I have a table, department , with several bit fields to indicate department types
One is Warehouse (when true, indicate the department is warehouse)
And I have another table, ManagersForWarehouses with following structure:
ID autoinc
WarehouseID int (foreign key reference DepartmentID from departments)
ManagerID int (foreign key reference EmployeeID from employees)
StartDate
EndDate
To set new manager for warehouse, I insert in this table with EndDate null, and I have a trigger that sets EndDate for previous record for that warehouse = StartDate for new manager, so a single manager appears for a warehouse at a certain time.
I want to add two check constraints as follows, but not sure how to do this
do not allow to insert into ManagersForWarehouses if WarehouseID is not marked as warehouse
Do not allow to uncheck Warehouse if there are records in ManagersForWarehouses
Thanks
In Departments table, add a unique constraint (DepartmentID, IsWarehouse)
In ManagersForWarehouses table, add column IsWarehouse, and have a CHECK constraint enforce that IsWarehouse='Y'
In ManagersForWarehouses table, add a FK on (WarehouseID , IsWarehouse) referrring to Departments(DepartmentID, IsWarehouse), and with ON UPDATE CASCADE clause.
And you are all set with 100% rock solid integrity, without any loopholes. Only trusted constraints do not have any loopholes. Triggers are less reliable.
You should use a triggers to do this. One on each table. Make sure to account for multiple record inserts, updates or deletes. To do anything else would be putting your data at risk. Data does NOT get into databases only from applications. You cannot afford to enforce this sort of thing from the application unless you want data that is incorrect and useless.
Don't put this kind of constraints on the DB table.
It is better to enforce such a thing using code (business logic).
EDIT: Not related to your question, but you should not use the trigger as well to perform updates to other rows. I don't have solid links to back it up with.
Related
Is it possible to add a foreign key constraint on a table which will allow values which do NOT exist in another table?
In the example below, both tables contain the field USER_ID. The constraint is that a customer and and an employee cannot have the same USER_ID value.
I am very limited in adding new tables or altering one of the tables in any way.
CUSTOMER
--------------------------
USER_ID varchar2(10)
EMPLOYEE
--------------------------
USER_ID varchar2(10)
I thought of a few workarounds, such as a view which contains the data from both tables or adding an insert trigger on the table I can modify.
No, no such thing exists, though it is possible to fake.
If you want to do this relationally (which would be a lot better than views/triggers) the easy option is to add a E to all employee IDs and a C to all customer IDs. However, this won't work if you have other attributes and you want to ensure they're not the same person (i.e. you're not just interested in the ID).
If this is the case you need to create a third table, let's call it PEOPLE:
create table people (
user_id varchar2(10) not null
, user_type varchar2(1) not null
, constraint pk_people primary key (user_id)
, constraint chk_people_user_types check ( user_type in ('C','E') )
);
C would stand for customer and E for employee in the check constraint. You then need to create a unique index/constraint on PEOPLE:
create index ui_people_id_type on people ( user_id, user_type );
Personally, I'd stop here and completely drop your CUSTOMER and EMPLOYEE tables; they no longer have any use and your problem has been solved.
If you don't have the ability to add new columns/tables you need to speak to the people who do and convince them to change it. Over-complicating things only leads to errors in logic and confusion (believe me - using a view means you need a lot of triggers to maintain your tables and you'll have to ensure that someone only ever updates the view). It's a lot easier to do things properly, even if they take longer.
However, if you really want to continue you alter your CUSTOMER and EMPLOYEE tables to include their USER_TYPE and ensure that it's always the same for every row in the table, i.e.:
alter table customers add user_type default 'C' not null;
alter table customers add constraint chk_customers_type
check ( user_type is not null and user_type = 'C' );
Unless you are willing to change the data model as someone else has suggested, the simplest way to proceed with the existing structure while maintaining mutual exclusion is to issue check constraints on the user_ids of both tables such that they validate only to mutually exclusive series.
For example, you could issue checks to ensure that only even numbers are assigned to customers and odd numbers to employees ( or vice-versa).
Or, since both IDS are varchar, stipulate using your check constraint that the ID begins with a known substring, such as 'EMP' or 'CUST'.
But these are only tricks and are hardly relational. Ideally, one would revise the data model. Hope this helps.
I have the following situation. My table is:
Table: CompanyEmployees
EmployeeID
Date of Birth
Date Joined
I also want to store the sales information for each employee. I have this:
Table: DealsCompleted
ID
EmployeeID
Deal Name
Deal Amount
My question is this- should there be a column in CompanyEmployees called "DealsCompletedID" which directly refers to the ID column in DealsCompleted, or is it acceptabe to just create a foreign key between the two Employee ID columns? Does this disadvantage the design or potentially distort the normalization?
I am unclear what the rule is as to whether I should include an extra column in CompanyEmployees or not.
EDIT Please assume there will only be one row in the deal table, per employee.
A FOREIGN KEY should point from one table to its referenced row in a parent table, the two tables should generally not reference each other (with foreign keys defined in both).
The FOREIGN KEY is most appropriately defined in the DealsCompleted table, which points back to CompanyEmployees.EmployeeID. Think about it this way - The CompanyEmployees table stores information about employees. Deals they completed do not really count as information about employees. However, the employee who completed a deal is a part of the information about a deal, so the key belongs there.
Having DealsCompleted.EmployeeID will allow for a proper one to many relationship between employees and deals. That is, one employee can have as many related rows in DealsCompleted as needed. Including a DealsCompleted column in the CompanyEmployees table on the other hand, would require you to either duplicate rows about employees, which breaks normalization, or include multiple DealCompletedID values in one column which is also incorrect.
Update after edit above Even if you plan for only a one-to-one relationship (one deal per employee), it is still more appropriate to reference the EmployeeID in DealsCompleted rather than the other way around (or both ways). ...And it allows you to expand to one-to-many, when the need arises.
Assuming that the relationship will always be one-to-one, as you state, then the answer depends on what is the primary entity within the Domain Model. If this database is at its core a database about Deals, and employee data is ancillary, then I would add an EmployeeId FK column in the Deal table. If otoh, this is a database about Employees, and Deals are ancillary, then eliminate the EmployeeId column in the Deal table, and add a DealId FK column to the Employeee table.
I have a database table called Lesson:
columns: [LessonID, LessonNumber, Description] ...plus some other columns
I have another table called Lesson_ScoreBasedSelection:
columns: [LessonID,NextLessonID_1,NextLessonID_2,NextLessonID_3]
When a lesson is completed, its LessonID is looked up in the Lesson_ScoreBasedSelection table to get the three possible next lessons, each of which are associated with a particular range of scores. If the score was 0-33, the LessonID stored in NextLessonID_1 would be used. If the score was 34-66, the LessonID stored in NextLessonID_2 would be used, and so on.
I want to constrain all the columns in the Lesson_ScoreBasedSelection table with foreign keys referencing the LessonID column in the lesson table, since every value in the Lesson_ScoreBasedSelection table must have an entry in the LessonID column of the Lesson table. I also want cascade updates turned on, so that if a LessonID changes in the Lesson table, all references to it in the Lesson_ScoreBasedSelection table get updated.
This particular cascade update seems like a very straightforward, one-way update, but when I try to apply a foreign key constraint to each field in the Lesson_ScoreBasedSelection table referencing the LessonID field in the Lesson table, I get the error:
Introducing FOREIGN KEY constraint 'c_name' on table 'Lesson_ScoreBasedSelection' may cause cycles or multiple cascade paths.
Can anyone explain why I'm getting this error or how I can achieve the constraints and cascading updating I described?
You can't have more than one cascading RI link to a single table in any given linked table. Microsoft explains this:
You receive this error message because
in SQL Server, a table cannot appear
more than one time in a list of all
the cascading referential actions that
are started by either a DELETE or an
UPDATE statement. For example, the
tree of cascading referential actions
must only have one path to a
particular table on the cascading
referential actions tree.
Given the SQL Server constraint on this, why don't you solve this problem by creating a table with SelectionID (PK), LessonID, Next_LessonID, QualifyingScore as the columns. Use a constraint to ensure LessonID and QualifyingScore are unique.
In the QualifyingScore column, I'd use a tinyint, and make it 0, 1, or 2. That, or you could do a QualifyingMinScore and QualifyingMaxScore column so you could say,
SELECT * FROM NextLesson
WHERE LessonID = #MyLesson
AND QualifyingMinScore <= #MyScore
AND #MyScore <= QualifyingMaxScore
Cheers,
Eric
I'm designing a database and I'm not sure how to define one of the relationships. Here's the situation:
An invoice is created
If the product is not in stock then it needs to be manufactured and so a work order is created.
The relationship is one-to-one. However work orders are sometimes created for other purposes so the WorkOrder table will also be linked to other tables in a similar one-to-one relationship. Also, some Invoices won't have a work order at all. This means I can't define these relationships in the normal way by using the same primary key in both tables. Instead of doing this I've created a linking table and then set unique indexes on both fields to define the one-to-one relationship (see image).
(source: markevans.org)
.
Is this the best way?
Cheers
Mark
EDIT: I just realised that this design will allow a single work order to be linked to an invoice and also to one of the other tables I mentioned via 2 linking tables. I guess no solution is perfect.
Okay, this answer is SQL Server specific, but should be adaptable to other RDBMSs, with a little work. So far as I see, we have the following constraints:
An invoice may be associated with 0 or 1 Work Orders
A Work Order must be associated with an invoice or an ABC or a DEF
I'd design the WorkOrder table as follows:
CREATE TABLE WorkOrder (
WorkOrderID int IDENTITY(1,1) not null,
/* Other Columns */
InvoiceID int null,
ABCID int null,
DEFID int null,
/* Etc for other possible links */
constraint PK_WorkOrder PRIMARY KEY (WorkOrderID),
constraint FK_WorkOrder_Invoices FOREIGN KEY (InvoiceID) references Invoice (InvoiceID),
constraint FK_WorkOrder_ABC FOREIGN KEY (ABCID) references ABC (ABCID),
/* Etc for other FKs */
constraint CK_WorkOrders_SingleFK CHECK (
CASE WHEN InvoiceID is null THEN 0 ELSE 1 END +
CASE WHEN ABCID is null THEN 0 ELSE 1 END +
CASE WHEN DEFID is null THEN 0 ELSE 1 END
/* + other FK columns */
= 1
)
)
So, basically, this table is constrained to only FK to one other table, no matter how many PKs are defined. If necessary, a computed column could tell you the "Type" of item that this is linked to, based on which FK column is non-null, or the type and a single int column could be real columns, and InvoiceID, ABCID, etc could be computed columns.
The final thing to ensure is that an invoice only has 0 or 1 Work Orders. If your RDMBS ignores nulls in unique constraints, this is as simple as applying such a constraint to each FK column. For SQL Server, you need to use a filtered index (>=2008) or an indexed view (<=2005). I'll just show the filtered index:
CREATE UNIQUE INDEX IX_WorkItems_UniqueInvoices on
WorkItem (InvoiceID) where (InvoiceID is not null)
Another way to deal with keeping WorkOrders straight is to include a WorkOrder type column in WorkOrder (e.g. 'Invoice','ABC','DEF'), including a computed or column constrained by check constraint to contain the matching value in the link table, and introduce a second foreign key:
CREATE TABLE WorkOrder (
WorkOrderID int IDENTITY(1,1) not null,
Type varchar(10) not null,
constraint PK_WorkOrder PRIMARY KEY (WorkOrderID),
constraint UQ_WorkOrder_TypeCheck UNIQUE (WorkOrderID,Type),
constraint CK_WorkOrder_Types CHECK (Type in ('INVOICE','ABC','DEF'))
)
CREATE TABLE Invoice_WorkOrder (
InvoiceID int not null,
WorkOrderID int not null,
Type varchar(10) not null default 'INVOICE',
constraint PK_Invoice_WorkOrder PRIMARY KEY (InvoiceID),
constraint UQ_Invoice_WorkOrder_OrderIDs UNIQUE (WorkOrderID),
constraint FK_Invoice_WorkOrder_Invoice FOREIGN KEY (InvoiceID) references Invoice (InvoiceID),
constraint FK_Invoice_WorkOrder_WorkOrder FOREIGN KEY (WorkOrderID) references WorkOrder (WorkOrderID),
constraint FK_Invoice_WorkOrder_TypeCheck FOREIGN KEY (WorkOrderID,Type) references WorkOrder (WorkOrderID,Type),
constraint CK_Invoice_WorkOrder_Type CHECK (Type = 'INVOICE')
)
The only disadvantage to this model, although closer to your original proposal, is that you can have a work order that isn't actually linked to any other item (although it claims to be for an e.g INVOICE).
What you have looks to be a perfectly normal way to construct your tables.
If you think you might like to use only one link table between your WorkOrder table and whatever other tables that may have WorkOrders, you could use a link table like:
WorkOrders
OtherId (Could be InvoiceId, or an ID for SomethingElse that may have a WorkOrder)
OtherType (ENUM - something like 'Invoice', 'SomethingElse')
WorkOrderId
So the issue is that you can have invoices that don't have work orders and work orders that don't have invoices but the two need to be linked when there is a link. I would say based upon that description that your database diagram is pretty good. This would open you up to allowing more than a one-to-one relationship. This way down the road you can consider having two work orders for one invoice. You might also have one work order that handles two invoices. This opens you up to a lot of possibilities that you may not need now but that you might in the future.
I would recommend your current design. In the future, you may want to add more information about the link between invoice and work order. This middle table will allow you to add this information.
In the interest of fairness to the other side of the coin, you do need to consider speed/number of tables/etc. that this will cause. For example, you have now created a third table which increased your table count by 50% in this example. Look at the rest of your database. If you did this everywhere, you would probably have the most normalized database but it might not be the most performant because of all the joins that are necessary. Basically, this isn't a "one-size-fits-all" solution. Instead it is a design choice. Personally, I hate nullable foreign key fields. I find they don't give me the granularity I usually want with my database designs.
Your schema corresponds to a many-to-many link between the 2 tables. You are de facto opening here the possibility to have one work order for multiple invoices, and multiple work orders for one invoice. The model offers then possibilities far above the rules you are setting.
You could use a simpler schema, that will reflect the (0,1) relation between work orders and invoices, and the (0,1) relation between Invoices and Work orders:
a Work Order can be independant from
an invoice, or linked to one specific
invoice: it has a (0,1) relation to Invoice table
An invoice can have no work orders, or one work orders: it has a (0,1) relation to Work Orders Table
Such a relation can be translated by the following model and rules
Invoice
id_Invoice, Primary Key
WorkOrder
id_WorkOrder, Primary Key
id_Invoice, Foreign Key, Nulls accepted, unique value
With such a structure, it will be easy to add new 'dependants' to work orders table. If, for example, you want to open the possibility to launch work orders from restocking orders (where you want to have minimal quantities of some items in stock), you can then just add the corresponding field to the WorkOrder table:
id_RestockingOrder, ForeignKey, Nulls accepted, unique value
You'll be then able to 'see' from where your WorkOrder comes: an invoice, a restocking order, etc.
Seems it corresponds to your needs.
Edit:
as noted by #mark, SQL Server will not allow multiple null values, in contradiction with ANSI specs (check here for some more details), As we do not want to wait for SQL Server 2011 to have this rule implemented, there is a workaround here, where you can build a view excluding the null values and set a unique index on this view. I must admit that I did not like this solution ...
There is still the possibility to implement the 'unique if not null' rule in your code. It will still be simpler than implementing the many-to-many model (with the Invoice_WorkOrder table) you are proposing and manage all additional unicity rules that you'll need to implement.
There is no real need for the link table, just have them linked directly and allow for NULL in the reference field of the work order. Because a work order can be linked to multiple tables what I would do is add a reference id on every work order to every table that can link from it. So you would have:
Invoice
PK - ID
FK - WorkOrderID
SomeOtherTable
PK - ID
FK - WorkOrderID
WorkOrder
PK - ID
FK - InvoiceID (allow NULL)
FK - SomeOtherTableID (allow NULL)
To make sure a WorkOrder is linked to only one item, you have to use code to validate the row (or perhaps a stored procedure which I cannot come up with right now).
EDIT: PS, if you want to use a link table, give it a generic name and add all the linked tables with the same sort of construct I just described allowing for NULL's. In my eyes adding the extra table makes the schema larger than it needs to be, but if a work order contains a lot of big text fields it could increase performance slightly and reduce database size with all the indexes flying around. In anything but the largest applications, I would consider it over-normalization though, but that is a matter of style.
What kind of sql tricks you use to enter data into two tables with a circular reference in between.
Employees
EmployeeID <PK>
DepartmentID <FK> NOT NULL
Departments
DepartmentID <PK>
EmployeeID <FK> NOT NULL
The employee belongs to a department, a department has to have a manager (department head).
Do I have to disable constraints for the insert to happen?
I assume your Departments.EmployeeID is a department head. What I'd do is make that column nullable; then you can create the department first, then the employee.
Q: Do I have to disable constraints for the insert to happen?
A: In Oracle, no, not if the foreign key constraints are DEFERRABLE (see example below)
For Oracle:
SET CONSTRAINTS ALL DEFERRED;
INSERT INTO Departments values ('foo','dummy');
INSERT INTO Employees values ('bar','foo');
UPDATE Departments SET EmployeeID = 'bar' WHERE DepartmentID = 'foo';
COMMIT;
Let's unpack that:
(autocommit must be off)
defer enforcement of the foreign key constraint
insert a row to Department table with a "dummy" value for the FK column
insert a row to Employee table with FK reference to Department
replace "dummy" value in Department FK with real reference
re-enable enforcement of the constraints
NOTES: disabling a foreign key constraint takes effect for ALL sessions, DEFERRING a constraint is at a transaction level (as in the example), or at the session level (ALTER SESSION SET CONSTRAINTS=DEFERRED;)
Oracle has allowed for foreign key constraints to be defined as DEFERRABLE for at least a decade. I define all foreign key constraints (as a matter of course) to be DEFERRABLE INITIALLY IMMEDIATE. That keeps the default behavior as everyone expects, but allows for manipulation without requiring foreign keys to be disabled.
see AskTom: http://www.oracle.com/technology/oramag/oracle/03-nov/o63asktom.html
see AskTom: http://asktom.oracle.com/pls/asktom/f?p=100:11:0::::P11_QUESTION_ID:10954765239682
see also: http://www.idevelopment.info/data/Oracle/DBA_tips/Database_Administration/DBA_12.shtml
[EDIT]
A: In Microsoft SQL Server, you can't defer foreign key constraints like you can in Oracle. Disabling and re-enabling the foreign key constraint is an approach, but I shudder at the prospect of 1) performance impact (the foreign key constraint being checked for the ENTIRE table when the constraint is re-enabled), 2) handling the exception if (when?) the re-enable of the constraint fails. Note that disabling the constraint will affect all sessions, so while the constraint is disabled, other sessions could potentially insert and update rows which will cause the reenable of the constraint to fail.
With SQL Server, a better approach is to remove the NOT NULL constraint, and allow for a NULL as temporary placeholder while rows are being inserted/updated.
For SQL Server:
-- (with NOT NULL constraint removed from Departments.EmployeeID)
insert into Departments values ('foo',NULL)
go
insert into Employees values ('bar','foo')
go
update Departments set EmployeeID = 'bar' where DepartmentID = 'foo'
go
[/EDIT]
This problem could be solved with deferable constraints. Such constraints are checked when the whole transaction is commited, thus allowing you to insert both employee and department in the same transaction, referring to each other. (Assuming the data model makes sense)
Refactor the schema by removing the circular reference.
Delete an ID column from either of the table schema.
Departments.EmployeeID doesn't seem to belong there in my opinion.
I can't think of a non hackish way to do this. I think you will need to remove the constraint or do some type of silly dummy values that get updated after all the inserts.
I'd recommend refactoring the DB schema. I can't think of any reasons why you would want it to work this way.
Maybe something like, Employee, EmployeeDepartment (EmployeeId, DepartmentId) and Department would be a better way to accomplish the same goal.
You could create a row in the Department table for 'Unassigned'
To create a new department with a new Employee you then would
Create the Employee (EmployeeA) in the 'Unassigned' Department
Create the new department (DepartmentA) with the employee EmployeeA
Update EmployeeA to be in DepartmentA
This wouldn't invalidate your current schema, and you could set up a task to be run regularly to check there are no members of the Unassigned department.
You would also need to create a default employee to be the Employee of Unassigned
EDIT:
The solution proposed by chaos is much simpler though
There are a few good designs I've used. All involve removing the "manager" EmployeeID from the Department table and removing the DepartmentID from the Employee table. I've seen a couple answers which mention it, but I'll clarify how we used it:
I typically end up with an EmployeeDepartment relationship link table - many to many, usually with flags like IsManager, IsPrimaryManager, IsAdmin, IsBackupManager etc., which clarify the relationship Some may be constrained so that there is only one Primary Manager allowed per department (although a person can be a PrimaryManager of multiple departments). If you don't like the single table, then you can have multiple tables: EmployeeDepartment, ManagerDepartment, etc. but then you could have situations where a person is a manager but not an employee, etc.
We also typically allowed people to be members of multiple departments.
For simplified access, you can provide views which perform the join appropriately.
Yes, in this instance you will have to disable a foreign key.
You need to get rid of one or the other reference permanently . This is not a viable design structure. Which has to be entered first? Department or Employee? Unless your departments are all one employee big, the structure doesn't make sense anyway as each employee would have to have a distinct departmentid.