Related
It is safe to say that the EAV/CR database model is bad. That said,
Question: What database model, technique, or pattern should be used to deal with "classes" of attributes describing e-commerce products which can be changed at run time?
In a good E-commerce database, you will store classes of options (like TV resolution then have a resolution for each TV, but the next product may not be a TV and not have "TV resolution"). How do you store them, search efficiently, and allow your users to setup product types with variable fields describing their products? If the search engine finds that customers typically search for TVs based on console depth, you could add console depth to your fields, then add a single depth for each tv product type at run time.
There is a nice common feature among good e-commerce apps where they show a set of products, then have "drill down" side menus where you can see "TV Resolution" as a header, and the top five most common TV Resolutions for the found set. You click one and it only shows TVs of that resolution, allowing you to further drill down by selecting other categories on the side menu. These options would be the dynamic product attributes added at run time.
Further discussion:
So long story short, are there any links out on the Internet or model descriptions that could "academically" fix the following setup? I thank Noel Kennedy for suggesting a category table, but the need may be greater than that. I describe it a different way below, trying to highlight the significance. I may need a viewpoint correction to solve the problem, or I may need to go deeper in to the EAV/CR.
Love the positive response to the EAV/CR model. My fellow developers all say what Jeffrey Kemp touched on below: "new entities must be modeled and designed by a professional" (taken out of context, read his response below). The problem is:
entities add and remove attributes weekly (search keywords dictate future attributes)
new entities arrive weekly (products are assembled from parts)
old entities go away weekly (archived, less popular, seasonal)
The customer wants to add attributes to the products for two reasons:
department / keyword search / comparison chart between like products
consumer product configuration before checkout
The attributes must have significance, not just a keyword search. If they want to compare all cakes that have a "whipped cream frosting", they can click cakes, click birthday theme, click whipped cream frosting, then check all cakes that are interesting knowing they all have whipped cream frosting. This is not specific to cakes, just an example.
There's a few general pros and cons I can think of, there are situations where one is better than the other:
Option 1, EAV Model:
Pro: less time to design and develop a simple application
Pro: new entities easy to add (might even
be added by users?)
Pro: "generic" interface components
Con: complex code required to validate simple data types
Con: much more complex SQL for simple
reports
Con: complex reports can become almost
impossible
Con: poor performance for large data sets
Option 2, Modelling each entity separately:
Con: more time required to gather
requirements and design
Con: new entities must be modelled and
designed by a professional
Con: custom interface components for each
entity
Pro: data type constraints and validation simple to implement
Pro: SQL is easy to write, easy to
understand and debug
Pro: even the most complex reports are relatively simple
Pro: best performance for large data sets
Option 3, Combination (model entities "properly", but add "extensions" for custom attributes for some/all entities)
Pro/Con: more time required to gather requirements and design than option 1 but perhaps not as much as option 2 *
Con: new entities must be modelled and designed by a professional
Pro: new attributes might be easily added later on
Con: complex code required to validate simple data types (for the custom attributes)
Con: custom interface components still required, but generic interface components may be possible for the custom attributes
Con: SQL becomes complex as soon as any custom attribute is included in a report
Con: good performance generally, unless you start need to search by or report by the custom attributes
* I'm not sure if Option 3 would necessarily save any time in the design phase.
Personally I would lean toward option 2, and avoid EAV wherever possible. However, for some scenarios the users need the flexibility that comes with EAV; but this comes with a great cost.
It is safe to say that the EAV/CR database model is bad.
No, it's not. It's just that they're an inefficient usage of relational databases. A purely key/value store works great with this model.
Now, to your real question: How to store various attributes and keep them searchable?
Just use EAV. In your case it would be a single extra table. index it on both attribute name and value, most RDBMs would use prefix-compression to on the attribute name repetitions, making it really fast and compact.
EAV/CR gets ugly when you use it to replace 'real' fields. As with every tool, overusing it is 'bad', and gives it a bad image.
// At this point, I'd like to take a moment to speak to you about the Magento/Adobe PSD format.
// Magento/PSD is not a good ecommerce platform/format. Magento/PSD is not even a bad ecommerce platform/format. Calling it such would be an
// insult to other bad ecommerce platform/formats, such as Zencart or OsCommerce. No, Magento/PSD is an abysmal ecommerce platform/format. Having
// worked on this code for several weeks now, my hate for Magento/PSD has grown to a raging fire
// that burns with the fierce passion of a million suns.
http://code.google.com/p/xee/source/browse/trunk/XeePhotoshopLoader.m?spec=svn28&r=11#107
The internal models are wacky at best, like someone put the schema into a boggle game, sealed that and put it in a paint shacker...
Real world: I'm working on a midware fulfilment app and here are one the queries to get address information.
CREATE OR REPLACE VIEW sales_flat_addresses AS
SELECT sales_order_entity.parent_id AS order_id,
sales_order_entity.entity_id,
CONCAT(CONCAT(UCASE(MID(sales_order_entity_varchar.value,1,1)),MID(sales_order_entity_varchar.value,2)), "Address") as type,
GROUP_CONCAT(
CONCAT( eav_attribute.attribute_code," ::::: ", sales_order_entity_varchar.value )
ORDER BY sales_order_entity_varchar.value DESC
SEPARATOR '!!!!!'
) as data
FROM sales_order_entity
INNER JOIN sales_order_entity_varchar ON sales_order_entity_varchar.entity_id = sales_order_entity.entity_id
INNER JOIN eav_attribute ON eav_attribute.attribute_id = sales_order_entity_varchar.attribute_id
AND sales_order_entity.entity_type_id =12
GROUP BY sales_order_entity.entity_id
ORDER BY eav_attribute.attribute_code = 'address_type'
Exacts address information for an order, lazily
--
Summary: Only use Magento if:
You are being given large sacks of money
You must
Enjoy pain
I'm surprised nobody mentioned NoSQL databases.
I've never practiced NoSQL in a production context (just tested MongoDB and was impressed) but the whole point of NoSQL is being able to save items with varying attributes in the same "document".
Where performance is not a major requirement, as in an ETL type of application, EAV has another distinct advantage: differential saves.
I've implemented a number of applications where an over-arching requirement was the ability to see the history of a domain object from its first "version" to it's current state. If that domain object has a large number of attributes, that means each change requires a new row be inserted into it's corresponding table (not an update because the history would be lost, but an insert). Let's say this domain object is a Person, and I have 500k Persons to track with an average of 100+ changes over the Persons life-cycle to various attributes. Couple that with the fact that rare is the application that has only 1 major domain object and you'll quickly surmize that the size of the database would quickly grow out of control.
An easy solution is to save only the differential changes to the major domain objects rather than repeatedly saving redundant information.
All models change over time to reflect new business needs. Period. Using EAV is but one of the tools in our box to use; but it should never be automatically classified as "bad".
I'm struggling with the same issue. It may be interesting for you to check out the following discussion on two existing ecommerce solutions: Magento (EAV) and Joomla (regular relational structure):
https://forum.virtuemart.net/index.php?topic=58686.0
It seems, that Magento's EAV performance is a real showstopper.
That's why I'm leaning towards a normalized structure. To overcome the lack of flexibility I'm thinking about adding some separate data dictionary in the future (XML or separate DB tables) that could be edited, and based on that, application code for displaying and comparing product categories with new attributes set would be generated, together with SQL scripts.
Such architecture seems to be the sweetspot in this case - flexible and performant at the same time.
The problem could be frequent use of ALTER TABLE in live environment. I'm using Postgres, so its MVCC and transactional DDL will hopefully ease the pain.
I still vote for modeling at the lowest-meaningful atomic-level for EAV. Let standards, technologies and applications that gear toward certain user community to decide content models, repetition needs of attributes, grains, etc.
If it's just about the product catalog attributes and hence validation requirements for those attributes are rather limited, the only real downside to EAV is query performance and even that is only a problem when your query deals with multiple "things" (products) with attributes, the performance for the query "give me all attributes for the product with id 234" while not optimal is still plenty fast.
One solution is to use the SQL database / EAV model only for the admin / edit side of the product catalog and have some process that denormalizes the products into something that makes it searchable. Since you already have attributes and hence it's rather likely that you want faceting, this something could be Solr or ElasticSearch. This approach avoids basically all downsides to the EAV model and the added complexity is limited to serializing a complete product to JSON on update.
EAV has many drawbacks:
Performance degradation over time
Once the amount of data in the application grows beyond a certain size, the retrieval and manipulation of that data is likely to become less and less efficient.
The SQL queries are very complex and difficult to write.
Data Integrity problems.
You can't define foreign keys for all the fields needed.
You have to define and maintain your own metadata.
I have a slightly different problem: instead of many attributes with sparse values (which is possibly a good reason to use EAV), I want to store something more like a spreadsheet. The columns in the sheet can change, but within a sheet all cells will contain data (not sparse).
I made a small set of tests to benchmark two designs: one using EAV, and the other using a Postgres ARRAY to store cell data.
EAV
Array
Both schemas have indexes on appropriate columns, and the indexes are used by the planner.
It turned out the array-based schema was an order of magnitude faster for both inserts and queries. From quick tests, it seemed that both scaled linearly. The tests aren't very thorough, though. Suggestions and forks welcome - they're under an MIT licence.
I am trying to understand the Single Responsibility principle but I have tough time in grasping the concept. I am reading the book "Design Patterns and Best Practices in Java by Lucian-Paul Torje; Adrian Ianculescu; Kamalmeet Singh ."
In this book I am reading Single responsibility principle chapter ,
where they have a car class as shown below:
They said Car has both Car logic and database operations. In future if we want to change database then we need to change database logic and might need to change also car logic. And vice versa...
The solution would be to create two classes as shown below:
My question is even if we create two classes , let’s consider we are adding a new property called ‘price’ to the class CAR [Or changing the property ‘model’ to ‘carModel’ ] then don’t you think we also need to update CarDAO class like changing the SQL or so on.
So What is the use of SRP here?
Great question.
First, keep in mind that this is a simplistic example in the book. It's up to the reader to expand on this a little and imagine more complex scenarios. In all of these scenarios, further imagine that you are not the only developer on the team; instead, you are working in a large team, and communication between developers often take the form of negotiating class interfaces i.e. APIs, public methods, public attributes, database schemas. In addition, you often will have to worry about rollbacks, backwards compatibility, and synchronizing releases and deploys.
Suppose, for example, that you want to swap out the database, say, from MySQL to PostgreSQL. With SRP, you will reimplement CarDAO, change whatever dialect-specific SQL was used, and leave the Car logic intact. However, you may have to make a small change, possibly in configuration, to tell Car to use the new PostgreSQL DAO. A reasonable DI framework would make this simple.
Suppose, in another example, that you want to delegate CarDAO to another developer to integrate with memcached, so that reads, while eventually consistent, are fast. Again, this developer would not need to know anything about the business logic in Car. Instead, they only need to operate behind the CRUD methods of CarDAO, and possibly declare a few more methods in the CarDAO API with different consistency guarantees.
Suppose, in yet another example, your team hires a database engineer specializing in compliance law. In preparation for the upcoming IPO, the database engineer is tasked with keeping an audit log of all changes across all tables in the company's 35 databases. With SRP, our intrepid DBA would not have to worry about any of the business logic using any of our tables; instead, their mutation tracking magic can be deftly injected into DAOs all over, using decorators or other aspect programming techniques. (This could also be done of the other side of the SQL interface, by the way.)
Alright one last one - suppose now that a systems engineer is brought onto the team, and is tasked with sharding this data across multiple regions (data centers) in AWS. This engineer could take SRP even further and add a component whose only role is to tell us, for each ID, the home region of each entity. Each time we do a cross-region read, the new component bumps a counter; each week, an automated tool migrates data frequently read across regions into a new home region to reduce latency.
Now, let's take our imagination even further, and assume that business is booming - suddenly, you are working for a Fortune 500 company with multiple departments spanning multiple countries. Business Analysts from the Finance Department want to use your table to plot quarterly growth in auto sales in their post-IPO investor reports. Instead of giving them access to Car (because the logic used for reporting might be different from the logic used to prepare data for rendering on a web UI), you could, potentially, create a read-only interface for CarDAO with a short list of carefully curated public attributes that you now have to maintain across department boundaries. God forbid you have to rename one of these attributes: be prepared for a 3-month sunset plan and many many sad dashboards and late-night escalations. (And please don't give them direct access to the actual SQL table, because the implicit assumption will be that the entire table is the public interface.) Oops, my scars may be showing.
A corollary is that, if you need to change the business logic in Car (say, add a method that computes the lower sale price of each Tesla after an embarrassing recall), you wouldn't touch the CarDAO, since if car.brand == 'Tesla; price = price * 0.6 has nothing to do with data access.
Additional Reading: CQRS
For adding new property you need to change both classes only if that property should be saved to database. If it is a property used in business logic then you do not need to change DAO. Also if you change your database from one vendor to another or from SQL to NoSQL you will have to make changes only in DAO class. And if you need to change some business logic then you need to change only Car class.
Single responsibility principle as stated by Robert C. Martin means that
A class should have only one reason to change.
Keeping this principle in mind will generally lead to smaller and highly cohesive classes, which in turn means that less people need to work on these classes simultaneously, and the code becomes more robust.
In your example, keeping data access and business logic (price calculation) logic separate means that you are less likely to break the other when making changes.
Occasionally I run into situations where all of the conditions below are true for two highly similar, but not quite identical entities or objects. This makes it difficult for me to decide how to model them, either on the database end or in terms of object modeling. I'm going to try to spell out the issue and my questions in detail, because I've found it to be a really difficult modeling problem to define. I'm trying to do both data and object modeling with these entities, so I'm going to use the terminology of both disciplines a little loosely.
1) Both entities share many identical properties, but have a few unique ones not found in the other.
2) One is not a supertype or subtype of another.
3) The overlap is not due to object inheritance.
4) The objects are used for different purposes in the same domain, but often in close proximity in any workflow. This frequently leads those with even moderate domain knowledge to confuse the entities. On the other hand, this fine separation in purposes leads to greater differences between the methods of the associated objects than their properties.
5) In some situations it may be possible to create bridge tables on the database side to express M2M relationships between the entities. Nevertheless, they have so many properties (or columns, on the database side) in common that it might make sense to store them in the same table.
Some cases in point I've run into include:
1) "Product vs. Project confusion" - especially in software marketing, where Products and Projects share many of the same properties. Normally a product will have multiple projects associated with it, but it is also unusual yet conceivable for a project to be used in multiple products.
2) The subtle differences between Features and Components in software development. A feature is developer-centric a means of supplying a benefit, from the customer's point of view, while a component is a means of implementing features on the developer's side. This is a really subtle distinction which nevertheless counts for a lot. For further discussion see Rod Maupin's post at http://www.installationdeveloper.com/347/features-and-components-101/
3) Templates vs. Types in a lot of different problem domains. For example, when identifying types of guitars through a TypeID column, the TypeTable it refers to would probably have columns corresponding to colors, string sizes, body shapes, etc. A template, on the other hand, is something you'd build a guitar from, so it would have different methods than a Type, perhaps linked to an "Apply Template" or "Make Item from Template" menu command. Nevertheless, it would have many of the same columns or properties as a Type, such as color, shape, string size etc. This distinction raises its head in thousands of different object types and templates in many problem domains, not just this narrow example. To complicate matters further, in some situations it might be helpful to associate multiple Templates with a particular Type, and/or vice-versa.
I haven't run into this problem of overlapping entities often, but when it does occur, it becomes a real bottleneck and leads to a lot of waste time refactoring the data and object models. I've read books on both topics and done a lot searches of data/object modeling webpages about the issue, but have yet to see it discussed. The only hits for "overlap" and "data model" I could find on StackOverflow were for differentiating between similar columns in one table or entity, not across tables or entities. My questions are:
1) Is there is a formal name for this issue?
2) Is there a simple shortcut or trick of the trade to identify such overlapping entities at the beginning of the modeling process, rather than much further down the line, when late recognition makes refactoring an issue?
3) How should such overlapping entities be handled? I assume that in terms of OOP, they ought to have separate objects since their methods tend to be different. Inheriting one from the other would be awkward though. A more difficult question would be whether or not it would make sense to use separate tables on the database end. Combining them might require a complex series of views plus waste storage space when the properties/columns they don't have in common are left null. Storing them in separate tables might also be wasteful though, if the common properties could be stored in single columns.
It's a tricky issue to even recognize, let alone handle. I have only a moderate amount of experience with data/object modeling, so the input of someone who really knows what they're doing would be helpful. Thanks :)
Your question concerns both database modeling aspects that object-oriented (programming) modeling aspects. Let’s start from an abstract point of view.
You say:
1) Both entities share many identical properties, but have a few unique ones not found in the other.
2) One is not a supertype or subtype of another.
and:
3) The overlap is not due to object inheritance.
But note that inheritance should not to be confused with subtyping, even if many times they are tied together! See for instance Inheritance (object-oriented programming) in Wikipedia, where this statement is supported by two citations [1,2].
In other words, even if A is not a subtype of B, and B is not a subtype of A, you can find a C from which both A and B inherits attributes.
So, you can think or not of this C as an “abstract supertype” of both A and B; but in any case it is convenient consider it as common ancestor, at least from a database point of view, so that factorize the common attributes in a “supertable”.
Then, from the object-oriented programming side, you can see A or B as subtype of C or simple as two different things, depending on the characteristics of your Object-Relational Mapping tools, from the problem at hand, etc.
Of course, this way of modelling things does not prohibit that A and B, in addition to inherit from C, have one or more relations between them, as in the example Products-Projects that you have done.
So, here is my answer to your four final questions:
1) Yes, it is called inheritance.
2) You can check if two entities have a significant number of common attributes.
3) You can model them in the database with a common table, that perhaps has some common property like integrity constraints, and with two tables that have a foreign key to it. Of course this rule is not to be applied blindly, but can have exception as all the human rules. From the programming point of view, on the other hand, you can decide to model them both with a supertype or not. This dependes on many factors, and should be decided on a case by case basis.
It is safe to say that the EAV/CR database model is bad. That said,
Question: What database model, technique, or pattern should be used to deal with "classes" of attributes describing e-commerce products which can be changed at run time?
In a good E-commerce database, you will store classes of options (like TV resolution then have a resolution for each TV, but the next product may not be a TV and not have "TV resolution"). How do you store them, search efficiently, and allow your users to setup product types with variable fields describing their products? If the search engine finds that customers typically search for TVs based on console depth, you could add console depth to your fields, then add a single depth for each tv product type at run time.
There is a nice common feature among good e-commerce apps where they show a set of products, then have "drill down" side menus where you can see "TV Resolution" as a header, and the top five most common TV Resolutions for the found set. You click one and it only shows TVs of that resolution, allowing you to further drill down by selecting other categories on the side menu. These options would be the dynamic product attributes added at run time.
Further discussion:
So long story short, are there any links out on the Internet or model descriptions that could "academically" fix the following setup? I thank Noel Kennedy for suggesting a category table, but the need may be greater than that. I describe it a different way below, trying to highlight the significance. I may need a viewpoint correction to solve the problem, or I may need to go deeper in to the EAV/CR.
Love the positive response to the EAV/CR model. My fellow developers all say what Jeffrey Kemp touched on below: "new entities must be modeled and designed by a professional" (taken out of context, read his response below). The problem is:
entities add and remove attributes weekly (search keywords dictate future attributes)
new entities arrive weekly (products are assembled from parts)
old entities go away weekly (archived, less popular, seasonal)
The customer wants to add attributes to the products for two reasons:
department / keyword search / comparison chart between like products
consumer product configuration before checkout
The attributes must have significance, not just a keyword search. If they want to compare all cakes that have a "whipped cream frosting", they can click cakes, click birthday theme, click whipped cream frosting, then check all cakes that are interesting knowing they all have whipped cream frosting. This is not specific to cakes, just an example.
There's a few general pros and cons I can think of, there are situations where one is better than the other:
Option 1, EAV Model:
Pro: less time to design and develop a simple application
Pro: new entities easy to add (might even
be added by users?)
Pro: "generic" interface components
Con: complex code required to validate simple data types
Con: much more complex SQL for simple
reports
Con: complex reports can become almost
impossible
Con: poor performance for large data sets
Option 2, Modelling each entity separately:
Con: more time required to gather
requirements and design
Con: new entities must be modelled and
designed by a professional
Con: custom interface components for each
entity
Pro: data type constraints and validation simple to implement
Pro: SQL is easy to write, easy to
understand and debug
Pro: even the most complex reports are relatively simple
Pro: best performance for large data sets
Option 3, Combination (model entities "properly", but add "extensions" for custom attributes for some/all entities)
Pro/Con: more time required to gather requirements and design than option 1 but perhaps not as much as option 2 *
Con: new entities must be modelled and designed by a professional
Pro: new attributes might be easily added later on
Con: complex code required to validate simple data types (for the custom attributes)
Con: custom interface components still required, but generic interface components may be possible for the custom attributes
Con: SQL becomes complex as soon as any custom attribute is included in a report
Con: good performance generally, unless you start need to search by or report by the custom attributes
* I'm not sure if Option 3 would necessarily save any time in the design phase.
Personally I would lean toward option 2, and avoid EAV wherever possible. However, for some scenarios the users need the flexibility that comes with EAV; but this comes with a great cost.
It is safe to say that the EAV/CR database model is bad.
No, it's not. It's just that they're an inefficient usage of relational databases. A purely key/value store works great with this model.
Now, to your real question: How to store various attributes and keep them searchable?
Just use EAV. In your case it would be a single extra table. index it on both attribute name and value, most RDBMs would use prefix-compression to on the attribute name repetitions, making it really fast and compact.
EAV/CR gets ugly when you use it to replace 'real' fields. As with every tool, overusing it is 'bad', and gives it a bad image.
// At this point, I'd like to take a moment to speak to you about the Magento/Adobe PSD format.
// Magento/PSD is not a good ecommerce platform/format. Magento/PSD is not even a bad ecommerce platform/format. Calling it such would be an
// insult to other bad ecommerce platform/formats, such as Zencart or OsCommerce. No, Magento/PSD is an abysmal ecommerce platform/format. Having
// worked on this code for several weeks now, my hate for Magento/PSD has grown to a raging fire
// that burns with the fierce passion of a million suns.
http://code.google.com/p/xee/source/browse/trunk/XeePhotoshopLoader.m?spec=svn28&r=11#107
The internal models are wacky at best, like someone put the schema into a boggle game, sealed that and put it in a paint shacker...
Real world: I'm working on a midware fulfilment app and here are one the queries to get address information.
CREATE OR REPLACE VIEW sales_flat_addresses AS
SELECT sales_order_entity.parent_id AS order_id,
sales_order_entity.entity_id,
CONCAT(CONCAT(UCASE(MID(sales_order_entity_varchar.value,1,1)),MID(sales_order_entity_varchar.value,2)), "Address") as type,
GROUP_CONCAT(
CONCAT( eav_attribute.attribute_code," ::::: ", sales_order_entity_varchar.value )
ORDER BY sales_order_entity_varchar.value DESC
SEPARATOR '!!!!!'
) as data
FROM sales_order_entity
INNER JOIN sales_order_entity_varchar ON sales_order_entity_varchar.entity_id = sales_order_entity.entity_id
INNER JOIN eav_attribute ON eav_attribute.attribute_id = sales_order_entity_varchar.attribute_id
AND sales_order_entity.entity_type_id =12
GROUP BY sales_order_entity.entity_id
ORDER BY eav_attribute.attribute_code = 'address_type'
Exacts address information for an order, lazily
--
Summary: Only use Magento if:
You are being given large sacks of money
You must
Enjoy pain
I'm surprised nobody mentioned NoSQL databases.
I've never practiced NoSQL in a production context (just tested MongoDB and was impressed) but the whole point of NoSQL is being able to save items with varying attributes in the same "document".
Where performance is not a major requirement, as in an ETL type of application, EAV has another distinct advantage: differential saves.
I've implemented a number of applications where an over-arching requirement was the ability to see the history of a domain object from its first "version" to it's current state. If that domain object has a large number of attributes, that means each change requires a new row be inserted into it's corresponding table (not an update because the history would be lost, but an insert). Let's say this domain object is a Person, and I have 500k Persons to track with an average of 100+ changes over the Persons life-cycle to various attributes. Couple that with the fact that rare is the application that has only 1 major domain object and you'll quickly surmize that the size of the database would quickly grow out of control.
An easy solution is to save only the differential changes to the major domain objects rather than repeatedly saving redundant information.
All models change over time to reflect new business needs. Period. Using EAV is but one of the tools in our box to use; but it should never be automatically classified as "bad".
I'm struggling with the same issue. It may be interesting for you to check out the following discussion on two existing ecommerce solutions: Magento (EAV) and Joomla (regular relational structure):
https://forum.virtuemart.net/index.php?topic=58686.0
It seems, that Magento's EAV performance is a real showstopper.
That's why I'm leaning towards a normalized structure. To overcome the lack of flexibility I'm thinking about adding some separate data dictionary in the future (XML or separate DB tables) that could be edited, and based on that, application code for displaying and comparing product categories with new attributes set would be generated, together with SQL scripts.
Such architecture seems to be the sweetspot in this case - flexible and performant at the same time.
The problem could be frequent use of ALTER TABLE in live environment. I'm using Postgres, so its MVCC and transactional DDL will hopefully ease the pain.
I still vote for modeling at the lowest-meaningful atomic-level for EAV. Let standards, technologies and applications that gear toward certain user community to decide content models, repetition needs of attributes, grains, etc.
If it's just about the product catalog attributes and hence validation requirements for those attributes are rather limited, the only real downside to EAV is query performance and even that is only a problem when your query deals with multiple "things" (products) with attributes, the performance for the query "give me all attributes for the product with id 234" while not optimal is still plenty fast.
One solution is to use the SQL database / EAV model only for the admin / edit side of the product catalog and have some process that denormalizes the products into something that makes it searchable. Since you already have attributes and hence it's rather likely that you want faceting, this something could be Solr or ElasticSearch. This approach avoids basically all downsides to the EAV model and the added complexity is limited to serializing a complete product to JSON on update.
EAV has many drawbacks:
Performance degradation over time
Once the amount of data in the application grows beyond a certain size, the retrieval and manipulation of that data is likely to become less and less efficient.
The SQL queries are very complex and difficult to write.
Data Integrity problems.
You can't define foreign keys for all the fields needed.
You have to define and maintain your own metadata.
I have a slightly different problem: instead of many attributes with sparse values (which is possibly a good reason to use EAV), I want to store something more like a spreadsheet. The columns in the sheet can change, but within a sheet all cells will contain data (not sparse).
I made a small set of tests to benchmark two designs: one using EAV, and the other using a Postgres ARRAY to store cell data.
EAV
Array
Both schemas have indexes on appropriate columns, and the indexes are used by the planner.
It turned out the array-based schema was an order of magnitude faster for both inserts and queries. From quick tests, it seemed that both scaled linearly. The tests aren't very thorough, though. Suggestions and forks welcome - they're under an MIT licence.
I have been programming in C# and Java for a little over a year and have a decent grasp of object oriented programming, but my new side project requires a database-driven model. I'm using C# and Linq which seems to be a very powerful tool but I'm having trouble with designing a database around my object oriented approach.
My two main question are:
How do I deal with inheritance in my database?
Let's say I'm building a staff rostering application and I have an abstract class, Event. From Event I derive abstract classes ShiftEvent and StaffEvent. I then have concrete classes Shift (derived from ShiftEvent) and StaffTimeOff (derived from StaffEvent). There are other derived classes, but for the sake of argument these are enough.
Should I have a separate table for ShiftEvents and StaffEvents? Maybe I should have separate tables for each concrete class? Both of these approaches seem like they would give me problems when interacting with the database. Another approach could be to have one Event table, and this table would have nullable columns for every type of data in any of my concrete classes. All of these approaches feel like they could impede extensibility down the road. More than likely there is a third approach that I have not considered.
My second question:
How do I deal with collections and one-to-many relationships in an object oriented way?
Let's say I have a Products class and a Categories class. Each instance of Categories would contain one or more products, but the products themselves should have no knowledge of categories. If I want to implement this in a database, then each product would need a category ID which maps to the categories table. But this introduces more coupling than I would prefer from an OO point of view. The products shouldn't even know that the categories exist, much less have a data field containing a category ID! Is there a better way?
Linq to SQL using a table per class solution:
http://blogs.microsoft.co.il/blogs/bursteg/archive/2007/10/01/linq-to-sql-inheritance.aspx
Other solutions (such as my favorite, LLBLGen) allow other models. Personally, I like the single table solution with a discriminator column, but that is probably because we often query across the inheritance hierarchy and thus see it as the normal query, whereas querying a specific type only requires a "where" change.
All said and done, I personally feel that mapping OO into tables is putting the cart before the horse. There have been continual claims that the impedance mismatch between OO and relations has been solved... and there have been plenty of OO specific databases. None of them have unseated the powerful simplicity of the relation.
Instead, I tend to design the database with the application in mind, map those tables to entities and build from there. Some find this as a loss of OO in the design process, but in my mind the data layer shouldn't be talking high enough into your application to be affecting the design of the higher order systems, just because you used a relational model for storage.
I had the opposite problem: how to get my head around OO after years of database design. Come to that, a decade earlier I had the problem of getting my head around SQL after years of "structured" flat-file programming. There are jsut enough similarities betwwen class and data entity decomposition to mislead you into thinking that they're equivalent. They aren't.
I tend to agree with the view that once you're committed to a relational database for storage then you should design a normalised model and compromise your object model where unavoidable. This is because you're more constrained by the DBMS than you are with your own code - building a compromised data model is more likley to cause you pain.
That said, in the examples given, you have choices: if ShiftEvent and StaffEvent are mostly similar in terms of attributes and are often processed together as Events, then I'd be inclined to implement a single Events table with a type column. Single-table views can be an effective way to separate out the sub-classes and on most db platforms are updatable. If the classes are more different in terms of attributes, then a table for each might be more appropriate. I don't think I like the three-table idea:"has one or none" relationships are seldom necessary in relational design. Anyway, you can always create an Event view as the union of the two tables.
As to Product and Category, if one Category can have many Products, but not vice versa, then the normal relational way to represent this is for the product to contain a category id. Yes, it's coupling, but it's only data coupling, and it's not a mortal sin. The column should probably be indexed, so that it's efficient to retrieve all products for a category. If you're really horrified by the notion then pretend it's a many-to-many relationship and use a separate ProductCategorisation table. It's not that big a deal, although it implies a potential relationship that doesn't really exist and might mislead somone coming to the app in future.
In my opinion, these paradigms (the Relational Model and OOP) apply to different domains, making it difficult (and pointless) to try to create a mapping between them.
The Relational Model is about representing facts (such as "A is a person"), i.e. intangible things that have the property of being "unique". It doesn't make sense to talk about several "instances" of the same fact - there is just the fact.
Object Oriented Programming is a programming paradigm detailing a way to construct computer programs to fulfill certain criteria (re-use, polymorphism, information hiding...). An object is typically a metaphor for some tangible thing - a car, an engine, a manager or a person etc. Tangible things are not facts - there may be two distinct objects with identical state without them being the same object (hence the difference between equals and == in Java, for example).
Spring and similar tools provide access to relational data programmatically, so that the facts can be represented by objects in the program. This does not mean that OOP and the Relational Model are the same, or should be confused with eachother. Use the Realational Model to design databases (collections of facts) and OOP to design computer programs.
TL;DR version (Object-Relational impedance mismatch distilled):
Facts = the recipe on your fridge.
Objects = the content of your fridge.
Frameworks such as
Hibernate http://www.hibernate.org/
JPA http://java.sun.com/developer/technicalArticles/J2EE/jpa/
can help you to smoothly solve this problem of inheritance. e.g. http://www.java-tips.org/java-ee-tips/enterprise-java-beans/inheritance-and-the-java-persistenc.html
I also got to understand database design, SQL, and particularly the data centered world view before tackling the object oriented approach. The object-relational-impedance-mismatch still baffles me.
The closest thing I've found to getting a handle on it is this: looking at objects not from an object oriented progamming perspective, or even from an object oriented design perspective but from an object oriented analysis perspective. The best book on OOA that I got was written in the early 90s by Peter Coad.
On the database side, the best model to compare with OOA is not the relational model of data, but the Entity-Relationship (ER) model. An ER model is not really relational, and it doesn't specify the logical design. Many relational apologists think that is ER's weakness, but it is actually its strength. ER is best used not for database design but for requirements analysis of a database, otherwise known as data analysis.
ER data analysis and OOA are surprisingly compatible with each other. ER, in turn is fairly compatible with relational data modeling and hence to SQL database design. OOA is, of course, compatible with OOD and hence to OOP.
This may seem like the long way around. But if you keep things abstract enough, you won't waste too much time on the analysis models, and you'll find it surprisingly easy to overcome the impedance mismatch.
The biggest thing to get over in terms of learning database design is this: data linkages like the foreign key to primary key linkage you objected to in your question are not horrible at all. They are the essence of tying related data together.
There is a phenomenon in pre database and pre object oriented systems called the ripple effect. The ripple effect is where a seemingly trivial change to a large system ends up causing consequent required changes all over the entire system.
OOP contains the ripple effect primarily through encapsulation and information hiding.
Relational data modeling overcomes the ripple effect primarily through physical data independence and logical data independence.
On the surface, these two seem like fundamentally contradictory modes of thinking. Eventually, you'll learn how to use both of them to good advantage.
My guess off the top of my head:
On the topic of inheritance I would suggest having 3 tables: Event, ShiftEvent and StaffEvent. Event has the common data elements kind of like how it was originally defined.
The last one can go the other way, I think. You could have a table with category ID and product ID with no other columns where for a given category ID this returns the products but the product may not need to get the category as part of how it describes itself.
The big question: how can you get your head around it? It just takes practice. You try implementing a database design, run into problems with your design, you refactor and remember for next time what worked and what didn't.
To answer your specific questions... this is a little bit of opinion thrown in, as in "how I would do it", not taking into account performance needs and such. I always start fully normalized and go from there based on real-world testing:
Table Event
EventID
Title
StartDateTime
EndDateTime
Table ShiftEvent
ShiftEventID
EventID
ShiftSpecificProperty1
...
Table Product
ProductID
Name
Table Category
CategoryID
Name
Table CategoryProduct
CategoryID
ProductID
Also reiterating what Pierre said - an ORM tool like Hibernate makes dealing with the friction between relational structures and OO structures much nicer.
There are several possibilities in order to map an inheritance tree to a relational model.
NHibernate for instance supports the 'table per class hierarchy', table per subclass and table per concrete class strategies:
http://www.hibernate.org/hib_docs/nhibernate/html/inheritance.html
For your second question:
You can create a 1:n relation in your DB, where the Products table has offcourse a foreign key to the Categories table.
However, this does not mean that your Product Class needs to have a reference to the Category instance to which it belongs to.
You can create a Category class, which contains a set or list of products, and you can create a product class, which has no notion of the Category to which it belongs.
Again, you can easy do this using (N)Hibernate;
http://www.hibernate.org/hib_docs/reference/en/html/collections.html
Sounds like you are discovering the Object-Relational Impedance Mismatch.
The products shouldn't even know that
the categories exist, much less have a
data field containing a category ID!
I disagree here, I would think that instead of supplying a category id you let your orm do it for you. Then in code you would have something like (borrowing from NHib's and Castle's ActiveRecord):
class Category
[HasMany]
IList<Product> Products {get;set;}
...
class Product
[BelongsTo]
Category ParentCategory {get;set;}
Then if you wanted to see what category the product you are in you'd just do something simple like:
Product.ParentCategory
I think you can setup the orm's differently, but either way for the inheritence question, I ask...why do you care? Either go about it with objects and forget about the database or do it a different way. Might seem silly, but unless you really really can't have a bunch of tables, or don't want a single table for some reason, why would you care about the database? For instance, I have the same setup with a few inheriting objects, and I just go about my business. I haven't looked at the actual database yet as it doesn't concern me. The underlying SQL is what is concerning me, and the correct data coming back.
If you have to care about the database then you're going to need to either modify your objects or come up with a custom way of doing things.
I guess a bit of pragmatism would be good here. Mappings between objects and tables always have a bit of strangeness here and there. Here's what I do:
I use Ibatis to talk to my database (Java to Oracle). Whenever I have an inheretance structure where I want a subclass to be stored in the database, I use a "discriminator". This is a trick where you have one table for all the Classes (Types), and have all fields which you could possibly want to store. There is one extra column in the table, containing a string which is used by Ibatis to see which type of object it needs to return.
It looks funny in the database, and sometimes can get you into trouble with relations to fields which are not in all Classes, but 80% of the time this is a good solution.
Regarding your relation between category and product, I would add a categoryId column to the product, because that would make life really easy, both SQL wise and Mapping wise. If you're really stuck on doing the "theoretically correct thing", you can consider an extra table which has only 2 colums, connecting the Categories and their products. It will work, but generally this construction is only used when you need many-to-many relations.
Try to keep it as simple as possible. Having a "academic solution" is nice, but generally means a bit of overkill and is harder to refactor because it is too abstract (like hiding the relations between Category and Product).
I hope this helps.