I'm new to databases and trying to understand why a junction or association table is needed when creating a many-to-many relationship.
Most of what I'm finding on Stackoverflow and elsewhere describe it in either highly technical relational theory terms or it's just described as 'that's the way it's done' without qualifying why.
Are there any relational database designs out there that support having a many-to-many relationship without the use of an association table? Why is it not possible to have, for example, a column on on table that holds the relationships to another and vice a versa.
For example, a Course table that holds a list of courses and a Student table that holds a bunch of student info — each course can have many students and each student can take many classes.
Why is it not possible to have a column on each row in either table (possibly in csv format) that contains the relationships to the others in a list or something similar?
In a relational database, no column holds more than a single value in each row. Therefore, you would never store data in a "CSV format" -- or any other multiple value system -- in a single column in a relational database. Making repeated columns that hold instances of the same item (Course1, Course2, Course3, etc) is also not allowed. This is the very first rule of relational database design and is referred to as First Normal Form.
There are very good reasons for the existence of these rules (it is enormously easier to verify, constrain, and query the data) but whether or not you believe in the benefits the rules are, none-the-less, part of the definition of relational databases.
I do not know the answer to your question, but I can answer a similar question: Why do we use a junction table for many-to-many relationships in databases?
First, if the student table keeps track of which courses the student is in and the course keeps track of which students are in it, then we have duplication. This can lead to problems. What if a student knows it is in a course, but the course doesn't know that it has that student. Every time you made a course change you would have to make sure to change it in both tables. Inevitably this will not happen every time and the data will become inconsistent.
Second, where would we store this information? A list is not a possible type for a field in a database. So do we put a course column in the student table? No, because that would only allow each student to take one course, a many-to-one relationship from students to courses. Do we put a student column in the courses table? No, because then we have one student in each course.
What does work is having a new table that has one student and one course per row. This tells us that a student is in a class without duplicating any data.
"Junction tables" come from ER/ORM presentations/methods/products that don't really understand the relational model.
In the relational model (and in original ER information modeling) application relationships are represented by relations/tables. Each table holds tuples of values that are in that relationship to each other, ie that are so related, ie that satisfy that relationship, ie that participate in the relationship.
A relationship is expressed independently of any particular situation as a predicate, a fill-in-the-(named-)blanks statement. Rows that fill in the named blanks to give a true statement from the predicate in a particular situation go in the table. We pick sufficient predicates (hence base tables) to describe every situation. Both many-to-1 and many-to-many application relationships get tables.
The reason why you don't see a lot of many-to-many relationships along with columns about the participants rather than about their participation in the relationship is that such tables are better split into ones about the participants and one for the relationship. Eg columns in a many-to-many table that are about participants 1. can't say anything about entities that don't participate and 2. say the same thing about an entity every time it participates. Information modeling techniques that focus on identifying independent entity types first then relationships between them tend to lead to designs with few such problems. The reason why you don't see many-to-many relationships in two tables is that that is redundant and susceptible to the error of the tables disagreeing. The problem with collection-valued columns (sequences/lists/arrays) is that you cannot generically query about their parts using usual query notation and implementation because the DBMS doesn't see the parts organized into a table.
See this recent answer or this one.
I have been asked to add a new address book table to our database (SQL Server 2012).
To simplify the related part of the database, there are three tables each linked to each other in a one to many fashion: Company (has many) Products (has many) Projects and the idea is that one or many addresses will be able to exist at any one of these levels. The thinking is that in the front-end system, a user will be able to view and select specific addresses for the project they specify and more generic addresses relating to its parent product and company.
The issue now if how best to model this in the database.
I have thought of two possible ideas so far so wonder if anyone has had a similar type of relationship to model themselves and how they implemented it?
Idea one:
The new address table will additionally contain three fields: companyID, productID and projectID. These fields will be related to the relevant tables and be nullable to represent company and product level addresses. e.g. companyID 2, productID 1, projectID NULL is a product level address.
My issue with this is that I am storing the relationship information in the table so if a project is ever changed to be related to a different product, the data in this table will be incorrect. I could potentially NULL all but the level I am interested in but this will make getting parent addresses a little harder to get
Idea two:
On the address table have a typeID and a genericID. genericID could contain the IDs from the Company, Product and Project tables with the typeID determining which table it came from. I am a little stuck how to set up the necessary constraints to do this though and wonder if this is going to get tricky to deal with in the future
Many thanks,
I will suggest using Idea one and preventing Idea two.
Second Idea is called Polymorphic Association anti pattern
Objective: Reference Multiple Parents
Resulting side effect: Using dual-purpose foreign key will violating first normal form (atomic issue), loosing referential integrity
Solution: Simplify the Relationship
The simplification of the relationship could be obtained in two ways:
Having multiple null-able forging keys (idea number 1): That will be
simple and applicable if the tables(product, project,...) that using
the relation are limited. (think about when they grow up to more)
Another more generic solution will be using inheritance. Defining a
new entity as the base table for (product, project,...) to satisfy
Addressable. May naming it organization-unit be more rational. Primary key of this organization_unit table will be the primary key of (product, project,...). Other collections like Address, Image, Contract ... tables will have a relation to this base table.
It sounds like you could use Junction tables http://en.wikipedia.org/wiki/Junction_table.
They will give you the flexibility you need to maintain your foreign key restraints, as well as share addresses between levels or entities if that is desired.
One for Company_Address, Product_Address, and Project_Address
Inventory Items :
Paper Size
-----
A0
A1
A2
etc
Paper Weight
------------
80gsm
150gsm etc
Paper mode
----------
Colour
Bw
Paper type
-----------
glass
silk
normal
Tabdividers and tabdivider Type
--------
Binding and Binding Types
--
Laminate and laminate Types
--
Such Inventory items and these all needs to be stored in invoice table
How do you store them in Database using proper RDBMS.
As per my opinion for each list a master table and retrieval with JOINS. However this may be a little bit complex adding too many tables into the database.
This normalisation is having bit of problem when storing all this information against a Invoice. This is causing too many columns in invoice table.
Other way putting all of them into a one table with more columns and then each row will be a combination of them.. (hacking algorithm 4 list with 4 items over 24 records which will have reference ID).
Which one do you think the best and why!!
Your initial idea is correct. And anyone claiming that four tables is "a little bit complex" and/or "too many tables" shouldn't be doing database work. This is what RDBMS's are designed (and tuned) to do.
Each of these 4 items is an individual property of something so they can't simply be put, as is, into a table that merges them. As you had thought, you start with:
PaperSize
PaperWeight
PaperMode
PaperType
These are lookup tables and hence should have non-auto-incrementing ID fields.
These will be used as Foreign Key fields for the main paper-based entities.
Or if they can only exist in certain combinations, then there would need to be a relationship table to capture/manage what those valid combinations are. But those four paper "properties" would still be separate tables that Foreign Key to the relationship table. Some people would put an separate ID field on that relationship table to uniquely identify the combination via a single value. Personally, I wouldn't do that unless there was a technical requirement such as Replication (or some other process/feature) that required that each table had a single-field key. Instead, I would just make the PK out of the four ID fields that point to those paper "property" lookup tables. Then those four fields would still go into any paper-based entities. At that point the main paper entity tables would look about the same as they would if there wasn't the relationship table, the difference being that instead of having 4 FKs of a single ID field each, one to each of the paper "property" tables, there would be a single FK of 4 ID fields pointing back to the PK of the relationship table.
Why not jam everything into a single table? Because:
It defeats the purpose of using a Relational Database Management System to flatten out the data into a non-relational structure.
It is harder to grow that structure over time
It makes finding all paper entities of a particular property clunkier
It makes finding all paper entities of a particular property slower / less efficient
maybe other reasons?
EDIT:
Regarding the new info (e.g. Invoice Table, etc) that wasn't in the question when I was writing the above, that should be abstracted via a Product/Inventory table that would capture these combinations. That is what I was referring to as the main paper entities. The Invoice table would simply refer to a ProductID/InventoryID (just as an example) and the Product/Inventory table would have these paper property IDs. I don't see why these properties would be in an Invoice table.
EDIT2:
Regarding the IDs of the "property" lookup tables, one reason that they should not be auto-incrementing is that their values should be taken from Enums in the app layer. These lookup tables are just a means of providing a "data dictionary" so that the database layer can have insight into what these values mean.
Sorry for that noob question but is there any real needs to use one-to-one relationship with tables in your database? You can implement all necessary fields inside one table. Even if data becomes very large you can enumerate column names that you need in SELECT statement instead of using SELECT *. When do you really need this separation?
1 to 0..1
The "1 to 0..1" between super and sub-classes is used as a part of "all classes in separate tables" strategy for implementing inheritance.
A "1 to 0..1" can be represented in a single table with "0..1" portion covered by NULL-able fields. However, if the relationship is mostly "1 to 0" with only a few "1 to 1" rows, splitting-off the "0..1" portion into a separate table might save some storage (and cache performance) benefits. Some databases are thriftier at storing NULLs than others, so a "cut-off point" where this strategy becomes viable can vary considerably.
1 to 1
The real "1 to 1" vertically partitions the data, which may have implications for caching. Databases typically implement caches at the page level, not at the level of individual fields, so even if you select only a few fields from a row, typically the whole page that row belongs to will be cached. If a row is very wide and the selected fields relatively narrow, you'll end-up caching a lot of information you don't actually need. In a situation like that, it may be useful to vertically partition the data, so only the narrower, more frequently used portion or rows gets cached, so more of them can fit into the cache, making the cache effectively "larger".
Another use of vertical partitioning is to change the locking behavior: databases typically cannot lock at the level of individual fields, only the whole rows. By splitting the row, you are allowing a lock to take place on only one of its halfs.
Triggers are also typically table-specific. While you can theoretically have just one table and have the trigger ignore the "wrong half" of the row, some databases may impose additional limits on what a trigger can and cannot do that could make this impractical. For example, Oracle doesn't let you modify the mutating table - by having separate tables, only one of them may be mutating so you can still modify the other one from your trigger.
Separate tables may allow more granular security.
These considerations are irrelevant in most cases, so in most cases you should consider merging the "1 to 1" tables into a single table.
See also: Why use a 1-to-1 relationship in database design?
My 2 cents.
I work in a place where we all develop in a large application, and everything is a module. For example, we have a users table, and we have a module that adds facebook details for a user, another module that adds twitter details to a user. We could decide to unplug one of those modules and remove all its functionality from our application. In this case, every module adds their own table with 1:1 relationships to the global users table, like this:
create table users ( id int primary key, ...);
create table users_fbdata ( id int primary key, ..., constraint users foreign key ...)
create table users_twdata ( id int primary key, ..., constraint users foreign key ...)
If you place two one-to-one tables in one, its likely you'll have semantics issue. For example, if every device has one remote controller, it doesn't sound quite good to place the device and the remote controller with their bunch of characteristics in one table. You might even have to spend time figuring out if a certain attribute belongs to the device or the remote controller.
There might be cases, when half of your columns will stay empty for a long while, or will not ever be filled in. For example, a car could have one trailer with a bunch of characteristics, or might have none. So you'll have lots of unused attributes.
If your table has 20 attributes, and only 4 of them are used occasionally, it makes sense to break the table into 2 tables for performance issues.
In such cases it isn't good to have everything in one table. Besides, it isn't easy to deal with a table that has 45 columns!
If data in one table is related to, but does not 'belong' to the entity described by the other, then that's a candidate to keep it separate.
This could provide advantages in future, if the separate data needs to be related to some other entity, also.
The most sensible time to use this would be if there were two separate concepts that would only ever relate in this way. For example, a Car can only have one current Driver, and the Driver can only drive one car at a time - so the relationship between the concepts of Car and Driver would be 1 to 1. I accept that this is contrived example to demonstrate the point.
Another reason is that you want to specialize a concept in different ways. If you have a Person table and want to add the concept of different types of Person, such as Employee, Customer, Shareholder - each one of these would need different sets of data. The data that is similar between them would be on the Person table, the specialist information would be on the specific tables for Customer, Shareholder, Employee.
Some database engines struggle to efficiently add a new column to a very large table (many rows) and I have seen extension-tables used to contain the new column, rather than the new column being added to the original table. This is one of the more suspect uses of additional tables.
You may also decide to divide the data for a single concept between two different tables for performance or readability issues, but this is a reasonably special case if you are starting from scratch - these issues will show themselves later.
First, I think it is a question of modelling and defining what consist a separate entity. Suppose you have customers with one and only one single address. Of course you could implement everything in a single table customer, but if, in the future you allow him to have 2 or more addresses, then you will need to refactor that (not a problem, but take a conscious decision).
I can also think of an interesting case not mentioned in other answers where splitting the table could be useful:
Imagine, again, you have customers with a single address each, but this time it is optional to have an address. Of course you could implement that as a bunch of NULL-able columns such as ZIP,state,street. But suppose that given that you do have an address the state is not optional, but the ZIP is. How to model that in a single table? You could use a constraint on the customer table, but it is much easier to divide in another table and make the foreign_key NULLable. That way your model is much more explicit in saying that the entity address is optional, and that ZIP is an optional attribute of that entity.
not very often.
you may find some benefit if you need to implement some security - so some users can see some of the columns (table1) but not others (table2)..
of course some databases (Oracle) allow you to do this kind of security in the same table, but some others may not.
You are referring to database normalization. One example that I can think of in an application that I maintain is Items. The application allows the user to sell many different types of items (i.e. InventoryItems, NonInventoryItems, ServiceItems, etc...). While I could store all of the fields required by every item in one Items table, it is much easier to maintain to have a base Item table that contains fields common to all items and then separate tables for each item type (i.e. Inventory, NonInventory, etc..) which contain fields specific to only that item type. Then, the item table would have a foreign key to the specific item type that it represents. The relationship between the specific item tables and the base item table would be one-to-one.
Below, is an article on normalization.
http://support.microsoft.com/kb/283878
As with all design questions the answer is "it depends."
There are few considerations:
how large will the table get (both in terms of fields and rows)? It can be inconvenient to house your users' name, password with other less commonly used data both from a maintenance and programming perspective
fields in the combined table which have constraints could become cumbersome to manage over time. for example, if a trigger needs to fire for a specific field, that's going to happen for every update to the table regardless of whether that field was affected.
how certain are you that the relationship will be 1:1? As This question points out, things get can complicated quickly.
Another use case can be the following: you might import data from some source and update it daily, e.g. information about books. Then, you add data yourself about some books. Then it makes sense to put the imported data in another table than your own data.
I normally encounter two general kinds of 1:1 relationship in practice:
IS-A relationships, also known as supertype/subtype relationships. This is when one kind of entity is actually a type of another entity (EntityA IS A EntityB). Examples:
Person entity, with separate entities for Accountant, Engineer, Salesperson, within the same company.
Item entity, with separate entities for Widget, RawMaterial, FinishedGood, etc.
Car entity, with separate entities for Truck, Sedan, etc.
In all these situations, the supertype entity (e.g. Person, Item or Car) would have the attributes common to all subtypes, and the subtype entities would have attributes unique to each subtype. The primary key of the subtype would be the same as that of the supertype.
"Boss" relationships. This is when a person is the unique boss or manager or supervisor of an organizational unit (department, company, etc.). When there is only one boss allowed for an organizational unit, then there is a 1:1 relationship between the person entity that represents the boss and the organizational unit entity.
The main time to use a one-to-one relationship is when inheritance is involved.
Below, a person can be a staff and/or a customer. The staff and customer inherit the person attributes. The advantage being if a person is a staff AND a customer their details are stored only once, in the generic person table. The child tables have details specific to staff and customers.
In my time of programming i encountered this only in one situation. Which is when there is a 1-to-many and an 1-to-1 relationship between the same 2 entities ("Entity A" and "Entity B").
When "Entity A" has multiple "Entity B" and "Entity B" has only 1 "Entity A"
and
"Entity A" has only 1 current "Entity B" and "Entity B" has only 1 "Entity A".
For example, a Car can only have one current Driver, and the Driver can only drive one car at a time - so the relationship between the concepts of Car and Driver would be 1 to 1. - I borrowed this example from #Steve Fenton's answer
Where a Driver can drive multiple Cars, just not at the same time. So the Car and Driver entities are 1-to-many or many-to-many. But if we need to know who the current driver is, then we also need the 1-to-1 relation.
Another use case might be if the maximum number of columns in the database table is exceeded. Then you could join another table using OneToOne
I'm developing an Asset management application.
Looking through the excel tracker that was being used previously, I was able to identify some attributes that were common to all categories of assets (basically non-technical attributes such as Purchase Order No. , Warranty Info etc.) for which I think I will make a separate table.
But when storing technical-attributes, there are many categories of assets for which I need only one or two additional attributes to be stored.
Should a make a single table for all these attributes and store NULLs wherever applicable or should I make a separate table each category containing just the asset ID and the addition columns? Which approach is better/more pragmatic?
Is cluttering the database with too many tables ok? I have around 10 such categories.
There are 3 known approaches to this:
Single table
In this model, you have a single table with all known columns, and allow them to be null for types that don't have that attribute. This gives you a simple database, and fairly simple SQL, but doesn't allow support for common features that relational databases give you, like insisting on non-null columns for a data type, or creating unique indices where that makes sense.
It also tends to lead to messy SQL, with developers forgetting over time what columns mean, so you could get a column being used for multiple purposes.
It does make it easy to join to other tables - so if you have an asset and a purchase related to that asset, the "purchase" table joins to the "asset" table on "assetID".
Table per subtype
In this case, you build a table for each subtype, and enforce the data characteristics of that subtype with not null, unique etc.
This creates a clearer separation of subtypes, and is less likely to degrade into big ball of mud, but makes joins very hard - to join from "purchase" to "asset", you have to know which table holds that particular asset.
Common table for common fields, table per subtype
In this model, you have a single table for the fields that are common between subtypes - you say you've identified this already - and have further tables for each subtype to store the unique attributes.
This solves the joining problem between "asset" and "purchase", keeps the data pretty self-describing.
It does mean client logic needs to implement the "join asset_master to asset_subtype" issue.
I prefer option 3 - it's the best trade-off between maintainability and managability.
Databases should be able to handle lots of columns and lots of tables, so both approaches should work from that perspective.
If you don't have any additional requirements, I'd use the single table approach. It is the easiest, and the only thing you are loosing is the ability to put not null constraints on the fields that exist only form some categories