Consider the problem above where the 'CommonChild' entity can be a child of either sub-type A or B, but not C. How would I go about designing the physical model in a relational [SQL] database?
Ideally, the solution would allow...
for an identifying relationship between CommonChild and it's related sub-type.
a 1:N relationship.
Possible Solutions
Add an additional sub-type to the super-type and move sub-type A and B under the new sub-type. The CommonChild can then have a FK constraint on the newly created sub-type. Works for the above, but not if an additional entity is added which can have a relationship with sub-type A and C, but not B.
Add a FK constraint between the CommonChild and SuperType. Use a trigger or check constraint (w/ UDF) against the super-type's discriminator before allowing a new tuple into CommonChild. Seems straight forward, but now CommonChild almost seems like new subtype itself (which it is not).
My model is fundamentally flawed. Remodel and the problem should go away.
I'm looking for other possible solutions or confirmation of one of the above solutions I've already proposed.
Thanks!
EDIT
I'm going to implement the exclusive foreign key solution provided by Branko Dimitrijevic (see accepted answer).
I am going to make a slight modifications in this case as:
the super-type, sub-type, and "CommonChild" all have the same PKs and;
the PKs are 3 column composites.
The modification is to to create an intermediate table whose sole role is to enforce the exclusive FK constraint between the sub-types and the "CommonChild" (exact model provided by Dimitrijevic minus the "CommonChild's" attributes.). The CommonChild's PK will have a normal FK constraint to the intermediate table.
This will prevent the "CommonChild" from having 2 sets of 3 column composite FKs. Plus, since the identifying relationship is maintained from super-type to "CommonChild", [read] queries can effectively ignore the intermediate table altogether.
Looks like you need a variation of exclusive foreign keys:
CREATE TABLE CommonChild (
Id AS COALESCE(SubTypeAId, SubTypeBId) PERSISTED PRIMARY KEY,
SubTypeAId int REFERENCES SubTypeA (SuperId),
SubTypeBId int REFERENCES SubTypeB (SuperId),
Attr6 varchar,
CHECK (
(SubTypeAId IS NOT NULL AND SubTypeBId IS NULL)
OR (SubTypeAId IS NULL AND SubTypeBId IS NOT NULL)
)
);
There are couple of thing to note here:
There are two NULL-able FOREIGN KEYs.
There is a CHECK that allows exactly one of these FKs be non-NULL.
There is a computed column Id which equals one of the FKs (whichever is currently non-NULL) which is also a PRIMARY KEY. This ensures that:
One parent cannot have multiple children.
A "grandchild" table can reference the CommonChild.Id directly from its FK. The SuperType.Id is effectively popagated all the way down.
We don't have to mess with NULL-able UNIQUE constraints, which are problematic in MS SQL Server (see below).
A DBMS-agnostic way of of doing something similar would be...
CREATE TABLE CommonChild (
Id int PRIMARY KEY,
SubTypeAId int UNIQUE REFERENCES SubTypeA (SuperId),
SubTypeBId int UNIQUE REFERENCES SubTypeB (SuperId),
Attr6 varchar,
CHECK (
(SubTypeAId IS NOT NULL AND SubTypeAId = Id AND SubTypeBId IS NULL)
OR (SubTypeAId IS NULL AND SubTypeBId IS NOT NULL AND SubTypeBId = Id)
)
)
Unfortunately a UNIQUE column containing more than one NULL is not allowed by MS SQL Server, which is not the case in most DBMSes. However, you can just omit the UNIQUE constraint if you don't want to reference SubTypeAId or SubTypeBId directly.
Wondering what am I missing here?
Admittedly, it is hard without having the wording of the specific problem, but things do feel a bit upside-down.
Related
This is a relational database design question, not specific to any RDBMS. A simplified case:
I have two tables Cars and Trucks. They have both have a column, say RegistrationNumber, that must be unique across the two tables.
This could probably be enforced with some insert/update triggers, but I'm looking for a more "clean" solution if possible.
One way to achieve this could be to add a third table, Vehicles, that holds the RegistrationNumber column, then add two additional columns to Vehicles, CarID and TruckID. But then for each row in Vehicles, one of the columns CarID or TruckID would always be NULL, because a RegistrationNumber applies to either a Car or a Truck leaving the other column with a NULL value.
Is there anyway to enforce a unique RegistrationNumber value across multiple tables, without introducing NULL columns or relying on triggers?
This is a bit complicated. Having the third table, Vehicles is definitely part of the solution. The second part is guaranteeing that a vehicle is either a car or a truck, but not both.
One method is the "list-all-the-possibilities" method. This is what you propose with two columns. In addition, this should have a constraint to verify that only one of the ids is filled in. A similar approach is to have the CarId and TruckId actually be the VehicleId. This reduces the number of different ids floating around.
Another approach uses composite keys. The idea is:
create table Vehicles (
Vehicle int primary key,
Registration varchar(255),
Type varchar(255),
constraint chk_type check (type in ('car', 'truck')),
constraint unq_type_Vehicle unique (type, vehicle), -- this is redundant, but necessary
. . .
);
create table car (
VehicleId int,
Type varchar(255), -- always 'car' in this table
constraint fk_car_vehicle foreign key (VehicleId) references Vehicles(VehicleId),
constraint fk_car_vehicle_type foreign key (Type, VehicleId) references Vehicles(Type, VehicleId)
);
See the following tags: class-table-inheritance shared-primary-key
You have already outlined class table inheritance in your question. The tag will just add a few details, and show you some other questions whose answers may help.
Shared primary key is a handy way of enforcing the one-to-one nature of IS-A relationships such as the relationship between a vehicle and a truck. It also allows a foreign key in some other table to reference a vehicle and also a truck or a car, as the case may be.
You can add the third table Vehicles containing a single column RegistratioNumber on which you apply the unique constraint, then on the existing tables - Cars and Trucks - use the RegistrationNumber as a foreign key on the Vehicles table. In this way you don't need an extra id, avoid the null problem and enforce the uniqueness of the registration number.
Update - this solution doesn't prevent a car and a truck to share the same registration number. To enforce this constraint you need to add either triggers or logic beyond plain SQL. Otherwise you may want to take a look at Gordon's solution that involves Composite Foreign Keys.
I have a column that sometimes will be null. This column is also a foreign key, so I want to know if I'll have problems with performance or with data consistency if this column will have weight
I know its a foolish question but I want to be sure.
There is no problem necessarily with this, other than it is likely indication that you might have poorly normalized design. There might be performance implications due to the way indexes are structured and the sparseness of the column with nulls, but without knowing your structure or intended querying scenarios any conclusions one might draw would be pure speculation.
A better solution might be a shared primary key where table A has a primary key, and there is zero or one records in B with the same primary key.
If table A can have one or zero B, but more than one A can refer to B, then what you have is a one to many relationship. This can be represented as Pieter laid out in his answer. This allows multiple A records to refer to the same B, and in turn each B may optionally refer to an A.
So you see there are two optional structures to address this problem, and choosing each is not guesswork. There is a distinct rational between why you would choose one or the other, but it depends on the nature of your relationships you are modelling.
Instead of this design:
create table Master (
ID int identity not null primary key,
DetailID int null references Detail(ID)
)
go
create table Detail (
ID int identity not null primary key
)
go
consider this instead
create table Master (
ID int identity not null primary key
)
go
create table Detail (
ID int identity not null primary key,
MasterID int not null references Master(ID)
)
go
Now the Foreign Key is never null, rather the existence (or not) of the Detail record indicates whether it exists.
If a Detail can exist for multiple records, create a mapping table to manage the relationship.
I'd like to hear your suggestions on this very basic question:
Imagine these three tables:
--DROP TABLE a_to_b;
--DROP TABLE a;
--DROP TABLE b;
CREATE TABLE A
(
ID NUMBER NOT NULL ,
NAME VARCHAR2(20) NOT NULL ,
CONSTRAINT A_PK PRIMARY KEY ( ID ) ENABLE
);
CREATE TABLE B
(
ID NUMBER NOT NULL ,
NAME VARCHAR2(20) NOT NULL ,
CONSTRAINT B_PK PRIMARY KEY ( ID ) ENABLE
);
CREATE TABLE A_TO_B
(
id NUMBER NOT NULL,
a_id NUMBER NOT NULL,
b_id NUMBER NOT NULL,
somevalue1 VARCHAR2(20) NOT NULL,
somevalue2 VARCHAR2(20) NOT NULL,
somevalue3 VARCHAR2(20) NOT NULL
) ;
How would you design table a_to_b?
I'll give some discussion starters:
synthetic id-PK column or combined a_id,b_id-PK (dropping the "id" column)
When synthetic: What other indices/constraints?
When combined: Also index on b_id? Or even b_id,a_id (don't think so)?
Also combined when these entries are referenced themselves?
Also combined when these entries perhaps are referenced themselves in the future?
Heap or Index-organized table
Always or only up to x "somevalue"-columns?
I know that the decision for one of the designs is closely related to the question how the table will be used (read/write ratio, density, etc.), but perhaps we get a 20/80 solution as blueprint for future readers.
I'm looking forward to your ideas!
Blama
I have always made the PK be the combination of the two FKs, a_id and b_id in your example. Adding a synthetic id field to this table does no good, since you never end up looking for a row based on a knowledge of its id.
Using the compound PK gives you a constraint that prevents the same instance of the relationship between a and b from being inserted twice. If duplicate entries need to be permitted, there's something wrong with your data model at the conceptual level.
The index you get behind the scenes (for every DBMS I know of) will be useful to speed up common joins. An extra index on b_id is sometimes useful, depending on the kinds of joins you do frequently.
Just as a side note, I don't use the name "id" for all my synthetic pk columns. I prefer a_id, b_id. It makes it easier to manage the metadata, even though it's a little extra typing.
CREATE TABLE A_TO_B
(
a_id NUMBER NOT NULL REFERENCES A (a_id),
b_id NUMBER NOT NULL REFERENCES B (b_id),
PRIMARY KEY (a_id, b_id),
...
) ;
It's not unusual for ORMs to require (or, in more clueful ORMs, hope for) an integer column named "id" in addition to whatever other keys you have. Apart from that, there's no need for it. An id number like that makes the table wider (which usually degrades I/O performance just slightly), and adds an index that is, strictly speaking, unnecessary. It isn't necessary to identify the entity--the existing key does that--and it leads new developers into bad habits. (Specifically, giving every table an integer column named "id", and believing that that column alone is the only key you need.)
You're likely to need one or more of these indexed.
a_id
b_id
{a_id, b_id}
{b_id, a_id}
I believe Oracle should automatically index {a_id, b_id}, because that's the primary key. Oracle doesn't automatically index foreign keys. Oracle's indexing guidelines are online.
In general, you need to think carefully about whether you need ON UPDATE CASCADE or ON DELETE CASCADE. In Oracle, you only need to think carefully about whether you need ON DELETE CASCADE. (Oracle doesn't support ON UPDATE CASCADE.)
the other comments so far are good.
also consider adding begin_dt and end_dt to the relationship. in this way, you can manage a good number of questions about each relationship through time. (consider baseline issues)
Just trying to figure out the best way to design my table for the following scenario:
I have several areas in my system (documents, projects, groups and clients) and each of these can have comments logged against them.
My question is should I have one table like this:
CommentID
DocumentID
ProjectID
GroupID
ClientID
etc
Where only one of the ids will have data and the rest will be NULL or should I have a separate CommentType table and have my comments table like this:
CommentID
CommentTypeID
ResourceID (this being the id of the project/doc/client)
etc
My thoughts are that option 2 would be more efficient from an indexing point of view. Is this correct?
Option 2 is not a good solution for a relational database. It's called polymorphic associations (as mentioned by #Daniel Vassallo) and it breaks the fundamental definition of a relation.
For example, suppose you have a ResourceId of 1234 on two different rows. Do these represent the same resource? It depends on whether the CommentTypeId is the same on these two rows. This violates the concept of a type in a relation. See SQL and Relational Theory by C. J. Date for more details.
Another clue that it's a broken design is that you can't declare a foreign key constraint for ResourceId, because it could point to any of several tables. If you try to enforce referential integrity using triggers or something, you find yourself rewriting the trigger every time you add a new type of commentable resource.
I would solve this with the solution that #mdma briefly mentions (but then ignores):
CREATE TABLE Commentable (
ResourceId INT NOT NULL IDENTITY,
ResourceType INT NOT NULL,
PRIMARY KEY (ResourceId, ResourceType)
);
CREATE TABLE Documents (
ResourceId INT NOT NULL,
ResourceType INT NOT NULL CHECK (ResourceType = 1),
FOREIGN KEY (ResourceId, ResourceType) REFERENCES Commentable
);
CREATE TABLE Projects (
ResourceId INT NOT NULL,
ResourceType INT NOT NULL CHECK (ResourceType = 2),
FOREIGN KEY (ResourceId, ResourceType) REFERENCES Commentable
);
Now each resource type has its own table, but the serial primary key is allocated uniquely by Commentable. A given primary key value can be used only by one resource type.
CREATE TABLE Comments (
CommentId INT IDENTITY PRIMARY KEY,
ResourceId INT NOT NULL,
ResourceType INT NOT NULL,
FOREIGN KEY (ResourceId, ResourceType) REFERENCES Commentable
);
Now Comments reference Commentable resources, with referential integrity enforced. A given comment can reference only one resource type. There's no possibility of anomalies or conflicting resource ids.
I cover more about polymorphic associations in my presentation Practical Object-Oriented Models in SQL and my book SQL Antipatterns.
Read up on database normalization.
Nulls in the way you describe would be a big indication that the database isn't designed properly.
You need to split up all your tables so that the data held in them is fully normalized, this will save you a lot of time further down the line guaranteed, and it's a lot better practice to get into the habit of.
From a foreign key perspective, the first example is better because you can have multiple foreign key constraints on a column but the data has to exist in all those references. It's also more flexible if the business rules change.
To continue from #OMG Ponies' answer, what you describe in the second example is called a Polymorphic Association, where the foreign key ResourceID may reference rows in more than one table. However in SQL databases, a foreign key constraint can only reference exactly one table. The database cannot enforce the foreign key according to the value in CommentTypeID.
You may be interested in checking out the following Stack Overflow post for one solution to tackle this problem:
MySQL - Conditional Foreign Key Constraints
The first approach is not great, since it is quite denormalized. Each time you add a new entity type, you need to update the table. You may be better off making this an attribute of document - I.e. store the comment inline in the document table.
For the ResourceID approach to work with referential integrity, you will need to have a Resource table, and a ResourceID foreign key in all of your Document, Project etc.. entities (or use a mapping table.) Making "ResourceID" a jack-of-all-trades, that can be a documentID, projectID etc.. is not a good solution since it cannot be used for sensible indexing or foreign key constraint.
To normalize, you need to the comment table into one table per resource type.
Comment
-------
CommentID
CommentText
...etc
DocumentComment
---------------
DocumentID
CommentID
ProjectComment
--------------
ProjectID
CommentID
If only one comment is allowed, then you add a unique constraint on the foreign key for the entity (DocumentID, ProjectID etc.) This ensures that there can only be one row for the given item and so only one comment. You can also ensure that comments are not shared by using a unique constraint on CommentID.
EDIT: Interestingly, this is almost parallel to the normalized implementation of ResourceID - replace "Comment" in the table name, with "Resource" and change "CommentID" to "ResourceID" and you have the structure needed to associate a ResourceID with each resource. You can then use a single table "ResourceComment".
If there are going to be other entities that are associated with any type of resource (e.g. audit details, access rights, etc..), then using the resource mapping tables is the way to go, since it will allow you to add normalized comments and any other resource related entities.
I wouldn't go with either of those solutions. Depending on some of the specifics of your requirements you could go with a super-type table:
CREATE TABLE Commentable_Items (
commentable_item_id INT NOT NULL,
CONSTRAINT PK_Commentable_Items PRIMARY KEY CLUSTERED (commentable_item_id))
GO
CREATE TABLE Projects (
commentable_item_id INT NOT NULL,
... (other project columns)
CONSTRAINT PK_Projects PRIMARY KEY CLUSTERED (commentable_item_id))
GO
CREATE TABLE Documents (
commentable_item_id INT NOT NULL,
... (other document columns)
CONSTRAINT PK_Documents PRIMARY KEY CLUSTERED (commentable_item_id))
GO
If the each item can only have one comment and comments are not shared (i.e. a comment can only belong to one entity) then you could just put the comments in the Commentable_Items table. Otherwise you could link the comments off of that table with a foreign key.
I don't like this approach very much in your specific case though, because "having comments" isn't enough to put items together like that in my mind.
I would probably go with separate Comments tables (assuming that you can have multiple comments per item - otherwise just put them in your base tables). If a comment can be shared between multiple entity types (i.e., a document and a project can share the same comment) then have a central Comments table and multiple entity-comment relationship tables:
CREATE TABLE Comments (
comment_id INT NOT NULL,
comment_text NVARCHAR(MAX) NOT NULL,
CONSTRAINT PK_Comments PRIMARY KEY CLUSTERED (comment_id))
GO
CREATE TABLE Document_Comments (
document_id INT NOT NULL,
comment_id INT NOT NULL,
CONSTRAINT PK_Document_Comments PRIMARY KEY CLUSTERED (document_id, comment_id))
GO
CREATE TABLE Project_Comments (
project_id INT NOT NULL,
comment_id INT NOT NULL,
CONSTRAINT PK_Project_Comments PRIMARY KEY CLUSTERED (project_id, comment_id))
GO
If you want to constrain comments to a single document (for example) then you could add a unique index (or change the primary key) on the comment_id within that linking table.
It's all of these "little" decisions that will affect the specific PKs and FKs. I like this approach because each table is clear on what it is. In databases that's usually better then having "generic" tables/solutions.
Of the options you give, I would go for number 2.
Option 2 is a good way to go. The issue that I see with that is you are putting the resouce key on that table. Each of the IDs from the different resources could be duplicated. When you join resources to the comments you will more than likely come up with comments that do not belong to that particular resouce. This would be considered a many to many join. I would think a better option would be to have your resource tables, the comments table, and then tables that cross reference the resource type and the comments table.
If you carry the same sort of data about all comments regardless of what they are comments about, I'd vote against creating multiple comment tables. Maybe a comment is just "thing it's about" and text, but if you don't have other data now, it's likely you will: date the comment was entered, user id of person who made it, etc. With multiple tables, you have to repeat all these column definitions for each table.
As noted, using a single reference field means that you could not put a foreign key constraint on it. This is too bad, but it doesn't break anything, it just means you have to do the validation with a trigger or in code. More seriously, joins get difficult. You can just say "from comment join document using (documentid)". You need a complex join based on the value of the type field.
So while the multiple pointer fields is ugly, I tend to think that's the right way to go. I know some db people say there should never be a null field in a table, that you should always break it off into another table to prevent that from happening, but I fail to see any real advantage to following this rule.
Personally I'd be open to hearing further discussion on pros and cons.
Pawnshop Application:
I have separate tables for Loan, Purchase, Inventory & Sales transactions.
Each tables rows are joined to their respective customer rows by:
customer.pk [serial] = loan.fk [integer];
= purchase.fk [integer];
= inventory.fk [integer];
= sale.fk [integer];
I have consolidated the four tables into one table called "transaction", where a column:
transaction.trx_type char(1) {L=Loan, P=Purchase, I=Inventory, S=Sale}
Scenario:
A customer initially pawns merchandise, makes a couple of interest payments, then decides he wants to sell the merchandise to the pawnshop, who then places merchandise in Inventory and eventually sells it to another customer.
I designed a generic transaction table where for example:
transaction.main_amount DECIMAL(7,2)
in a loan transaction holds the pawn amount,
in a purchase holds the purchase price,
in inventory and sale holds sale price.
This is clearly a denormalized design, but has made programming alot easier and improved performance. Any type of transaction can now be performed from within one screen, without the need to change to different tables.
I have data that kinda looks like this...
Elements
Class | Synthetic ID (pk)
A | 2
A | 3
B | 4
B | 5
C | 6
C | 7
Elements_Xref
ID (pk) | Synthetic ID | Real ID (fk)
. | 2 | 77-8F <--- A class
. | 3 | 30-7D <--- A class
. | 6 | 21-2A <--- C class
. | 7 | 30-7D <--- C class
So I have these elements that are assigned synthetic IDs and are grouped into classes. But these synthetic IDs are then paired with Real IDs that we actually care about. There is also a constraint that a Real ID cannot recur in a single class. How can I capture all of this in one coherent design?
I don't want to jam the Real ID into the upper table because
It is nullable (there are periods where we don't know what the Real ID of something should be).
It's a foreign key to more data.
Obviously this could be done with triggers acting as constraints, but I'm wondering if this could be implemented with regular constraints/unique indexes. Using SQL Server 2005.
I've thought about having two main tables SyntheticByClass and RealByClass and then putting IDs of those tables into another xref/link table, but that still doesn't guarantee that the classes of both elements match. Also solvable via trigger.
Edit: This is keyword stuffing but I think it has to do with normalization.
Edit^2: As indicated in the comments below, I seem to have implied that foreign keys cannot be nullable. Which is false, they can! But what cannot be done is setting a unique index on fields where NULLs repeat. Although unique indexes support NULL values, they cannot constraint more than one NULL in a set. Since the Real ID assignment is initially sparse, multiple NULL Real IDs per class is more than likely.
Edit^3: Dropped the redundant Elements.ID column.
Edit^4: General observations. There seems to be three major approaches at work, one of which I already mentioned.
Triggers. Use a trigger as a constraint to break any data operations that would corrupt the integrity of the data.
Index a view that joins the tables. Fantastic, I had no idea you could do that with views and indexes.
Create a multi-column foreign key. Didn't think of doing this, didn't know it was possible. Add the Class field to the Xref table. Create a UNIQUE constraint on (Class + Real ID) and a foreign key constraint on (Class + Synthetic ID) back to the Elements table.
Comments from before the question was made into a 'bonus' question
What you'd like to be able to do is express that the join of Elements and Elements_Xref has a unique constraint on Class and Real ID. If you had a DBMS that supported SQL-92 ASSERTION constraints, you could do it.
AFAIK, no DBMS supports them, so you are stuck with using triggers.
It seems odd that the design does not constrain Real ID to be unique across classes; from the discussion, it seems that a given Real ID could be part of several different classes. Were the Real ID 'unique unless null', then you would be able to enforce the uniqueness more easily, if the DBMS supported the 'unique unless null' concept (most don't; I believe there is one that does, but I forget which it is).
Comments before edits made 2010-02-08
The question rules out 'jamming' the Real_ID in the upper table (Elements); it doesn't rule out including the Class in the lower table (Elements_Xref), which then allows you to create a unique index on Class and Real_ID in Elements_Xref, achieving (I believe) the required result.
It isn't clear from the sample data whether the synthetic ID in the Elements table is unique or whether it can repeat with different classes (or, indeed whether a synthetic ID can be repeated in a single class). Given that there seems to be an ID column (which presumably is unique) as well as the Synthetic ID column, it seems reasonable to suppose that sometimes the synthetic ID repeats - otherwise there are two unique columns in the table for no very good reason. For the most part, it doesn't matter - but it does affect the uniqueness constraint if the class is copied to the Elements_Xref table. One more possibility; maybe the Class is not needed in the Elements table at all; it should live only in the Elements_Xref table. We don't have enough information to tell whether this is a possibility.
Comments for changes made 2010-02-08
Now that the Elements table has the Synthetic ID as the primary key, things are somewhat easier. There's a comment that the 'Class' information actually is a 'month', but I'll try to ignore that.
In the Elements_Xref table, we have an unique ID column, and then a Synthetic ID (which is not marked as a foreign key to Elements, but presumably must actually be one), and the Real ID. We can see from the sample data that more than one Synthetic ID can map to a given Real ID. It is not clear why the Elements_Xref table has both the ID column and the Synthetic ID column.
We do not know whether a single Synthetic ID can only map to a single Real ID or whether it can map to several Real ID values.
Since the Synthetic ID is the primary key of Elements, we know that a single Synthetic ID corresponds to a single Class.
We don't know whether the mapping of Synthetic ID to Real ID varies over time (it might as Class is date-related), and whether the old state has to be remembered.
We can assume that the tables are reduced to the bare minimum and that there are other columns in each table, the contents of which are not directly material to the question.
The problem states that the Real ID is a foreign key to other data and can be NULL.
I can't see a perfectly non-redundant design that works.
I think that the Elements_Xref table should contain:
Synthetic ID
Class
Real ID
with (Synthetic ID, Class) as a 'foreign key' referencing Elements, and a NOT NULL constraint on Real ID, and a unique constraint on (Class, Real ID).
The Elements_Xref table only contains rows for which the Real ID is known - and correctly enforces the uniqueness constraint that is needed.
The weird bit is that the (Synthetic ID, Class) data in Elements_Xref must match the same columns in Elements, even though the Synthetic ID is the primary key of Elements.
In IBM Informix Dynamic Server, you can achieve this:
CREATE TABLE elements
(
class CHAR(1) NOT NULL,
synthetic_id SERIAL NOT NULL PRIMARY KEY,
UNIQUE(class, synthetic_id)
);
CREATE TABLE elements_xref
(
class CHAR(1) NOT NULL,
synthetic_id INTEGER NOT NULL REFERENCES elements(synthetic_id),
FOREIGN KEY (class, synthetic_id) REFERENCES elements(class, synthetic_id),
real_id CHAR(5) NOT NULL,
PRIMARY KEY (class, real_id)
);
I would:
Create a UNIQUE constraint on Elements(Synthetic ID, Class)
Add Class column to Elements_Xref
Add a FOREIGN KEY constraint on Elements_Xref table, referring to (Synthetic ID, Class)
At this point we know for sure that Elements_Xref.Class always matches Elements.Class.
Now we need to implement "unique when not null" logic. Follow the link and scroll to section "Use Computed Columns to Implement Complex Business Rules":
Indexes on Computed Columns: Speed Up Queries, Add Business Rules
Alternatively, you can create an indexed view on (Class, RealID) with WHERE RealID IS NOT NULL in its WHERE clause - that will also enforce "unique when not null" logic.
Create an indexed view for Elements_Xref with Where Real_Id Is Not Null and then create a unique index on that view
Create View Elements_Xref_View With SchemaBinding As
Select Elements.Class, Elements_Xref.Real_Id
From Elements_Xref
Inner Join Element On Elements.Synthetic_Id = Elements_Xref.Synthetic_Id
Where Real_Id Is Not Null
Go
Create Unique Clustered Index Elements_Xref_Unique_Index
On Elements_Xref_View (Class, Real_Id)
Go
This serves no other purpose other than simulating a unique index that treats nulls properly i.e. null != null
You can
Create a view from the the result set of joining Elements_Xref and Elements together on Synthetic ID
add a unique constraint on class, and [Real ID]. In other news, this is also how you do functional indexes in MSSQL, by indexing views.
Here is some sql:
CREATE VIEW unique_const_view AS
SELECT e.[Synthetic ID], e.Class, x.[Real ID]
FROM Elements AS e
JOIN [Elements_Xref] AS x
ON e.[Synthetic ID] = x.[Synthetic ID]
CREATE UNIQUE INDEX unique_const_view_index ON unique_const_view ( Class, [Real ID] );
Now, apparently, unbeknownst to myself this solution doesn't work in Microsoft-land-place because with MS SQL Server duplicate nulls will violate a UNIQUE constraint: this is against the SQL spec. This is where the problem is discussed about.
This is the Microsoft workaround:
create unique nonclustered index idx on dbo.DimCustomer(emailAddress)
where EmailAddress is not null;
Not sure if that is 2005, or just 2008.
I think a trigger is your best option. Constraints can't cross to other tables to get information. Same thing with a unique index (although I suppose a materialized view with an index might be possible), they are unique within the table. When you put the trigger together, remember to do it in a set-based fashion not row-by-row and test with a multi-row insert and multi-row update where the real key is repeated in the dataset.
I don't think either of your two reasons are an obstacle to putting Real ID in Elements. If a given element has 0 or 1 Real IDs (but never more than 1), it should absolutely be in the Elements table. This would then allow you to constrain uniqueness within Class (I think).
Could you expand on your two reasons not to do this?
Create a new table real_elements with fields Real ID, Class and Synthetic ID with a primary key of Class, RealId and add elements when you actually add a RealID
This constrains Real IDs to be unique for a class and gives you a way to match a class and real ID to the synthetic ID
As for Real ID being a foreign key do you mean that if it is in two classes then the data keyed off it will be the same. If so the add another table with key Real Id. This key is then a foreign key into real_elements and any other table needing real ID as foreign key