First off, I've done a lot of research on the web (including this site) and have found lots of conflicting information on how the model and controller communicate in an MVC pattern. Here is my specific question (I'm using AS3), but it's a general MVC question...
I have two main components... a list of recipes and a form that displays a selected recipe. The form has an edit state that allows you to edit the recipe and then save or cancel the changes. What is the best way (using MVC principles) to handle changes made to a recipe? So far, I have the save button trigger an event which is captured by the controller.
Should I have the save button (the view) pass an object with the current state of the fields along with the event (some logic in view)? Should I allow the controller to hold access to the view and have the controller figure out what's in the fields on its own (added coupling)? Should events be made every time a field in the form is changed and the controller keeps track of the state of each field (lots of events)? Or is their another way? Note: I don't want to bind the fields to the model because I only want the data to save if the save button is clicked.
Any help would be greatly appreciated. Thanks!
Should I have the save button (the view) pass an object with the
current state of the fields along with the event (some logic in view)?
yes, this is not 'logic in the view', it does not decide anything, simply reporting an action and its current state
Should I allow the controller to hold access to the view and have the
controller figure out what's in the fields on its own (added
coupling)?
no, this would become very messy, pass a VO with the event
Should events be made every time a field in the form is changed and
the controller keeps track of the state of each field (lots of
events)?
is an option, but this basically is the same as hitting the save button, the trigger is different (TextField.onChange), but you can dispatch the same event (setup is form = view, dispatches one general event with a VO, not an event for each field)
Or is their another way?
MVC flow w/ events:
onClick save btn: a RecipeEvent.SAVE is dispatched (from the view), with a VO (value object) containing the Recipe data (e.g. RecipeVO)
the controller catches this, and as the controller is where the logic resides, it decides what to do with it: update the RecipesModel (either directly by calling a method on the model, or by a custom event e.g. RecipeModelEvent.SAVE)
the model stores the data, and dispatches the RecipeEvent.UPDATE event (with the RecipeVO)
the views updated itself accordingly (check if the RecipeVO.ID is same, update data representation e.g. title…)
optionally, a controller could save the data to a back-end/remote database
As for the event listeners: views listen to the model, controller listen to the views.
About de-coupling:
use interfaces (by using an IModel you can swap out the model easily by another implementation, as you register the event listeners against the interface instead of the actual implementation)
Obviously, all this results in a lot of registering/removing event listeners, and keeping reference to the models/views/controllers to be able to register to the appropriate instance. An alternative is to use a framework as RobotLegs, as it makes a central event-bus available + easy/automatic cleanup of event listeners in the mediator class of a view.
I think this is really a database question. what I would do is create a stored procedure that first checks to see if the recipe already exists. If yes then update it. If not then add a new recipe. (you'll have to bind your entity to a stored procedure. other MVC frameworks can do this. I don't know about actionscript)
if that's not an option then I guess you would have to cache the original form in a helper class in the controller and then compare it to what the user is trying to save. And have the controller decide whether to update the recipe.
I think it's much cleaner to use the first way, but I've never used actionscript so...
Colin Moock's lecture on MVC in ActionScript is quite old, but still one of the best explications: http://www.moock.org/lectures/mvc/
Your model should populate your view, and your view should send input events to the controller, which should decide what to do with the input. As g10 says, wait until the save button is clicked and then pass an object with the modified fields up to the controller for processing. The controller can then decide whether or not to accept it, and whether to update an existing model object or create a new one.
Related
I'm using redux and es6. I want to dispatch an action before components will mount. The problem is that with redux newest syntaxis componentWillMount doesn't exists any more. So... where should i dispatch this action.
My case: I have a component that needs user's info (such as name, for example). I need to get the user's name before that component mounts.
thanks,
Connect your component to a field in your state, let's call it "name". When the component renders, make sure it checks if name is not empty; if it is then it renders it, if not then it renders nothing.
In the componentDidMount, fire your dispatch as normal - this in turn will reduce and end up changing the value of "name" in your state. This will cause a re-render and will then show properly.
componentWillMount only exists server side, so it is usually not the best idea to dispatch actions from it, as if you just had your client side code, they wouldn't work.
I have an Android app which uses a SupportActionBar that contains a bunch of tabs. These tabs each have a fragment that in turn are connected to a ViewModel in my core project. This works great and when i start the app they are all initialized right away and setup correctly.
What i would like to do is to call on a method on one of these ViewModels from my main activity that contains all the tabs and fragments.
I read in another post that in WP you could cast the DataContext to the ViewModel but that might not work in Android. I haven't been able to do this, maybe because my DataContext is not the currently displayed ViewModel but the MainViewModel connected to my main activity. Or maybe it's not supposed to be done that way, i'm not sure.
I'm trying to do this:
var test = (MessagesViewModel)this.DataContext;
test.GetViewDataFromApi();
To update the data in the view when i press the tab. I can't use the Init function for this for example since the ViewModel isn't recreated everytime i show the view.
Are you trying to update some data in the tab's fragment when tab is selected?
If that's the case, one way to do it is to
1) handle the tab selection event to get the current tab(maybe using TabListener),
2) get the fragment (MvxFagment) in the selected tab
3) get the (IMvxViewModel) view-model from the fragment
4) call the method you need to update data on the view-model
I assume you are using a MvxFragment (https://github.com/MvvmCross/MvvmCross/blob/v3.1/Cirrious/Cirrious.MvvmCross.Droid.Fragging/Fragments/MvxFragment.cs?source=cc) so you can access the view-model from the MvxFragment's ViewModel property.
What exactly will this addActionListener Do.....we we call button.addActionListener(this) what will happen
It basically adds this (the current object) to a list of objects that will be notified when the component has an action performed on it, such as a button being pressed.
It's a way of registering your interest in what is happening to the component and is useful in that you don't have to keep polling a component to check its status.
Your object (or class, really) simply implements the interface methods for listening (such as actionPerformed) and that method will be called for each event that happens.
The Java tutorials have a large variety of different articles on the various listeners that you're likely to use.
Is it possible to raise built-in MS Access form events programmatically? I have a
feeling it isn't but thought I would check. (I am using Access 2003).
For instance, I want to do something like this within a private sub on the
form:
RaiseEvent Delete(Cancel)
and have it trigger the Access.Form delete event -- i.e. without actually
deleting a bound record.
Note my delete event is not handled by the form itself but by an external
class, so I can't simply call Form_Delete(Cancel).
I can understand your confusion -- I didn't explain any of the bigger context. Sorry.
Basically, the situation is I have an 'index' i.e. 'continuous-forms' form which is bound to a read-only query. The query has to be read only because it involves an outer join. But, I want to be able to delete underlying records from this form.
So my first thought was to do the deletion outside the form recordset, eg. using a delete query. And I was hoping to hook the standard Delete/BeforeDelConfirm/AfterDelConfirm events around this manual deletion routine by raising these events myself. But alas, this is not possible.
If the form itself handled these events, I could simply call the handlers (Form_Delete, etc), but my project has custom classes that handle form delete and update events (validation, confirmation, logging, etc.) for all the forms. (#Smandoli, it's not well-documented, I just discovered it a few months ago and use it extensively now -- maybe you know about it already -- you can set up external classes to handle your form events. See for example here)
Long story short, I found a workaround I'm satisfied with. It involves making the 'index' form a subform of another form that is bound to a recordset that can be deleted from. So the deletion can be done in the outer form using the standard Access form events, based on the selection in the inner form.
#Knox, I disagree in principle that being able to raise 'built-in' events yourself is difficult to document and maintain. Plenty of other frameworks depend on it. In practice, I agree with you, since we all have to work within the limitations of our tools and 'best practices' that evolve around those limitations. The blessing and curse of Access is its tight binding between recordsets and forms...
In common, there is simply no need to fire standard form events on your own, normally this shows a wrong understanding of form events in general (if not even events in general).
The form events exists to react on user interaction with the form or to notify the code behind of something that generally happens (like the Form_Load event). The event subs are there to react on these event - nothing more.
It is an often seen thing that people wants to execute event subs directly, but that's also a wrong way. There is a reason why event subs are in general declared as "Private" and not "Public", it should prevent calling them directly from outside the code module, but in fact you should also not execute any of them inside the same code module. Event subs always has to be called exclusively by their events, although it's possible to call them directly.
If an event sub has any code which should be executed also elsewhere then create a private or public sub inside the same module (depending on if you want to execute them from outside or not) and then call this sub from the event sub. If you think you must execute the same sub from elsewhere you can now also call the same sub. This is not a question of "it is possible to call the event sub directly", it is mainly a design question. You should always be sure that an event sub was called only by the event itself and never by any code. The problem when calling an event sub by code is that you can get in trouble very fast if you execute a code and also a real event executes it. In the end you get a big chaos of code which is very hard to debug.
It is, by the way, of course possible to call the event subs from a class module which has a reference to the form (which is needed if you use the class module to handle general events). You only would need to declare the event subs as Public and then you can call them with the form reference, but as stated above: Don't do that.
If a class module is used to handle the events then you can do anything here, you don't need the form code.
If a query is read only and you want to delete a record of a base table no event sub could help you. They are fired when the user wants to delete something which he can't do because it's read only so DoCmd does also not help you.
Like David said in the comment above, simply create a Delete button anywhere you want which can then read out the ID of the current row in the continous form and start a "DELETE" SQL command, then simply requery the continous form and you're done. You can also handle this in your standard class module because you can not only forward form events you can also forward control events on the same way. Create an Init procedure in your class module which takes all the controls from your form you want to handle with it any maybe additional the name of the base table in each continous form, then the class module can assign it to a standard "WithEvents" defined control variable of for example type CommandButton and save the base table name to a string variable. (Don't forget to set the OnClick event to "[Event Procedure]" in the Init procedure.)
In the Load event of your continous form where you probably initialize your class module you can then forward the base table name and the delete button control to the Init procedure of the class module which then can handle the deletion on a very generic way by starting a DELETE query on the base table and requery the form because it already has the form reference also. No need to call any event procedure.
Last but not least: Maybe there are frameworks which allow you to raise events directly but in common I would say that the creators of such frameworks also didn't understand the purpose of event procedures. If you have ever created an own event in a class module of your own you will see that they also cannot be raised outside the class module. Of course, you CAN create a "RaiseEvent" sub on your own to call them externally - and in fact, in case of own events it can make sense in some scenarios. In case of form (control...) events they should inform the code about something happened and there should be a reaction now. If you use events in own class modules you would normally also create a "WithEvents" variable in the outside module to get informed when an event happened in the other class module. An event should make it possible to make the module objects independent of each other. The module with the event will only raise the event and it doesn't know if anyone is listening to it or react on this event. It informs "the world" that there was something which happened in the class module, nothing else. Like a radio station which sends the daily news "to the world" but it doesn't know about if anyone listens to it. Normally, no listener of the radio station would go to the radio station and reads his own news for other listeners. Only the people at the radio station decides what to send and when. Same story.
I have a tree control in my GUI (with naturally lots of GUI/platform specific functions to handle the nodes).
I have a data model with its own complex set of nodes, children, properties etc..
I want the tree to display a representation of the model, be able to send messages to the nodes inside the model and be told to redraw itself when the model changes.
But I don't want the GUI code to need to know the details of the model's data types and I don't want to pollute the model by linking it to the GUI classes.
I can't get my head around how the controller is supposed to do this and what functions it should provide?
(this is in C++ but that shouldn't matter)
GUI "controls" don't quite fit neatly into a model-view-controller pattern because they typically have their own internal model rather than accepting a reference to one. If the control is structured that way, you'll need an adapter class that "data-binds" the control's internal model to the underlying data model.
This can accomplish something similar to what model-view-controller would, except that the adapter class plays the role both of a view hookup component (updating the GUI from the data model) and a controller (interpreting GUI events into model actions).
Qt provides a group of classes for model-view programming. You can hook a tree view to a filesystem model, for example, and neither directly know anything about each other (except a pointer to the model in the view).
Your requirements are:
Tree displays representation of model
Nodes in tree can send messages to nodes inside model
Tree redraws itself based on model changes
I don't know exactly what kind of data you're working with here, but a hierarchical model is a fairly simple thing. I'll take it as a given you know how to iterate hierarchical data and populate a tree view.
Your controller should have member function(s) for sending messages to the model. The parameters should be a model element and the message you want to send. This way, the UI is completely unaware of how the message gets to the element, but can get the messages through.
The last requirement is more tricky, and depends on a few things (e.g., the lifetime of the controller, application architecture, etc.) I'm going to assume the controller lives as long as the tree view does. If that's the case, then your controller should provide a way to set a callback on model changes. Then, when the controller changes the model, it can callback to the UI without being aware of the UI.
i think your troubles start with an unfortunate choice of words. the 'control' thingies doesn't have anything to do with the 'controller' in MVC. that's why some GUI libraries use other names (widgets is a common one).
your 'tree control' is the view, not the controller. it should be tied to the GUI, both for display, and to get GUI events and turn them into 'tree events'.
the controller gets those 'tree events' and does the necessary modifications to the model. that's where you couple the 'action' with the 'response'.
First Solution: You can implement a "Subject observer" pattern between model and view, with model as the subject and view as the observer. Whenever there is a change in the state of model, it can fire a event to all the observers those are registered, they can update themselves.
Second Solution: Introduce a controller, which registers with model as observer. Upon receiving a event for update from Model, it can update the view. Even you can decouple view from controller using one more subject observer pattern between controller and view
Third Solution: Use MVP pattern. Model view Presenter. This pattern used whenever there is no much computation in controller i.e., job of the controller is just to update its corresponding view. Here controller becomes Presenter.
You need a controller that sits outside the display widget but has the state of the tree (in MFc there are CTreeView/CTreeCtrl classes - there is a similiar separation in Qt) the tree controller handles all the data storage and calls redraws on the tree widget.
Changes in the tree widget get sent to the tree controller - so this controller needs to know about the gui functions.
The model will need set/get functions for all the relevant parameters for the nodes. But these can return simple types so aren't dependent on the gui.
Updating the view form the model requires sending a message, if you don't want the model to know about your gui messaging the best you can do is register a callback function (a void pointer to a function) from the tree controller - and call this to do an update.
This update function can then query the model for the changes.