Identifying Primary Keys for table(s) which have no natural PK
Hello,
At my organization, a person used the loader to bring in 4 tables which have no natural PK's. The user is using BI Studio to demo to upper management. He wants to be able to get down to a single row returned, but the table(s) don't appear to have anything unique, and the data pulled from the warehouse has no timestamps, and only dates. Since there can be multiple occurrences of case names, supervisor names, worker names, case numbers, dates, etc. I have no idea how to get anything for sure down to a unique single row returned and this table has no other table in relation to it:
Name Null Type
APPROVAL_PERIOD VARCHAR2(50)
PAGE_NUMBER VARCHAR2(50)
SERVICING_AGENCY VARCHAR2(50)
ELIGIBILITY_OVERRIDES_1 VARCHAR2(50)
SERVICE_AUTH_OVERRIDES_1 VARCHAR2(50)
SUPERVISOR VARCHAR2(50)
ELIGIBILITY_OVERRIDES_2 VARCHAR2(50)
SERVICE_AUTH_OVERRIDES_2 VARCHAR2(50)
WORKER VARCHAR2(50)
ELIGIBILITY_OVERRIDES_3 VARCHAR2(50)
SERVICE_AUTH_OVERRIDES_3 VARCHAR2(50)
TYPE_OF_OVERRIDE VARCHAR2(50)
CASE_NUMBER VARCHAR2(50)
CASE_NAME VARCHAR2(50)
CC_PROGRAM VARCHAR2(50)
PERIOD_BEGIN DATE
OVERRIDE_REASON VARCHAR2(50)
OVERRIDE_VERSION VARCHAR2(2)
APPROVED_DATE DATE
If anyone has some advice on how to proceed with this data I would appreciate it.
Data model normalization is usually different for data warehouses.
Data often is denormalized purposely.
To get "unique" rows you have to aggregate data one or another way.
Work with OLAP/BI is oriented rather on aggregations of facts by dimensions, versus looking to individual rows in fact table.
Aggregations are usually done by BI Studio automatically when user slices and dices, a fact column will be either summarized, or max'ed or aggregated otherwise.
Getting down "to the row" may make sense when user wants to go (drill through) to original data that came from OLTP database.
Usually it is required for investigations or like that.
Normally OLAP/BI applications do no go that deep and manipulate aggregated data on more high level.
He wants to be able to get down to a single row returnedWhat is a reason for that? What will change if he will see that same worker had multiple interactions for a case for one day?
Worker Case Date
Joe Doe SC12345 2013-02-07
Joe Doe SC12345 2013-02-07
Joe Doe SC12345 2013-02-07
Joe Doe SC12345 2013-02-07instead of normally aggregated result
Worker Case Date Count
Joe Doe SC12345 2013-02-07 4Edited by: Mark Malakanov (user11181920) on Feb 7, 2013 12:22 PM
If user still want to drill through, a drill through query needs to be specified and send to OLTP database where this data was originated from.
For above it should be like:
select w.WorkerName, c.Case#, i.Timestamp, ... other details
from Workers@OLTP w
join Cases@OLTP c on c.worker_id=w.worker_id
join Interactions@OLTP i on i.case#=c.case#
where w.WorkerName='Joe Doe' and c.Case#='SC12345'
and TRUNC(i.Timestamp)=TO_DATE('2013-02-07');Edited by: Mark Malakanov (user11181920) on Feb 7, 2013 12:22 PM
Edited by: Mark Malakanov (user11181920) on Feb 7, 2013 12:35 PM
Similar Messages
-
I have a table Gr_order which was populated ages ago and now it need to be populated
with some more look up data.
"Orderid" is primary key for the table but question is how do I insert "orderid" in insert statement in procedure. I could not find any sequence name. There are 1230
records which need to be inserted.
Desc Gr_order
Orderid Number Primary key
Desc Varchar2
INSERT INTO gr_order (orderid,desc)
VALUES (???, upper(each_pom.desc);Thanks
sandyYou find the code that inserts into the table and see if it uses the sequence ;-) . A sequence is not "attached" to a particular table - it is up to the inserting statement to decide if a sequence is to be used.
You might want to check if the table has a trigger - some developers like to emulate MS-sql server behaviour by having a trigger on the table that selects from a sequence. If it has such a trigger, then you just do your inserts without specifying order_id. But if the table does not have such a trigger, then you cannot know if the table was originally populated using a sequence or not (unless you can find the insert statements somewhere.) -
Database Copy can't copy primary key for tables
When I use Tools->Database Copy Wizard to copy my existing schema (for example: azteca_KSMMS) to a brand new schema (azteca), I find the primary key for some of the tables can't be copied, also the views seem not valid under the new schema (azteca). Please give me your comments about that. Thanks for your help.
KevinFor example I have a table called workorder, it has a primary key with name (CW_PKEY_519, on WORKORDERID column) from the source schema (azteca_KSMMS). After I database copy to new schema azteca, the primary key lost. The data of workorder and other indexes have been copied to azteca schema.
Kevin -
How to add primary key for table with existing data?
The table is occupied data already. There was no primary key before, so for every column, there are some duplicate values.
I want to add a new column, which should be of the datatype integer, and can automatically increment, starting from 001. I tried this in Oracle SQL Developer, but it says "ORA-01758: table must be empty to add mandatory (NOT NULL) column". How can I do it? Thanks!Hello, don't call the column ID, that could be an ID relating to anything. Perhaps be a little more specific like emp_id .
Then to update it:
UPDATE Employee emp
SET emp_id = (
SELECT row_num
FROM (
SELECT ROWID,
ROW_NUMBER() OVER (ORDER BY ROWID) row_num
FROM Employee emp2) emp2
WHERE emp.ROWID = emp2.ROWID);That will assign a new EMP_ID for each, beginning at 1.
Then:
ALTER TABLE Employee MODIFY emp_id NOT NULL;
ALTER TABLE Employee ADD CONSTRAINT EMP_PK PRIMARY KEY (EMP_ID);Then create a sequence for future population of the Employee table:
CREATE SEQUENCE emp_seq START WITH <maximum value of emp_id + 1>; -
Is it possible to create many primary key for a table?
hello sir/mam.
i am selva, i am new to abap,
is it possible to create many primary key for a table?
please guide me.Hi Selva,
Many Primary Keys = Composite Key.
i.e some set of columns will enable you to identify a unique row from a set of rows in the table.
This can be achieved by checking the primary key check box in se11.
Hope it helps.
Regards,
Maheswaran.B
Message was edited by: Maheswaran B -
ORA-02266: unique/primary keys in table referenced by enabled foreign keys
Hi,
I am trying to delete data from a table by dropping a partition. I have identified all the child tables by running the following command.
select 'select count(*) from '||table_name||' where employee_id = 100;'
from dba_constraints
where constraint_type='R'
and r_constraint_name in
(select constraint_name from dba_constraints
where constraint_type in ('P','U') and table_name='EMPLOYEE);
'SELECTCOUNT(*)FROM'||TABLE_NAME||'WHEREEMPLOYEE_ID_ID=100;'
select count(*) from PT_ORDERS where employee_id = 100;
select count(*) from PT_DEP where employee_id = 100;
select count(*) from PT_SKILLSET where employee_id = 100;
I dropped the partition for employee_id 100 in all of the child tables. The select count(*) returns 0 rows for each of the above.
When I try to run the below command on the EMPLOYEE table, I get 'ORA-02266: unique/primary keys in table referenced by enabled foreign keys'.
alter table EMPLOYEE drop partition EMP_ID_100;
I cant see why I am unable to drop this partition now as there is now child data in any of the referenced tables. Any suggestions or help on this would be greatly appreciated.
Thanks.
Rgs,
RobYou should disable foreign key constraints first and drop partition. Deletion of rows or dropping partitions in childs don't work in this case
as you have the global dependency:
<PRE>
SQL> create table scott.t (x int primary key, y int)
2 partition by list (y) (
3 partition p_1 values(1), partition p_2 values(2))
4 /
Table created.
SQL> create table scott.t_c (x int references scott.t(x), y int)
2 partition by list (y) (
3 partition p_1 values(1), partition p_2 values(2))
4 /
Table created.
SQL> insert into scott.t values(1,1)
2 /
1 row created.
SQL> insert into scott.t values(2,2)
2 /
1 row created.
SQL> insert into scott.t_c values(1,1)
2 /
1 row created.
SQL> insert into scott.t_c values(2,2)
2 /
1 row created.
SQL> commit;
Commit complete.
SQL> alter table scott.t_c drop partition p_2;
Table altered.
SQL> alter table scott.t drop partition p_2;
alter table scott.t drop partition p_2
ERROR at line 1:
ORA-02266: unique/primary keys in table referenced by enabled foreign keys
SQL> select constraint_name from dba_constraints
2 where owner = 'SCOTT' and constraint_type = 'P'
3 and table_name = 'T';
CONSTRAINT_NAME
SYS_C0011058
SQL> select constraint_name from dba_constraints
2 where owner = 'SCOTT' and constraint_type = 'R'
3 and r_constraint_name = 'SYS_C0011058';
CONSTRAINT_NAME
SYS_C0011059
SQL> alter table scott.t_c disable constraint SYS_C0011059;
Table altered.
SQL> alter table scott.t drop partition p_2;
Table altered.
SQL> alter table scott.t_c enable novalidate constraint SYS_C0011059;
Table altered.
</PRE>
I guess you should consider such option as Referencial partitioning (with some restrictions).
Best wishes,
Dmitry. -
Primary key for an column consisting duplicates
hi,
------->i have created a table and
-------> i have a column consisting of 1000 records (but where i have duplicates)
and now i want to create a primary key for the column
how can i do it....Hi,
You can find records which contains duplicate values for the table column using Oracle exceptions table. Please see a small demonstration:-
SQL> create table test1(id number);
Table created.
SQL> insert into test1 values(&id);
Enter value for id: 1
old 1: insert into test1 values(&id)
new 1: insert into test1 values(1)
1 row created.
SQL> /
Enter value for id: 2
old 1: insert into test1 values(&id)
new 1: insert into test1 values(2)
1 row created.
SQL> /
Enter value for id: 3
old 1: insert into test1 values(&id)
new 1: insert into test1 values(3)
1 row created.
SQL> /
Enter value for id: 1
old 1: insert into test1 values(&id)
new 1: insert into test1 values(1)
1 row created.
SQL> /
Enter value for id: 3
old 1: insert into test1 values(&id)
new 1: insert into test1 values(3)
1 row created.
SQL> /
Enter value for id: 4
old 1: insert into test1 values(&id)
new 1: insert into test1 values(4)
1 row created.
SQL> /
Enter value for id: 5
old 1: insert into test1 values(&id)
new 1: insert into test1 values(5)
1 row created.
SQL> commit;
Commit complete.
SQL> alter table test1 add constraint id_pk primary key(id);
alter table test1 add constraint id_pk primary key(id)
ERROR at line 1:
ORA-02437: cannot validate (SYS.ID_PK) - primary key violated
SQL> alter table test1 add constraint id_pk primary key(id) exceptions into exceptions;
alter table test1 add constraint id_pk primary key(id) exceptions into exceptions
ERROR at line 1:
ORA-02445: Exceptions table not found
SQL> @?/rdbms/admin/utlexcpt
Table created.
SQL> alter table test1 add constraint id_pk primary key(id) exceptions into exceptions;
alter table test1 add constraint id_pk primary key(id) exceptions into exceptions
ERROR at line 1:
ORA-02437: cannot validate (SYS.ID_PK) - primary key violated
SQL> desc exceptions
Name Null? Type
ROW_ID ROWID
OWNER VARCHAR2(30)
TABLE_NAME VARCHAR2(30)
CONSTRAINT VARCHAR2(30)
SQL> select * from exceptions;
ROW_ID OWNER TABLE_NAME CONSTRAINT
AAAc95AABAAA9EpAAD SYS TEST1 ID_PK
AAAc95AABAAA9EpAAA SYS TEST1 ID_PK
AAAc95AABAAA9EpAAE SYS TEST1 ID_PK
AAAc95AABAAA9EpAAC SYS TEST1 ID_PK
SQL> select * from test1 where rowid in(select row_id from exceptions);
ID
3
1
1
3
Thanks
Edited by: rarain on May 28, 2013 12:10 PM -
Hi everybody,
I am wondering if there is such thing as primary key for a view? Assuming, that my view is based on a single table and it's just a subset of columns (some of them renamed) including PK of the table it's based on.
The reason for my question is this - we're using Reverse POCO generator which automatically generates C# Model class and Configuration file for our tables and views. For the view is lists all the columns as a key and therefore I obviously can not use normal
way of updating that view. I posted that as an issue here https://efreversepoco.codeplex.com/workitem/115 but I am thinking there is no such thing as the "primary key" for a view.
The query used to generate the classes is extremely complex already but may be it can be modified to get the PK ?
SELECT [Extent1].[SchemaName],
[Extent1].[Name] AS TableName,
[Extent1].[TABLE_TYPE] AS TableType,
[UnionAll1].[Ordinal],
[UnionAll1].[Name] AS ColumnName,
[UnionAll1].[IsNullable],
[UnionAll1].[TypeName],
ISNULL([UnionAll1].[MaxLength],0) AS MaxLength,
ISNULL([UnionAll1].[Precision], 0) AS Precision,
ISNULL([UnionAll1].[Default], '') AS [Default],
ISNULL([UnionAll1].[DateTimePrecision], '') AS [DateTimePrecision],
ISNULL([UnionAll1].[Scale], 0) AS Scale,
[UnionAll1].[IsIdentity],
[UnionAll1].[IsStoreGenerated],
CASE WHEN ([Project5].[C2] IS NULL) THEN CAST(0 AS BIT)
ELSE [Project5].[C2]
END AS PrimaryKey
FROM (
SELECT QUOTENAME(TABLE_SCHEMA) + QUOTENAME(TABLE_NAME) [Id],
TABLE_SCHEMA [SchemaName],
TABLE_NAME [Name],
TABLE_TYPE
FROM INFORMATION_SCHEMA.TABLES
WHERE TABLE_TYPE IN ('BASE TABLE', 'VIEW')
) AS [Extent1]
INNER JOIN (
SELECT [Extent2].[Id] AS [Id],
[Extent2].[Name] AS [Name],
[Extent2].[Ordinal] AS [Ordinal],
[Extent2].[IsNullable] AS [IsNullable],
[Extent2].[TypeName] AS [TypeName],
[Extent2].[MaxLength] AS [MaxLength],
[Extent2].[Precision] AS [Precision],
[Extent2].[Default],
[Extent2].[DateTimePrecision] AS [DateTimePrecision],
[Extent2].[Scale] AS [Scale],
[Extent2].[IsIdentity] AS [IsIdentity],
[Extent2].[IsStoreGenerated] AS [IsStoreGenerated],
0 AS [C1],
[Extent2].[ParentId] AS [ParentId]
FROM (
SELECT QUOTENAME(c.TABLE_SCHEMA) + QUOTENAME(c.TABLE_NAME) + QUOTENAME(c.COLUMN_NAME) [Id],
QUOTENAME(c.TABLE_SCHEMA) + QUOTENAME(c.TABLE_NAME) [ParentId],
c.COLUMN_NAME [Name],
c.ORDINAL_POSITION [Ordinal],
CAST(CASE c.IS_NULLABLE
WHEN 'YES' THEN 1
WHEN 'NO' THEN 0
ELSE 0
END AS BIT) [IsNullable],
CASE WHEN c.DATA_TYPE IN ('varchar', 'nvarchar', 'varbinary')
AND c.CHARACTER_MAXIMUM_LENGTH = -1 THEN c.DATA_TYPE + '(max)'
ELSE c.DATA_TYPE
END AS [TypeName],
c.CHARACTER_MAXIMUM_LENGTH [MaxLength],
CAST(c.NUMERIC_PRECISION AS INTEGER) [Precision],
CAST(c.DATETIME_PRECISION AS INTEGER) [DateTimePrecision],
CAST(c.NUMERIC_SCALE AS INTEGER) [Scale],
c.COLLATION_CATALOG [CollationCatalog],
c.COLLATION_SCHEMA [CollationSchema],
c.COLLATION_NAME [CollationName],
c.CHARACTER_SET_CATALOG [CharacterSetCatalog],
c.CHARACTER_SET_SCHEMA [CharacterSetSchema],
c.CHARACTER_SET_NAME [CharacterSetName],
CAST(0 AS BIT) AS [IsMultiSet],
CAST(COLUMNPROPERTY(OBJECT_ID(QUOTENAME(c.TABLE_SCHEMA) + '.' + QUOTENAME(c.TABLE_NAME)), c.COLUMN_NAME, 'IsIdentity') AS BIT) AS [IsIdentity],
CAST(COLUMNPROPERTY(OBJECT_ID(QUOTENAME(c.TABLE_SCHEMA) + '.' + QUOTENAME(c.TABLE_NAME)), c.COLUMN_NAME, 'IsComputed')
| CASE WHEN c.DATA_TYPE = 'timestamp' THEN 1
ELSE 0
END AS BIT) AS [IsStoreGenerated],
c.COLUMN_DEFAULT AS [Default]
FROM INFORMATION_SCHEMA.COLUMNS c
INNER JOIN INFORMATION_SCHEMA.TABLES t
ON c.TABLE_CATALOG = t.TABLE_CATALOG
AND c.TABLE_SCHEMA = t.TABLE_SCHEMA
AND c.TABLE_NAME = t.TABLE_NAME
AND t.TABLE_TYPE IN ('BASE TABLE', 'VIEW')
) AS [Extent2]
UNION ALL
SELECT [Extent3].[Id] AS [Id],
[Extent3].[Name] AS [Name],
[Extent3].[Ordinal] AS [Ordinal],
[Extent3].[IsNullable] AS [IsNullable],
[Extent3].[TypeName] AS [TypeName],
[Extent3].[MaxLength] AS [MaxLength],
[Extent3].[Precision] AS [Precision],
[Extent3].[Default],
[Extent3].[DateTimePrecision] AS [DateTimePrecision],
[Extent3].[Scale] AS [Scale],
[Extent3].[IsIdentity] AS [IsIdentity],
[Extent3].[IsStoreGenerated] AS [IsStoreGenerated],
6 AS [C1],
[Extent3].[ParentId] AS [ParentId]
FROM (
SELECT QUOTENAME(c.TABLE_SCHEMA) + QUOTENAME(c.TABLE_NAME) + QUOTENAME(c.COLUMN_NAME) [Id],
QUOTENAME(c.TABLE_SCHEMA) + QUOTENAME(c.TABLE_NAME) [ParentId],
c.COLUMN_NAME [Name],
c.ORDINAL_POSITION [Ordinal],
CAST(CASE c.IS_NULLABLE
WHEN 'YES' THEN 1
WHEN 'NO' THEN 0
ELSE 0
END AS BIT) [IsNullable],
CASE WHEN c.DATA_TYPE IN ('varchar', 'nvarchar', 'varbinary')
AND c.CHARACTER_MAXIMUM_LENGTH = -1 THEN c.DATA_TYPE + '(max)'
ELSE c.DATA_TYPE
END AS [TypeName],
c.CHARACTER_MAXIMUM_LENGTH [MaxLength],
CAST(c.NUMERIC_PRECISION AS INTEGER) [Precision],
CAST(c.DATETIME_PRECISION AS INTEGER) AS [DateTimePrecision],
CAST(c.NUMERIC_SCALE AS INTEGER) [Scale],
c.COLLATION_CATALOG [CollationCatalog],
c.COLLATION_SCHEMA [CollationSchema],
c.COLLATION_NAME [CollationName],
c.CHARACTER_SET_CATALOG [CharacterSetCatalog],
c.CHARACTER_SET_SCHEMA [CharacterSetSchema],
c.CHARACTER_SET_NAME [CharacterSetName],
CAST(0 AS BIT) AS [IsMultiSet],
CAST(COLUMNPROPERTY(OBJECT_ID(QUOTENAME(c.TABLE_SCHEMA) + '.' + QUOTENAME(c.TABLE_NAME)), c.COLUMN_NAME, 'IsIdentity') AS BIT) AS [IsIdentity],
CAST(COLUMNPROPERTY(OBJECT_ID(QUOTENAME(c.TABLE_SCHEMA) + '.' + QUOTENAME(c.TABLE_NAME)), c.COLUMN_NAME, 'IsComputed')
| CASE WHEN c.DATA_TYPE = 'timestamp' THEN 1
ELSE 0
END AS BIT) AS [IsStoreGenerated],
c.COLUMN_DEFAULT [Default]
FROM INFORMATION_SCHEMA.COLUMNS c
INNER JOIN INFORMATION_SCHEMA.VIEWS v
ON c.TABLE_CATALOG = v.TABLE_CATALOG
AND c.TABLE_SCHEMA = v.TABLE_SCHEMA
AND c.TABLE_NAME = v.TABLE_NAME
WHERE NOT (
v.TABLE_SCHEMA = 'dbo'
AND v.TABLE_NAME IN ('syssegments', 'sysconstraints')
AND SUBSTRING(CAST(SERVERPROPERTY('productversion') AS VARCHAR(20)), 1, 1) = 8
) AS [Extent3]
) AS [UnionAll1]
ON (0 = [UnionAll1].[C1])
AND ([Extent1].[Id] = [UnionAll1].[ParentId])
LEFT OUTER JOIN (
SELECT [UnionAll2].[Id] AS [C1],
CAST(1 AS BIT) AS [C2]
FROM (
SELECT QUOTENAME(tc.CONSTRAINT_SCHEMA) + QUOTENAME(tc.CONSTRAINT_NAME) [Id],
QUOTENAME(tc.TABLE_SCHEMA) + QUOTENAME(tc.TABLE_NAME) [ParentId],
tc.CONSTRAINT_NAME [Name],
tc.CONSTRAINT_TYPE [ConstraintType],
CAST(CASE tc.IS_DEFERRABLE
WHEN 'NO' THEN 0
ELSE 1
END AS BIT) [IsDeferrable],
CAST(CASE tc.INITIALLY_DEFERRED
WHEN 'NO' THEN 0
ELSE 1
END AS BIT) [IsInitiallyDeferred]
FROM INFORMATION_SCHEMA.TABLE_CONSTRAINTS tc
WHERE tc.TABLE_NAME IS NOT NULL
) AS [Extent4]
INNER JOIN (
SELECT 7 AS [C1],
[Extent5].[ConstraintId] AS [ConstraintId],
[Extent6].[Id] AS [Id]
FROM (
SELECT QUOTENAME(CONSTRAINT_SCHEMA) + QUOTENAME(CONSTRAINT_NAME) [ConstraintId],
QUOTENAME(TABLE_SCHEMA) + QUOTENAME(TABLE_NAME) + QUOTENAME(COLUMN_NAME) [ColumnId]
FROM INFORMATION_SCHEMA.KEY_COLUMN_USAGE
) AS [Extent5]
INNER JOIN (
SELECT QUOTENAME(c.TABLE_SCHEMA) + QUOTENAME(c.TABLE_NAME) + QUOTENAME(c.COLUMN_NAME) [Id],
QUOTENAME(c.TABLE_SCHEMA) + QUOTENAME(c.TABLE_NAME) [ParentId],
c.COLUMN_NAME [Name],
c.ORDINAL_POSITION [Ordinal],
CAST(CASE c.IS_NULLABLE
WHEN 'YES' THEN 1
WHEN 'NO' THEN 0
ELSE 0
END AS BIT) [IsNullable],
CASE WHEN c.DATA_TYPE IN ('varchar', 'nvarchar', 'varbinary')
AND c.CHARACTER_MAXIMUM_LENGTH = -1 THEN c.DATA_TYPE + '(max)'
ELSE c.DATA_TYPE
END AS [TypeName],
c.CHARACTER_MAXIMUM_LENGTH [MaxLength],
CAST(c.NUMERIC_PRECISION AS INTEGER) [Precision],
CAST(c.DATETIME_PRECISION AS INTEGER) [DateTimePrecision],
CAST(c.NUMERIC_SCALE AS INTEGER) [Scale],
c.COLLATION_CATALOG [CollationCatalog],
c.COLLATION_SCHEMA [CollationSchema],
c.COLLATION_NAME [CollationName],
c.CHARACTER_SET_CATALOG [CharacterSetCatalog],
c.CHARACTER_SET_SCHEMA [CharacterSetSchema],
c.CHARACTER_SET_NAME [CharacterSetName],
CAST(0 AS BIT) AS [IsMultiSet],
CAST(COLUMNPROPERTY(OBJECT_ID(QUOTENAME(c.TABLE_SCHEMA) + '.' + QUOTENAME(c.TABLE_NAME)),
c.COLUMN_NAME, 'IsIdentity') AS BIT) AS [IsIdentity],
CAST(COLUMNPROPERTY(OBJECT_ID(QUOTENAME(c.TABLE_SCHEMA) + '.' + QUOTENAME(c.TABLE_NAME)),
c.COLUMN_NAME, 'IsComputed')
| CASE WHEN c.DATA_TYPE = 'timestamp' THEN 1
ELSE 0
END AS BIT) AS [IsStoreGenerated],
c.COLUMN_DEFAULT AS [Default]
FROM INFORMATION_SCHEMA.COLUMNS c
INNER JOIN INFORMATION_SCHEMA.TABLES t
ON c.TABLE_CATALOG = t.TABLE_CATALOG
AND c.TABLE_SCHEMA = t.TABLE_SCHEMA
AND c.TABLE_NAME = t.TABLE_NAME
AND t.TABLE_TYPE IN ('BASE TABLE', 'VIEW')
) AS [Extent6]
ON [Extent6].[Id] = [Extent5].[ColumnId]
UNION ALL
SELECT 11 AS [C1],
[Extent7].[ConstraintId] AS [ConstraintId],
[Extent8].[Id] AS [Id]
FROM (
SELECT CAST( NULL AS NVARCHAR (1)) [ConstraintId], CAST( NULL AS NVARCHAR (MAX)) [ColumnId] WHERE 1= 2
) AS [Extent7]
INNER JOIN (
SELECT QUOTENAME(c.TABLE_SCHEMA) + QUOTENAME(c.TABLE_NAME) + QUOTENAME(c.COLUMN_NAME) [Id],
QUOTENAME(c.TABLE_SCHEMA) + QUOTENAME(c.TABLE_NAME) [ParentId],
c.COLUMN_NAME [Name],
c.ORDINAL_POSITION [Ordinal],
CAST(CASE c.IS_NULLABLE
WHEN 'YES' THEN 1
WHEN 'NO' THEN 0
ELSE 0
END AS BIT) [IsNullable],
CASE WHEN c.DATA_TYPE IN ('varchar', 'nvarchar', 'varbinary')
AND c.CHARACTER_MAXIMUM_LENGTH = -1 THEN c.DATA_TYPE + '(max)'
ELSE c.DATA_TYPE
END AS [TypeName],
c.CHARACTER_MAXIMUM_LENGTH [MaxLength],
CAST(c.NUMERIC_PRECISION AS INTEGER) [Precision],
CAST(c.DATETIME_PRECISION AS INTEGER) AS [DateTimePrecision],
CAST(c.NUMERIC_SCALE AS INTEGER) [Scale],
c.COLLATION_CATALOG [CollationCatalog],
c.COLLATION_SCHEMA [CollationSchema],
c.COLLATION_NAME [CollationName],
c.CHARACTER_SET_CATALOG [CharacterSetCatalog],
c.CHARACTER_SET_SCHEMA [CharacterSetSchema],
c.CHARACTER_SET_NAME [CharacterSetName],
CAST(0 AS BIT) AS [IsMultiSet],
CAST(COLUMNPROPERTY(OBJECT_ID(QUOTENAME(c.TABLE_SCHEMA) + '.' + QUOTENAME(c.TABLE_NAME)),
c.COLUMN_NAME, 'IsIdentity') AS BIT) AS [IsIdentity],
CAST(COLUMNPROPERTY(OBJECT_ID(QUOTENAME(c.TABLE_SCHEMA) + '.' + QUOTENAME(c.TABLE_NAME)),
c.COLUMN_NAME, 'IsComputed')
| CASE WHEN c.DATA_TYPE = 'timestamp' THEN 1
ELSE 0
END AS BIT) AS [IsStoreGenerated],
c.COLUMN_DEFAULT [Default]
FROM INFORMATION_SCHEMA.COLUMNS c
INNER JOIN INFORMATION_SCHEMA.VIEWS v
ON c.TABLE_CATALOG = v.TABLE_CATALOG
AND c.TABLE_SCHEMA = v.TABLE_SCHEMA
AND c.TABLE_NAME = v.TABLE_NAME
WHERE NOT (
v.TABLE_SCHEMA = 'dbo'
AND v.TABLE_NAME IN ('syssegments', 'sysconstraints')
AND SUBSTRING(CAST(SERVERPROPERTY('productversion') AS VARCHAR(20)), 1, 1) = 8
) AS [Extent8]
ON [Extent8].[Id] = [Extent7].[ColumnId]
) AS [UnionAll2]
ON (7 = [UnionAll2].[C1])
AND ([Extent4].[Id] = [UnionAll2].[ConstraintId])
WHERE [Extent4].[ConstraintType] = N'PRIMARY KEY'
) AS [Project5]
ON [UnionAll1].[Id] = [Project5].[C1]
WHERE NOT ([Extent1].[Name] IN ('EdmMetadata', '__MigrationHistory'));
Thanks in advance.
UPDATE. I made a quick Google search and found this
http://stackoverflow.com/questions/1013333/entity-framework-and-sql-server-view/2715299#2715299
It sounds as an interesting idea to try although I am not sure it will work with POCO Generator. But I'm going to try it now anyway.
For every expert, there is an equal and opposite expert. - Becker's Law
My blog
My TechNet articlesUPDATE. Changing my view to be
USE [SiriusSQL_SysManager]
GO
/****** Object: View [dbo].[SiriusV_SalesPointsLicensesList] Script Date: 4/30/2015 1:28:54 PM ******/
DROP VIEW [dbo].[SiriusV_SalesPointsLicensesList]
GO
/****** Object: View [dbo].[SiriusV_SalesPointsLicensesList] Script Date: 4/30/2015 1:28:54 PM ******/
SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
create view [dbo].[SiriusV_SalesPointsLicensesList]
as
select ISNULL(salespoint,'') as Salespoint
, CAST(loc_suffix as smallint) as Id
, CAST(descrip as varchar(25)) as [Description]
, cast(mode as tinyint) as Mode
, cast(is_access as bit) as AccessWare
, cast(is_adv_inv as bit) as AdvInventory
, cast(is_cw_crdt as bit) as CardwareCredit
, cast(is_debit as bit) as DebitWare
, cast(is_dir_net as bit) as DirectNet
, cast(is_fbquick as bit) as FBQuickService
, cast(is_fbtable as bit) as FBTableService
, cast(is_grpschd as bit) as ResourceScheduling
, cast(is_guests as bit) as Guests
, cast(is_pass as bit) as PassWare
, cast(is_pms as bit) as PMSInterface
, cast(is_rental as bit) as RentWare
, cast(is_dual as bit) as DualBook
, cast(is_pod as bit) as PodBook
, cast(is_private as bit) as PrivateBook
, cast(is_resrv as bit) as Reservations
, cast(is_teesht as bit) as TeeSheet
, cast(is_rungk as bit) as RunGKInSales
, cast(is_resseat as bit) as ReservedSeating
, cast(is_kcashls as bit) as KioskCashless
, cast(is_ksales as bit) as KioskSales
, cast(is_kwilcal as bit) as KioskWillCall
, cast(is_payplan as bit) as PaymentPlan
, cast(is_test as bit) as Test
from dbo.sales_pt
where hidden = 0 ;
GO
worked!!! I now have just salespoint as the Key.
For every expert, there is an equal and opposite expert. - Becker's Law
My blog
My TechNet articles -
Issues while changing primary key in table
Hi
I have one table. In that two fields are primary key. I want to change the second PK as FK only. But when i am changing this field as FK. Its showing one error 'Primary Key Change not permitted for value Table ZCAUSECATMASTER' . How to avoid this error.
Please help me.Hi.....
Remove that primary key for the second field and assign your foregin key..table to that filed.....
So when you give entries in that second filed it will be validated with its foregin key table......its nothing but check table...
what all values in the foregin table only can given.....
regards
raja -
Primary Key for DSO (2LIS_11_VAKON)
Dear all:
I have to create a Standard DSO out of 2LIS_11_VAKON Datasource. My question is how do I define the Primary Key for this DSO...
I mean: (Doc.Number + Pos.Number) is enough?. or do I need some extra(s) field(s) in the Primary Key?
Regards.Hi
I have to create a Standard DSO out of 2LIS_11_VAKON Datasource. My question is how do I define the Primary Key for this DSO...
I mean: (Doc.Number + Pos.Number) is enough?. or do I need some extra(s) field(s) in the Primary Key?
You can Install the Standard content You cannot Create any Standard DSO.
You can Create a customized DSO . Though the DSO is a Table There is no need of Primery Key Function here. Hope You are Talking about Data Fields and Key Fields.
But there is a Standard DSO FOR 0SD_O05 Order: Condition Data.. you can install the same.
Both fields are needed to report analytical data based on your criteria.
http://help.sap.com/saphelp_nw70/helpdata/EN/3c/7b88408bc0bb4de10000000a1550b0/frameset.htm -
Setting a Primary Key for a newly added Column
Is it possible for setting a primary key for a newly added column in a table having records?
Hi,
km**** wrote:
oh if the table has records then it is not possible ah...No, the table can have rows at the time you add the Primary Key constraint. Solomon was just saying that the column(s) that you are making into the Primary Key must already have unique values. A Primary Key can not be NULL, so you must do the steps in this order:
(1) ALTER TABLE to add the column(s) (if this hasn't been done already)
(2) UPDATE the table, to put unique values in all rows for the Primary Key column(s)
(3) Add the PRIMARY KEY constraint
You need to do step (2) before you do step (3) -
Primary Key for Programmatic View
Hi,
I am using Jdev R2 for ADF. When I create a programmatic VO, Is it necessary to have a primary key for Programmatic VO, if yes, what can be the effective primary key?
- VinothHi Vino,
Primary keys are used to loop through the collection on tables, list of values and to access rows on a view object.
Juan Camilo -
How to genarate Primary Key for each record in XI Mapping
Hi,
I need to genarate primary key for each record in the paylod in mapping..
Eg: if i have a 10 reacords i need to have a primary key for each record..starting with the
Record Primary key
First 001
Secound 002
Tenth 010
If i need to write any user defined funtion... can any one provide the code for it.
Thanks
AmareshHi Bavesh,
I will explain with the example:
XML:
<Record 1>
<Primary key>
<value1>
<value2>
</Record 1>
<Record 2>
<Primary Key>
</Record 2>
like above i will be getting n number of recored and in each record i need to genarate primary key in XI.and the key sould be in sequence...
like for the first record 1 secound 2 like that primary key should add to the privous Primary key.
and this is for each payload.
Thanks
Amaresh -
Populate choice component with 2 primary key in table
There are 2 primary key in table. When I create my uix with jheadstart, the UIXs don't populate choice component. How can I create automatically populate choice component, with 2 PK in table?
A choice component can only have one value attribute. So if you have a composite primary key in the table you use to populate the choice, it will not work correctly, because based on this single attribute the choice cannot render the correct row as selected.
Steven Davelaar,
JHeadstart Team. -
No matching unique or primary key for this column-list --error
CREATE TABLE CUSTOMER(CUS_NUM NUMBER(10) PRIMARY KEY,CUS_NAME VARCHAR2(50),ADDRESS VARCHAR2(50));
CREATE TABLE ITEM(ITEM_NUM NUMBER(10),ITEM_DESC VARCHAR2(50),TYPE VARCHAR2(50),ITEM_STATUS VARCHAR2(50),
CONSTRAINT PK_ITEM PRIMARY KEY(ITEM_NUM,TYPE));
CREATE TABLE ITEM_TYPE(TYPE VARCHAR2(50) PRIMARY KEY,TYPE_NAME VARCHAR2(50),SKU VARCHAR2(50),RATING VARCHAR2(50));
CREATE TABLE ORDERS(ORDER_NUMBER NUMBER(10),CUS_NUM NUMBER(10),ITEM_NUM NUMBER(10),ORDER_POINT VARCHAR2(50),DISPATCH_POINT VARCHAR2(50),
PRIORITY VARCHAR2(50),ORDER_STATUS VARCHAR2(50),
CONSTRAINT PK_ORDERS PRIMARY KEY(ORDER_NUMBER,ITEM_NUM),
CONSTRAINT FK1_ORDERS FOREIGN KEY (CUS_NUM) REFERENCES CUSTOMER(CUS_NUM),
CONSTRAINT FK2_ORDERS FOREIGN KEY (ITEM_NUM) REFERENCES ITEM(ITEM_NUM));
Hi,friends i created tables as above when i creating 'ORDERS' table the error will fir as "ORA-02270: no matching unique or primary key for this column-list".
can any body solve this issue.
Cheers,
Sudhir reddy.For your own sanity, you should learn to format your code, or use a tool to do it for you. For the sanity of the forum members, you should preserve that code using the \ tagsCREATE TABLE CUSTOMER
CUS_NUM NUMBER(10) PRIMARY KEY,
CUS_NAME VARCHAR2(50),
ADDRESS VARCHAR2(50)
CREATE TABLE ITEM
ITEM_NUM NUMBER(10),
ITEM_DESC VARCHAR2(50),
TYPE VARCHAR2(50),
TYPE is a reserved word. Call it ITEM_TYPEITEM_STATUS VARCHAR2(50),
CONSTRAINT PK_ITEM PRIMARY KEY(ITEM_NUM,TYPE)
CREATE TABLE ITEM_TYPE
TYPE VARCHAR2(50) PRIMARY KEY,
again, with the reserved word.TYPE_NAME VARCHAR2(50),
SKU VARCHAR2(50),
RATING VARCHAR2(50)
CREATE TABLE ORDERS
ORDER_NUMBER NUMBER(10),
be consistent. NUM or NUMBER. Pick one and stick with it.CUS_NUM NUMBER(10),
ITEM_NUM NUMBER(10),
ORDER_POINT VARCHAR2(50),
DISPATCH_POINT VARCHAR2(50),
PRIORITY VARCHAR2(50),
ORDER_STATUS VARCHAR2(50),
CONSTRAINT PK_ORDERS PRIMARY KEY(ORDER_NUMBER,ITEM_NUM),
CONSTRAINT FK1_ORDERS FOREIGN KEY (CUS_NUM) REFERENCES CUSTOMER(CUS_NUM),
CONSTRAINT FK2_ORDERS FOREIGN KEY (ITEM_NUM) REFERENCES ITEM(ITEM_NUM)
Maybe you are looking for
-
Partner Application in release 2
Hello, I have to create a new PL/SQL partner application in the release 902. Where is the new sso sdk for the new Portal release? In the new pdk seems that it is the same of the 309 release (ssosdk307_032101 or ssosdk307)... And the pl/sql example co
-
Import oracle.jdbc.driver.* problem.
I've got classes111b.zip from this site, and put it into my $ORACLE_HOME/jdbc/lib; add this: "$ORACLE_HOME/jdbc/lib/classes111b.zip" to my CLASSPATH; use env command, I can see it, but when I visit my jsp page, there occured an error: Error: 500 Loca
-
Dear All, While doing MIGO, excise calculation is not been done by the users. I want the list of document in which excise should have been done by not done by the users. This would be the ABAP development but I need to know which table while give the
-
Looking to confirm the successfully MOVED Transport Requests & their order
Hi, I have manually ordered about 20 requests which were successfully transported from Dev to Quality; After 6 months of testing, I now need to move them to production. Is there a way to know these objects which were successfully moved by Basis to Qu
-
Meaning of (000) fields in the analytics subject areas
What is the meaning/difference between Total Liabilities and Total Liabilities (000), and all such pairs of these type of dollar fields (Potential Revenue, Closed Revenue...)