Good day
This question is more of a meta question than a specific problem based question.
It is always a good idea to separate any and all DB code from view related code and view files, correct?
Thus is my assumption when using JSF correct in that:
The xhtml file forms part of the view in the MVC.
The #Named backing beans also forms part of the view.
To ensure that one can relatively painlessly migrate away from JSF to another type of tech, one would ensure to not have ANY DB code inside the #Named backing beans.
All DB code should reside in a controller / service class.
The controller / service class will contain all the DB access code and business logic. This thus forms the controller of the MVC.
The #Entity classes are used to map the DB to JPA and this thus forms the model of the MVC.
Now, if my understanding of the above is correct, what would best methods and methodologies be when handling the following scenario:
I have an XHTML file displaying JSF components (Primefaces). The lists and component linked members that link the JSF components to the Java code all resides in the #Named backing bean.
Now for argument sake, let's say that the specific form is used to CRUD a supplier's information, which of the following methods is recommended as best practices (while attemping to maintain as much seperation of concern between the View and the Controller Java code) when for instance edits were made on the XHTML form:
Enforce ZERO DB code (thus never defining or using the Entity Manager) in the #Named backing bean file). The code to persist the changes after all input validation was successfully done, will reside in the Controller. To get these values to the Controller, we will have a function looking perhaps like this (basically send all the components on the xhtml form as parameters to a function in the Controller):public void supplierService (String supName, String SupAddress, String supTel....) The problem with this of course is that one may end up running into methods that takes tens of parameters. Very ugly and difficult to maintain.
Accept that separating some DB code is not possible and every #Named file must have the required JPA DB code to persist and or merge changes to the models (however if this is considered best practice, what is the use of having Controller classes?).
Create a temporary object of the same type as the model and set the attributes of this temporary object to the values obtained from the XHTML mapped components. Then only pass this temporary object to a method in the Controller. This method in the Controller will then persist and or merge the passed object's info. However I feel this may introduce unnecessary object instantiation overhead. Also I am not 100% sure what exactly happens 'behind the scenes' when I have a model named SupplierEntity.java that is mapped via JPA to a PostgreSQL DB and I call this code: SupplierEntity tempSup = new SupplierEntity(); Will JPA via Hibernate on Wildfly actually at this point create a new entity (record in the DB), and as such I cannot use this to create a temporary object to hold the values I am passing to the Controller as a temp instance of the underlying JPA entity, or will Hibernate (using JPA 2.1) ONLY create a new record when I do em.persist(mySupplier); and thus it is safe to use this method to pass objects to the controller's persisting method, instead of passing tens of parameters to the persisting method.
Something completely different than what I mentioned above is considered to be the best practice for separating the MVC components in JSF as much as possible, while still preventing having to pass 50 parameters to the Controller.
Please as said right in the start, this is a meta question regarding best practices. If Stackoverflow is not the right forum for these questions, instead of down voting this into oblivion, please let me know where I should ask instead and I will gladly delete the question from here and create it on the right forum.
Related
I have a general JSF problem, I found no nice solution for yet. See the picture for a general idea. I have a workaround solution (sorry for the typo in the image) in place that solves the problem by a listbox. However the desired solution is to display all existing versions next to each other (probably always around 1-3).
I have a view with a tree and picklist. There is a complex flow regarding the interaction between list and tree, e.g. you can only move models to subgroups, not top-level-groups and much more. I created a handler class that manages this behavior and translates it to service calls.
Now, a new requirement came up. There are several versions of this tree that should be displayed all together on one page. My gut feeling is that managing n versions in one handler is a big mess as I need to store several things in the handler already for one version.
In React, I would create a component that wraps the tree and all of the interaction. However, in JSF I'm not so sure what is the best practice here?
I would be happy about suggestions and ideas, I'm not expecting Code :)
I found a solution that fits my needs and I post it here hoping that it might help other people as well :)
So on my view I have several tree views with complex interactions. For example, if an item within the tree is moved, the operation is immediately reflected in the database. As I use JPA, I need to translate this to an entitymanager call.
The views are either displayed in a list or just one-at-a-time via a dropdown select.
Anyway, the idea is that every complex view component has its own controller with a reference to an entitymanager and a transaction, while having just one JSF handler class. If JSF would allow to create multiple handlers (like #{handler_1}, {handler_2}), the problem could be solved in a different way. But as JSF works name based and the name {#handler} always refers to the same container managed thing, this is no option.
The handler class is ViewScoped (or SessionScoped, if you prefer). For each tree component it has a ComponentController class that receives the EntityManager and the UserTransaction as well as the related data form the handler via constructor injection. This way, the handler can delegate all commands to the Controller while being DRY.
With this solution, the controller logic can be re-used regardless how many tree components exist. Each view elements binds a specific controller via handler.controllers.get(id).
All other solutions did not work for me as they are not able to perform database operations on view interactions.
To not have to keep repeating some validations, for example, who can see a button in a certain status of a document in the worlflow, I'm using session, scope, and session variables to store the user roles and application variable to store the Status related to each area.
I was evaluating whether it would be better from a performance and build point of view to implement a managed bean, to return the user roles and the possible statuses of each participating workflow area. Would it be the best structure in fact? What do you think? I do not have much experience in java. How could I construct the structure in java, several methods, one for roles and the other for set of status associated with the area that would name the related method? You could return the results of this method in arrays, or there is a better return structure.
Thanks a lot!
My best suggestion is to adopt the pageController Methodology. Then it's more like true MVC. This has been talked about on NotesIn9 screencast many times but basically you have a java object that's bound to your XPage. In effect it's a viewScoped bean that holds all your page logic. Then you can have methods like isGroupMember(), hasRole() etc and calculate that on the pageInit. There's little need to hold onto that in sessionScope in my opinion. So for example I have this in my pageController :
public boolean isGroupMember(String groupName) {
return JSFUtil.getXSPContext().getUser().getGroups().contains(groupName);
}
So that's available to each page. BUT I don't need to copy that snippet onto every page controller. In Java you can have your page controllers extend a more generic class. so I have a "base.pageController" class. All the specific page controllers extend that. So this isGroupMember() code goes into the base and then it's available to be used on every XPage. Doing it this way gives you the ability to have generic functions like this and then hold more specific function that are only for the individual page.
You can also have a hasRole() function etc...
Recommend you check out this video : http://www.notesin9.com/2016/08/25/notesin9-196-no-dependency-page-controllers/
Also for a question like this, I recommend you just use the xpages tag. Adding others like javabeans can bring people in who know nothing about XPages and XPages is unique enough of a beast that outsiders can cause some confusion on occasion.
I've written a new back-end Java component (extending UIComponentBase) as an alternative for the ExtLib Application Layout control. The control needs to show a collection of data that is looked up from another Notes application. The data is user dependant and doesn't change from page to page so, to avoid repeatedly doing a lookup to the other application, I want to store it in the session scope. (Note that because this is a layout control, there will only ever be one per page.)
I know I could use a session-scoped maanged bean (and have done in previous iterations) but the data only needs to be used in this control and shouldn't be used elsewhere on the page which it could be with a bean. So my question is, what's the best practice approach I should take here? Should I just directly store the data in the sessionMap or am I missing a trick with the component stateHolder? Or am I stuck with using a bean?
Thanks
(Edited for clarification)
It looks like you're talking about your own back-end Java components rather than Custom Controls within a single NSF.
I'm not sure at what level, when you write your own native XPages components, the properties are cached by the stateHolder when calling saveState(). I would presume no higher than View, for the reasons Frantisek says, that otherwise it would be unclear which instance to update if you had multiple on one XPage but one on another. It couldn't update both at the same time on the same page, so I would guess that each is a separate instance. As a result, the same component on multiple pages would be a separate discreet instance.
But there's nothing stopping you, in specific setters of the component, writing to sessionScope as well as the private property, and then doing the reverse on the getter. I'm not sure if you'd want to try the internal property before trying sessionScope or vice versa. It would depend how you wanted to handle the situation of the same sessionScope being updated from multiple pages (if the collection could change).
When I working with a DTO inside a ManagedBean, should I encapsulate the getters and setters?
class UserMBean {
private UserDTO user;
public String getName() {
return user.getName();
}
public void setName(String name) {
user.setName(name);
}
}
Or access directly the parametters from JSF:
<h:outputText value=#{userMBean.user.name} />
In general, people tend to use such encapsulation to hide the specific implementation of services or the structure of the model. It's an abstraction in order to make your code easier to modify without having to change the code on other places as well.
If you "flatten" your model this way, you actually just shift the knowledge of how your model looks like from the facelets to the managed bean. If you change your model now, you need to change the javacode instead of the xhtml code. One benefit in this scenario is that you now have a compile time error instead of runtime. On the other hand you kind of introduced duplicated code. Since your model is probably likely to change less often compared to service implementations, I usually "deeply" access my models in the xhtml code, because it saves time and is just more comfortable and keeps the managed beans more clean. In your simple case I would avoid the additional getter.
It is, as so often, a matter of taste and usecase specific. If you "deeply" access a property of your model through five (model) layers, and you do this in ten different places and you know that your model structure might change soon, it starts to make more sense to have a utility method to directly access this property with an additional getter in one place in a managed bean.
However there is one more case where it is necessary to offer direct access to properties. If you want to avoid someone accessing other properties. That means, you want to keep certain attributes non-accessible. Let's say your UserDTO also a property password and you don't want anyone to be able access that in your xhtml to display or change it, then roll with your second option and don't have a getUser(), but offer direct access to the username via getUserName().
I'm currently building my first JSF 2 application and I have some questions regarding flow of data between views.
Background
The application keeps track of competences of IT-consultants. To illustrate my questions, I will just describe a small part of it. A Competence entity has a many-to-one relation to a Area entity, putting the different competences in groups. I have views to handle these: Competence.xhtml, EditCompetence.xhtml and EditCompetenceArea.xhtml. These are backed up by a ManagedBean, AdminCompetenceController.
Competence.xhtml lists all Areas in a table, and lists all Competences in an Area if the user clicks on its row. It also has a new-button for Competences and Areas, and an Edit-button for each Competence and Area. these leads to EditCompetence.xhtml and EditCompetenceArea.xhtml respectivly, where the user can fill in information about a Competence or and Area.
Questions
What is the recommended way to handle flow of information between these pages? For example: EditCompetenceArea.xhtml needs to know which Area it should edit (or if it should make a new one). Should I have one backing bean per view or one for all of them? Should it be Session scoped and keep track of the selected Area/Competence with variables? Or should it be View scoped and send the id of the selected object with viewParams? Is it even possible to use Request scope? I've tried several of these and have run into practical problems with each method.
Is it possible to use viewParam to transfer information between views using different backing beans?
Is it possible to use viewParam with a Request scoped backing bean? The data seems to disappear from the bean before I can use it in the postback.
Also, if anyone can recommend any reading material (preferably free, online) regarding more general design patterns rather than specific smaller problems in JSF 2, I would really appreciate it.
To retain data in conversations with the same view, put the bean in view scope. The data will get lost after you navigate to a different view.
To retain data in conversations with different views, either put the bean in session scope to keep it alive among all requests/views, or use <h:inputHidden>, <f:setPropertyActionListener>, etc to retain request scoped data in subsequent request.
In your particular case, I'd go with single main #ViewScoped bean and a single CRUD view wherein the display table and the edit form is included/rendered conditionally.