]
Create table with Primary Key:
CREATE TABLE Authors (
AuthorID INT NOT NULL PRIMARY KEY,
Name VARCHAR(100) NOT NULL
)
GO
Alter table with Primary Key:
ALTER TABLE Authors
ADD CONSTRAINT pk_authors PRIMARY KEY (AuthorID)
GO
Kallol
Hi,
In Which case we can use Unique Key.Please give one example so i can understand.
Regards
Prashant
Prashant – to drop a PK from a table, when You don’t know a PK’s name, You need to run something like:
Declare @pkName nvarchar(500)
Declare @sql nvarchar(500)
select @pkName=name
from sysobjects
where xtype = ‘PK’ and parent_obj = object_id(‘YourTableName’)
IF NOT @pkName IS NULL
BEGIN
select @sql = ‘ALTER TABLE YourTableName DROP CONSTRAINT ‘ + @pkName
execute sp_executesql @sql
END
Thank U Very much pinal dave
Dear Pinal,
I like your site very much. Few concern about pk and unique key
1) how many columns could be included in unique key(i mean whether composite unique key is possible or not)
2) Could we edit values of the unique key/pk (keeping them unique)
3) how to use composite pk or fk as foreign key
4) which one is better choice composite pk or composite unique key
5) What if wanna modify composite pk column value
I could try in db to get my ans but wanna confirm frm you.
@spider
1) Yes, composite UNIQUE keys are allowed.
2) Yes, because you don’t edit the key, you edit the values. Then the UNIQUE CONSTRAINT will make sure the data is still unique. If it is, it will allow the edit.
3) Like any other FK. FOREIGN KEY(col1, col2) REFERENCES other_tab
If the parent is a UNIQUE key instead of the PRIMARY KEY
FOREIGN KEY(col1, col2) REFERENCES other_tab(col1, col2)
4) That depends on the situation. Assuming i understand the question, PK would be better.
5) See question 2.
We can create composite like
CREATE TABLE Authors (
AuthorID INT NOT NULL,
Name VARCHAR(100) NOT NULL,
PRIMARY KEY(AuthorID,Name)
)
GO
am I right??
Thanks 4 the quick reply :-)) you saved my effort.Thats the way to get more hits ,love and respect from the users :-))
Keep it up…
What is reason for having only one NULL IN unique key constraint?
Hi Pinal.
I am looking for the sql statement, by which executing, i can come to know which column(s) is having unique constraint and which is/are not having.
Hello Nilesh,
Please visit this blog:
http://blog.sqlauthority.com/2007/09/16/sql-server-2005-list-all-the-constraint-of-database-find-primary-key-and-foreign-key-constraint-in-database/
Regards,
Pinal Dave
Hi Pinal,
Thanx for the reply. It solved my problem…..
It is really very useful me…
Thanx again….
Pinal, I am fond of learning SQL deeply. And I have heard lots of about your knowledger and you. So I would love to attend your seminars. So can you plz tell me how can I get the information about your upcoming seminars?
Regards,
Nilesh Soni
Hellow Pinal,
You solved my last problem. But I’ve come again with another problem.
I need to change the data type of column ‘Science’ in table ‘tblMarks’.
I want to change the data type from int to varchar(100) and want to set the default value ‘abc’.
I’ve tried too many times with different sql statements and have checked lots of websites. But didn’t succeeded.
Here I have written a sql statement which is producing error.
ALTER TABLE [dbo].[tblMarks] Alter Column Science varchar(100) Not Null DEFAULT ‘abc’
Before changing the data type the column ‘Science’ does not have any default constraint created on it.
Plz help me.
Regards,
Nilesh Soni
@Nilesh
The easiest way to do this is to DROP the existing COLUMN and ADD a new one.
@Brian Tkatch
Thanx for reply..
But what about the data inserted in to table if I’ll drop the column….
@Nilesh
1) ADD a new COLUMN of the correct type.
ALTER TABLE [dbo].[tblMarks] ADD Science2 varchar(100) Not Null DEFAULT ‘abc’;
2) UPDATE [dbo].[tblMarks] SET Science2 = Science;
3) ALTER TABLE [dbo].[tblMarks] DROP Column Science;
4) 3 options.
4a) sp_rename
4b) Open tblMarks in Design (right-click) and change the name.
4c) ADD Science as the new COLUMN, UPDATE it to equal Science2, and DROP Science2.
Hello Nilesh,
What is the error produced by ALTER TABLE statement?
Try to perform alter column length using SSMS interface when table is not being used.
Regards,
Pinal Dave
Hi Pinal,
I’m getting below error while executing the alter statement “ALTER TABLE [dbo].[tblMarks] Alter Column Science varchar(100) Not Null DEFAULT ‘abc’”.
Msg 156, Level 15, State 1, Line 1
Incorrect syntax near the keyword ‘DEFAULT’.
and i’m unaware about SSMS interface.
Regards,
Nilesh Soni
Hello Nilesh,
Break the statement in two as below:
ALTER TABLE [dbo].[tblMarks] Alter Column Science varchar(100) Not Null
ALTER TABLE [dbo].[tblMarks] ADD CONSTRAINT Sc_Default DEFAULT ‘abc’ FOR Science
Regards,
Pinal Dave
Hi Brian Tkatch.
Thanx for reply.
But Brian my purpose is not for just a single time execution. I’m developing a tool which used to create the xml file at admin side or our side. the file will contains the information about changes in the database and the same tool will read the xml at the client side and will make the changes by creating and executing the sql statements. so the way you suggested is little bit long to altering the table. and i’m sure there is a way to alter the table with default value and not null attribute.
Thanx once again Brian…
Regards,
Nilesh Soni
Oh, i see. I didn’t realize this was for a script. I was trying to provide a simple solution for your scenario.
Pinal’s answer looks best. I think i learned something too.
hi
i created a table with columns such as p_id(unique),name(unique),address,city.
now i want to drop the constraint for name column.
whats the query to done the same
HAI….
I HAVE ANOTHER PROBLEM….
I CREATED A TABLE AS
CREATE TABLE PERSONS(PID INT, LASTNAME VARCHAR(8),FIRSTNAME VARCHAR(8))
NOW I WANT TO ADD PRIMARY KEY FOR PID COLUMN…
FOR THIS I WROTE A QUERY AS
ALTER TABLE PERSON
ADD PRIMARY KEY (PID)
I AM WAITING FOR UR ANSWER
WHILE I EXECUTE THIS ,I FACE SOME ERROR. I MENTION IT BELOW
Msg 8111, Level 16, State 1, Line 1
Cannot define PRIMARY KEY constraint on nullable column in table ‘primaryalter’.
Msg 1750, Level 16, State 0, Line 1
Could not create constraint. See previous errors.
PLS SOLVE MY PROBLEM..
hi,
I want Add Unique key constraint on my existing table which already have data.
I run this query..
ALTER TABLE tbl_Standard ADD CONSTRAINT Unique_Standard_Name UNIQUE(Standard_Name)
I think error is due to this names
Ex
in first row ‘ASTM C 295 – 98′
in 2nd row ‘ASTM C 40 – 98′
these both values are different but I am receiving error.. plz help me..
Error
“CREATE UNIQUE INDEX terminated because a duplicate key was found for index ID 5. Most significant primary key is ‘ASTM C’ “
Hello Rizwan,
its a bug identified in SQL Server 2000. For more details please check the below page:
http://support.microsoft.com/kb/308811
Regards,
Pinal Dave
Isn’t the other *BIG* difference between PRIMARY KEY & UNIQUE INDEX the fact hat UNIQUE INDEX allows you to add columns using the INCLUDE keyword, and PRIMARY KEY will not.
@Antek
To do that, you would have to compare PRIMARY KEY to UNIQUE KEY. Being they both CREATE their own INDEXes, they both have the same restriction.
Nope.
Primary key is basically unique index. By default it is also clustered index, which defines how data is ordered by default, but it can be non-clustered also and then it behaves as normal unique index.
Include can be used with non-clustered indexes and index doesn’t even have to be unique.
hi
iam using select query in my sp to reterieve the values from the temp table. In my sp iamgetting the values from the temp table by calling another sp by passing the parameter and then i am getting the values through select query using 2 right joins and 3 left outer join with where condition and with group by.
but this select statement is taking more than 15 minutes to execute. i have already created index for the temp table. canu plz give me any solution
Hello Aswathaman,
This is not sufficient to identify the issue and recommond a solution. You can get the correct solution by checking the Execution Plan of query. In execution plan, Idetify the most resource consuming tasks and apply a fix.
Regards,
Pinal Dave