I am very new to Scala and trying to understand its various constructs and their use-cases, so coming down to case-classes, They are great for pattern matching etc.
So looking at MVC point of view, should our all our Models be case classes to leverage this feature ?
I looked at the play framework's sample code snippets and found an example where a model class was defined as a case class.
If you have models as in swing component models, e.g., table models, in mind, then case classes might not be the best choice. Case classes are a good choice when they are (observationally) immutable, which is usually the case if you use them to represent data retrieved from a data base. For swing models, how ever, this might not be the case, e.g., if the user is allowed to change the table data.
It's not uncommon. There are various libraries, e.g. Salat for MongoDB, that will store and retrieve case class instances from your datastore of choice.
Related
I'm aware that copying entity classes and properties into DTOs is considered anti-pattern, so by Exposed domain model pattern the same #Entity can be used as both database entity class, and DTO for service and MVC layer. (see here https://codereview.stackexchange.com/questions/93511/data-transfer-objects-vs-entities-in-java-rest-server-application)
But suppose we have microservice architecture where the same set of properties is used as entity in one project with persistence, and as DTO in another project which uses the first one as a service. What's the proposed pattern in such a situation?
Because the second project doesn't need #Entity related functionality, and if we put that class in shared library, it will be tied unnecessary to JPA specific APIs and libraries. And the alternative is to again use separate DTO classes anti-pattern.
When your requirements for a DTO model exactly match your entity model you are either in a very early stage of the project or very lucky that you just have a simple model. If your model is very simple, then DTOs won't give you many immediate benefits.
At some point, the requirements for the DTO model and the entity model will diverge though. Imagine you add some audit aspects, statistics or denormalization to your entity/persistence model. That kind of data is usually never exposed via DTOs directly, so you will need to split the models. It is also often the case that the main driver for DTOs is the fact that you don't need all the data all the time. If you display objects in e.g. a dropdown you only need a label and the object id, so why would you load the whole entity state for such a use case?
The fact that you have annotations on your DTO models shouldn't bother you that much, what is the alternative? An XML-like mapping? Manual object wiring?
If your model is used by third parties directly, you could use a subclassing i.e. keep the main model free of annotations and have annotated subclasses in your project that extend the main model.
Since implementing a DTO approach correctly, I created Blaze-Persistence Entity Views which will not only simplify the way you define DTOs, but it will also improve the performance of your queries.
If you are interested, I even have an example for an external model that uses entity view subclasses to keep the main model clean.
Thank you for the answers, but emphasize in the question is on microservice (MS) architecture and reusing defined entity POJOs from one MS in another as POJOs. From what I've read on microservices it's closely related to another question - should MSs share any common functionality and classes at all, or be completely independent? It seems there is no definite agreement on it, and also no definite answer, or widely accepted pattern, to this.
From my recent experience here is what I adopted, and it works well so far.
Have common functionality across MSs - yes, in form of a commons project added as dependency to all MSs, with its dependencies set as optional. Share entity classes (expose them in commons) - no.
The main reason is that entity classes are closely related to data store for particular MS. And as the established rule is that MSs shouldn't share data stores, then it makes sense not to share entity classes for those data stores. It helps MSs to be more independent, and freedom to manage their data store in their own way. It means some more typing to add additional DTO classes and conversion between them, but it's a trade-off worth taking to retain MS independence. Reasons Christian Beikov and Maksim Gumerov mentioned apply as well.
What we do share (put in commons) are some common functionality and helper classes (for cloud, discovery, error handling, rest and json configuration...), and pure DTOs, where T is transfer between MSs (rest entities or message payloads).
I was starting a design involving Scala case classes as models, and I was wondering about a design decision.
Let's imagine we have two models, a User model and an Order model. The Order model references the User model as a foreign key.
case class User(id: UserId, [Other fields...], password: String)
case class Order(id: OrderId, [Other fields...], userId: UserId)
Then, given this design, we would have an Orders DAO with a method findByUser method.
My question is: would it be good design to have an orders method in User that calls this DAO method, and thus making the system more OO, or is it better to keep layers isolated and not include this method?
Thanks!
If you understand you correctly, you're asking about the Active Record Pattern. As any pattern, it has its pros and cons, you can find more about it online. Here are some of them:
http://www.mehdi-khalili.com/orm-anti-patterns-part-1-active-record
https://softwareengineering.stackexchange.com/questions/70291/what-are-the-drawbacks-to-the-activerecord-pattern
Active Record Design Pattern?
In a Play2 project, I firstly used the pattern, mainly because of Ebean support. However, since I needed a bit more logic for persisting some models, extension of it turned cumbersome. In the end, I separated everything: separate service, separate model, separate DAO. That also helped me easily switch between repository layer implementations (in the end I could experiment with Spring Data JPA, Hibernate etc. more freely)
I have a bit confusion in choosing whether to use NSManagedObjet class objects directly as models or whether to create separate classess for models and create data mappers to map data from these model class to NSManagedObject class objects.
Is there any harm in using Core Data objects as models ? What are the pros and cons of both approches?
Thanks in advance,
Regards,
tek3
I read your question and I take it you are not asking whether to use NSManagedObject directly or whether to subclass NSManagedObject, but if you should have your model as separate classes which use Core Data by explicit methods written by yourself.
Core Data is designed to act as the model layer for your application. I do not see any real benefits in having your own model classes, writing an interface for them and implementing it in core data behind the scenes, unless you really need the freedom to give up core data entirely at some point.
I recommend you create your model classes as subclasses of NSManagedObject. You are free to extend those in any way you deem necessary beyond what core data provides you, but at the same time your model classes will have full benefits from the core data framework: faulting, caching, data integrity assurances, cascade deletes, etc...
If you just use NSManagedObject, you will not have the benefit of the convenient "dot.notation" when referring to attributes and relationships.
Also, you will have to retrieve all values with valueForKey and set them with setValueForKey followed by strings. This can be extremely error prone and cumbersome and the resulting code is not nearly as readably as with the dot notation.
Finally, in your object classes you can implement all kinds of additional functionality. Where would you put this code otherwise? Remember the principle of encapsulation that helps produce independent and reusable code.
I am struggling with how to understand the correct usage of models. Currently i use the inheritance of Db_Table directly and declare all the business logic there. I know it's not correct way to do this.
One solution would be to use Doctrine ORM, but this requires learning curve and all the current components what i use needs to be rewritten paginator and auth. Also Doctrine1 adds a another dozen classes which need to be loaded.
So the current cleanest implementation what i have seen is to use the Data Mapper classes between the so called model and DbTabel. I haven't yet implemented this as it seems to head writing another ORM. But example could be something this: SQL table User
create class with setters, getters, business logic here /model/User.php
data mapper /model/mapper/UserMapper.php, the funcionality is basically writing all the update, save actions in here.
the data source /model/DbTable/User.php extends the Db_Table_Abstract
Problems are with relationships between other models.
I have found it beneficial to not have my models extend Db_Table, but to use composition instead. That means my model 'has a' Db_Table rather than 'is a' Db_Table.
That way I find it much easier to reference multiple tables in the same model, which is a common requirement. This is enough for a simple project. I am currently developing a more complex application and have used the Data Mapper pattern and have found that it has simplified my code more than I would have believed.
Specifically, I have created a class which provides all access to the database and exposes methods such as getUser() etc.. That way, if the DB changes, or my client wants something daft like storing records in XML or we split the servers or something I only have to rewrite one class.
Again, my models do not extend this class, but have an instance of it assigned as a property during construction.
I would say the 'correct' way depends on the situation. Following the YAGNI and KISS principles, it is not good to over-complicate your model setup unless you really believe that it will benefit you in the long run.
What is the application you are developing? How is your current setup of extending Db_Table holding you back?
Let's say I have two tables in a database: projects and users. I create two models, that extend Zend_Db_Table_Abstract: Model_DbTable_Users and Model_DbTable_Projects.
Now, is it a good pattern to create an instance of Model_DbTable_Projects inside the Model_DbTable_Users class ? In other words: is it OK to put any logic in this model, or should I create another class, that uses Model_DbTable_Users and Model_DbTable_Projects?
I use to put all the logic in models, that extend Zend_Db_Table_Abstract, but in large projects it can make code very unclean. So, can you give me any advice on models architecture(links on articles would be great!).
I was the project lead for the Zend Framework project through version 1.0. My contributions were mainly in the Zend_Db component.
I frequently advise that people should use the Domain Model pattern and avoid the Anemic Domain Model antipattern. Remember that a Table is not a Model.
Your Model is a class (extending no base class) for code that encapsulates your business logic. The relationship between a Model and a Table isn't IS-A, it's HAS-A (or HAS-MANY). The Model treats database persistence as an implementation detail. The consumer of a Model should have no clue about your database structure (this allows you to change database structure without changing the Model's interface).
I'm basically repeating the answer I gave to Models in the Zend Framework.
Here is some more reading:
http://weierophinney.net/matthew/archives/202-Model-Infrastructure.html
http://blog.astrumfutura.com/archives/373-The-M-in-MVC-Why-Models-are-Misunderstood-and-Unappreciated.html
http://n4.nabble.com/Another-Model-Design-Thread-td670076.html