I noticed in one of my exercises that an Order (Attributes OrderID, description...etc) requires a buyer, seller, and an account number which are stored as other entities. I'm wondering why Order is stored as an entity rather than an associative entity.
The idea of an associative entity is it is something we normally wouldn’t identify as as entity but which we need in order to link things together, see https://en.m.wikipedia.org/wiki/Associative_entity
Here the Order is a business entity in itself so the term wouldn’t apply. Entities can be artifacts of business processes as well as concrete things.
In all implementations of SQL (that I know of), there isn't a separate type of entity for "Associative Entities". They are simply regular entities with Foreign Key constraints.
There's quite a bit of information that can be "lost" in translation when going from something like an ER Diagram to actual DB schemas, but all you can do is try your best to reconcile the two.
Edit
Terminology changes too: Entities become Tables, and Attributes become Columns
Related
Problem description
I am currently working on a project which requires a relational database for storage.
After thinking about the data and its relations for a while I ran into a quite repetitive problem:
I encountered a common data schema for entity A which contains some fields e.g. name, description, value. This entity is connected with entity B in multiple n-1 relations. So entity B has n entities A in relation rel1 and n entities A in relation rel2.
Now I am trying to break down this datamodel into a schema for a relational database (e.g. Postgres, MySQL).
After some research, I have not really found "the best" solution for this particular problem.
Some similar questions I have found so far:
Stackoverflow
DBA Stackexchange
My ideas
So I have thought about possible solutions which I am going to present here:
1. Duplicate table
The relationship from entity B to entity A has a certain meaning to it. So it is possible to create multiple tables (1 per relationship). This would solve all immediate problems but essentially duplicate the tables which means that changes now have to be reflected to multiple tables (e.g. a new column).
2. Introduce a type column
Instead of multiple relationships, I could just say "Entity B is connected with n entity A". Additionally, I would add a type column that then tells me to which relation entity A belongs. I am not exactly sure how this is represented with common ORMs like Spring-Hibernate and if this introduces additional problems that I am currently unaware of.
3. Abstract the common attributes of entity A
Another option is to create a ADetails entity, which bundles all attributes of entity A.
Then I would create two entities that represent each relationship and which are connected to the ADetails entity in a 1-to-1 relationship. This would solve the interpretation problem of the foreign key but might be too much overhead.
My Question
In the context of a medium-large-sized project, are any of these solutions viable?
Are there certain Cons that rule out one particular approach?
Are there other (better) options I haven't thought about?
I appreciate any help on this matter.
Edit 1 - PPR (Person-Party-Role)
Thanks for the suggestion from AntC. PPR Description
I think the described situation matches my problem.
Let's break it down:
Entity B is an event. There exists only one event for the given participants to make this easier. So the relationship from event to participant is 1-n.
Entity A can be described as Groups, People, Organization but given my situation they all have the same attributes. Hence, splitting them up into separate tables felt like the wrong idea.
To explain the situation with the class diagram:
An Event (Entity B) has a collection of n Groups (Entity A), n People (Entity A) and n Organizations (Entity A).
If I understand correctly the suggestion is the following:
In my case the relationship between Event and Participant is 1-n
The RefRoles table represents the ParticipantType column that descibes to which relationship the Participant belongs (is it a customer or part of the service for the event for example)
Because all my Groups, People and Organizations have the same attributes the only table required at this point is the Participant table
If there are individual attributes in the future I would introduce a new table (e.g. People) that references the Participant in a 1-1 relationship.
If there are multiple tables going to be added, the foreign key of the multiple 1-1 relationship is mutually exclusive (so there can only be one Group/Person/Organization for a participant)
Solution suggested by AntC and Christian Beikov
Splitting up the tables does make sense while keeping the common attributes in one table.
At the moment there are no individual attributes but the type column is not required anymore because the foreign keys can be used to see which relationship the entity belongs to.
I have created a small example for this:
There exist 3 types (previously type column) of people for an event: Staff, VIP, Visitor
The common attributes are mapped in a 1-1-relationship to the person table.
To make it simple: Each Person (Staff, VIP, Visitor) can only participate in one event. (Would be n-m-relationship in a more advanced example)
The database schema would be the following:
This approach is better than the type column in my opinion.
It also solves having to interprete the entity based on its type in the application later on. It is also possible to resolve a type column in an ORM (see this question) but this approach avoids the struggle if the ORM you are using does not support resolving it.
IMO since you already use dedicated terms for these objects, they probably will diverge and splitting up a table afterwards is quite some work, also on the code side, so I would suggest you map dedicated entities/tables from the beginning.
I have to design a generic entity that would be able to refer to variated other entities.
In my example, that would be a commentary entity inside a web application. You could post commentaries on to users, classifieds, articles, varieties (botanical ones), and so on.
So that entity would be made like this:
As a matter of fact, the design (kind of) pattern would be this one:
What are the pros and cons of this kind of pattern?
What I see is:
Pros
It decreases the number of entities if the concept is the same (commentaries for example);
You can therefore easily manipulate heterogeneous objects;
You can aggregate these objects easily (e.g. this user's last commentaries in the whole site, presented easily in a same thread);
Cons
This allows you to fall in the ugly (you use it outrageously and your database and source code are ugly);
There is no control in the database, and this one must therefore be done inside the application code.
What are the performances impacts?
Conclusion
Is this kind of pattern suitable for a relational database? How can we do then?
Thank you by advance.
One more con :
This scheme relies on a mapping between values and names for the "entities" referred to by those values. Think of all the fun you'll have resolving issues that in the TEST system, the ORDER entity has number 734 but in production, it has number 256. You can use the entity names themselves as the values of your entity_id stuff, but you will never be able to avoid hardcoding values for them in your programs (or, say, in view definitions) anyway. Thereby defeating whatever advantage it was you thought you could win.
This kind of scheme is a disease mostly suffered by OO programmers. They see structures that are largely similar and they have this instinctive reflex "I must find a way to resue the existing thing for this". Forgetting that database design is not program design.
EDIT
(if it wasn't clear, this means my answer to your question "Is this kind of pattern suitable for a relational database?" is a principled "NO".)
This is the classic Polymorphic Association anti-pattern. There are a number of possible solutions:
1) Exclusive Arcs e.g. for the Commentary entity
Id
User_Id
Classified_Id
Article_Id
Variety_Id
Where User_Id, Classified_Id, Article_Id and Variety_Id are nullable and exactly one must be not null.
2) Reverse the Relationship e.g remove the Target_Entity and Target_Entity_Id from the Commentary entity and create four new entities
User_Commentary
Commentary_Id
User_Id
Classified_Commentary
Commentary_Id
Classified_Id
Article_Commentary
Commentary_Id
Article_Id
Variety_Commentary
Commentary_Id
Variety_Id
Where Commentary_Id is unique and relates to the Id in Commentary.
3) Create a super-type entity for User, Classified, Article and Variety and have the Commentary entity reference the unique attribute of this new entity.
You would need to decide which of these approaches you feel is most appropriate in your specific situation.
I have an entity called events can be either one or several types.
My question is how do the relationship because a event can be more than one type. These can't be an attribute nor an inheritance relationship. Could be a weak-strong entity types as events not exist without the body but I have not clear.
example:
Event is a workshop and a conference.
Thanks for yours help.
A rule of thumb is that you should always keep in mind this when designing a database: each entity with its own table. This is the base of a good Relational Database Design.
If you can't use a flag (attribute) neither use inheritance, then you'd have to artificially keep two entities, one for workshop and another for conference, keep some FK for each one and possibly a flag/trigger to ensure you're not using an FK for both at the same time.
OR
You could use some intermediary entity, but I understand this would be some sort of inheritance, because this proxy entity would act like a "super" entity for both workshop and conference (like an "event"...)
The first option is bad in terms of maintenance, I'd not recommend you that.
The second option is IMO more "intuitive".
If both entities are too close one another and will be linked to several other entities in the same way, I think you could just use a flag to differentiate them, with the risk that in the future, if anything changes to one of them, you have to refactor your schema, which is usually painful and risky. It's a premature optimization, and as we know, premature optimization is the root of all evil. So keeping them in separate entities may be a good option.
If the list of all possible Types that can be associated with an Entity are defined in a list (e.g. within a single table), then you can use the standard many-to-many relationship pattern:
TABLE EVENTS
EventId, primary key
TABLE TYPES
TypeId, primary key
TABLE EVENTTYPES
EventId, foreign key to Events
TypeId, foreign key to Types
...with the primray key on {EventId, Typeid}
This pattern was usefull: multivalued attributes.
http://www.tomjewett.com/dbdesign/dbdesign.php?page=hobbies.php
Every time I program I recognize this relationship between classes and tables, or am I imagining it.
You can have a class per database table or a table per class i.e. :
tables: customer, products, order.
classes: customer, products, order, may have methods such as addRecord, deleteRecord, updateRecord.
what is this called? Object-Relational? I am not a DBA.
It all depends on the type of database you're using. If you're using an object oriented database (OODB), then there is no relationship, as the objects and the persisted data are the same thing. For example, if you have a Customer class, and you save it in an OODB, then that instance of the customer is what is stored in the DB.
If you are using a relational database, then the class instances, and the persisted representation of them in the DB, can be the same thing, but many times they aren't. This is because most folks use normalization to represent their data in an efficient way (in a relational DB). This means, instead of having a table per class, you can have a class represented by more than one table. In the Customer example, the tables might now be Customer (with Name, date of birth, and other properties), and Order (with order pointing to products in yet another table). The reason for this has to do with cardinality, and the ability for Customers to have more than one order. When your business logic needs this information from the DB, the data access layer's job is to map the data (called ORM) from the DB into your classes.
If you are using yet another type of DB, then there will be a different relationship between the classes (domain model) and what's persisted in the DB.
But, as far as having a name for this relationship? No, there is no name.
In additon to Bob's answer, the following.
In object modeling, the relationship between classes and subclasses is taken care of by inheritance, and object modelers know how to use inheritance to good advantage. The relational data model and by extension the SQL databases do not implement inheritance for you. You have to design tables to give you some of the same results.
In ER (Entity-Relationship) modeling, the corresponding concept is called generalization/specialization. This tells you how to model a class/subclass relationship, but it doesn't tell you how to design the tables when you go to build your database.
There are three techniques that are pretty well understood that can be really helpful when dealing with classes and subclasses. Here are their tags: single-table-inheritance class-table-inheritance shared-primary-key. Unfortunately, many tutorials on database design never cover these techniques. They can be enormously useful to people who know object modeling and want to come up to speed on relational modeling.
There are couples of questions around asking for difference / explanation on identifying and non-identifying relationship in relationship database.
My question is, can you think of a simpler term for these jargons? I understand that technical terms have to be specific and unambiguous though. But having an 'alternative name' might help students relate more easily to the concept behind.
We actually want to use a more layman term in our own database modeling tool, so that first-time users without much computer science background could learn faster.
cheers!
I often see child table or dependent table used as a lay term. You could use either of those terms for a table with an identifying relationship
Then say a referencing table is a table with a non-identifying relationship.
For example, PhoneNumbers is a child of Users, because a phone number has an identifying relationship with its user (i.e. the primary key of PhoneNumbers includes a foreign key to the primary key of Users).
Whereas the Users table has a state column that is a foreign key to the States table, making it a non-identifying relationship. So you could say Users references States, but is not a child of it per se.
I think belongs to would be a good name for the identifying relationship.
A "weak entity type" does not have its own key, just a "partial key", so each entity instance of this weak entity type has to belong to some other entity instance so it can be identified, and this is an "identifying relationship". For example, a landlord could have a database with apartments and rooms. A room can be called kitchen or bathroom, and while that name is unique within an apartment, there will be many rooms in the database with the name kitchen, so it is just a partial key. To uniquely identify a room in the database, you need to say that it is the kitchen in this particular apartment. In other words, the rooms belong to apartments.
I'm going to recommend the term "weak entity" from ER modeling.
Some modelers conceptualize the subject matter as being made up of entities and relationships among entities. This gives rise to Entity-Relationship Modeling (ER Modeling). An attribute can be tied to an entity or a relationship, and values stored in the database are instances of attributes.
If you do ER modeling, there is a kind of entity called a "weak entity". Part of the identity of a weak entity is the identity of a stronger entity, to which the weak one belongs.
An example might be an order in an order processing system. Orders are made up of line items, and each line item contains a product-id, a unit-price, and a quantity. But line items don't have an identifying number across all orders. Instead, a line item is identified by {item number, order number}. In other words, a line item can't exist unless it's part of exactly one order. Item number 1 is the first item in whatever order it belongs to, but you need both numbers to identify an item.
It's easy to turn an ER model into a relational model. It's also easy for people who are experts in the data but know nothing about databases to get used to an ER model of the data they understand.
There are other modelers who argue vehemently against the need for ER modeling. I'm not one of them.
Nothing, absolutely nothing in the kind of modeling where one encounters things such as "relationships" (ER, I presume) is "technical", "precise" or "unambiguous". Nor can it be.
A) ER modeling is always and by necessity informal, because it can never be sufficient to capture/express the entire definition of a database.
B) There are so many different ER dialects out there that it is just impossible for all of them to use exactly the same terms with exactly the same meaning. Recently, I even discovered that some UK university that teaches ER modeling, uses the term "entity subtype" for the very same thing that I always used to name "entity supertype", and vice-versa !
One could use connection.
You have Connection between two tables, where the IDs are the same.
That type of thing.
how about
Association
Link
Correlation