declare @myconstraint varchar(8000);
select @myconstraint = name
from sys.default_constraints
where object_name(parent_object_id) = ‘mytable’
and col_name (parent_object_id, parent_column_id) = ‘mycolumn’;
if @myconstraint is not null
exec(‘alter table mytable drop constraint ‘+@myconstraint);

I would want to run a query against a database where in I wanted to specify a column/field, which has been referenced in multiple DB tables. And the result of the query should retrieve me information about the field’s reference in a particular DB table and the column name being referenced along with the field description/documentation, if any. I am not a developer or a DBA. But, I just need this information for documentation. Thanks a ton before hand for responding back to my query.

The following query returns all related tables for tbl1.

select
o1.name as Referencing_Object_name
, c1.name as referencing_column_Name
, o2.name as Referenced_Object_name
, c2.name as Referenced_Column_Name
, s.name as Constraint_name
from sysforeignkeys fk
inner join sysobjects o1 on fk.fkeyid = o1.id
inner join sysobjects o2 on fk.rkeyid = o2.id
inner join syscolumns c1 on c1.id = o1.id and c1.colid = fk.fkey
inner join syscolumns c2 on c2.id = o2.id and c2.colid = fk.rkey
inner join sysobjects s on fk.constid = s.id
and o2.name=’tbl1′ and c2.name=’ID’

I also need to get is there any data in the referenced tables.

We can do this by writing a dynamic query by checking count in each table.

Other than this i am looking for a simplest way.

When we delete a record from the main table, and it has some references,we usually get the below error msg.

The DELETE statement conflicted with the REFERENCE constraint “FK_tbl3_tbl1″. The conflict occurred in database ” db1 “, table “dbo.tbl3″, column ‘ID’.
The statement has been terminated.

So definitely SQL Server should have some system procedures, that is checking this dependencies(means existence of records in related tables).

What is that sp..?

Thanks

select * from table as t exists(select * from othertable ot where t.id=ot.id)

Is there any way to get the data in one table is referencing in any other tables

For eg: I have 3 tables say tbl1,tbl2,tbl3.

tbl1 contains 3 records say 1,2,3

But only 1 and 2 are referencing tbl2 and tbl3,

If we delete the record 1 or 2 from tbl1 ,it is not possible.since the data is using in tbl2 or tbl3.

But if delete 3 from tbl1 it is possible.

So my question is, the way to find out whether the data is referencing in some other places without writing the delete query.

Through proc I pass the values 1,2,3 etc.

Thanks.
Manoj

select * from INFORMATION_SCHEMA.CONSTRAINT_COLUMN_USAGE
select * from INFORMATION_SCHEMA.CHECK_CONSTRAINTS

select
‘[' + SCHEMA_NAME(o1.schema_id) + '].[' + o1.name + ']‘ as Referencing_Object_name
, s.name as Constraint_name
, c1.name as referencing_column_Name
, ‘[' + SCHEMA_NAME(o2.schema_id) + '].[' + o2.name + ']‘ as Referenced_Object_name
, c2.name as Referenced_Column_Name
, fk.keyno as orderKey
, ‘[' + SCHEMA_NAME(o1.schema_id) + '].[' + s.name + ']‘ Constraint_name_schema
into #tmp
from sysforeignkeys fk
inner join sys.objects o1 on fk.fkeyid = o1.object_id
inner join sys.objects o2 on fk.rkeyid = o2.object_id
inner join sys.columns c1 on c1.object_id = o1.object_id and c1.column_id = fk.fkey
inner join sys.columns c2 on c2.object_id = o2.object_id and c2.column_id = fk.rkey
inner join sys.objects s on fk.constid = s.object_id
and o2.name=’<>’ — this predicate for a specific table
order by 1, fk.keyno

select distinct
‘IF EXISTS (SELECT * FROM sys.foreign_keys WHERE object_id = OBJECT_ID(N”’+ Constraint_name_schema + ”’) AND parent_object_id = OBJECT_ID(N”’ + Referencing_Object_name + ”’))’ + char(13) + char(10) + char(9) +
‘ALTER TABLE ‘ +
Referencing_Object_name +
‘ DROP CONSTRAINT ‘ +
Constraint_name
from #tmp o1

select distinct
‘ALTER TABLE ‘ +
Referencing_Object_name +
‘ WITH NOCHECK ADD CONSTRAINT ‘ +
Constraint_name +
‘ FOREIGN KEY ‘ +
‘(‘ +
STUFF(( SELECT
‘],[' + c1.referencing_column_Name
FROM #tmp c1
where c1.Referencing_Object_name = o1.Referencing_Object_name
group by c1.referencing_column_Name,orderKey
order by orderKey
FOR XML PATH('')
), 1, 2, '') + ']‘
+ ‘)’ + char(13) + char(10) + char(9) +
‘ REFERENCES ‘ +
Referenced_Object_name +
‘(‘ +
STUFF(( SELECT
‘],[' + c1.Referenced_Column_Name
FROM #tmp c1
where c1.Referenced_Object_name = o1.Referenced_Object_name
group by c1.Referenced_Column_Name,orderKey
order by orderKey
FOR XML PATH('')
), 1, 2, '') + ']‘
+ ‘)’
from #tmp o1

ERROR_NUMBER() – catches error no,
ERROR_MESSAGE() – catches error msg,
ERROR_PROCEDURE() – catches procedure name ,
ERROR_LINE() – catches line no

How do i catch the constraint name if the error is about the violation of primary key?

eg If I get an error Violation of PRIMARY KEY constraint ‘pk24′. Cannot insert duplicate key in object ‘xyz’.
I would like to catch pk24 alone.

Is there a way out?

SELECT *, OBJECT_NAME(OBJECT_ID) AS ConstraintName,SCHEMA_NAME(schema_id) AS SchemaName,OBJECT_NAME(parent_object_id) AS TableName,type_desc AS ConstraintType FROM sys.objects WHERE (OBJECTPROPERTY(object_Id, N’IsConstraint’) = 1)

i don’t know if we gain a performance boost or not but it looks more reliable to me