How to properly use Grails Command Objects - grails

I've been struggling for a while reading and trying to understand command objects, but I've yet to understand how to use them in my particular scenario.
Here's what I have:
class Beneficary {
String name
//more attributes
static hasMany = [dependents = Dependent]
}
class Dependent {
DegreeKinship degreeKinship //enum
//several atrributes
static belongsTo = [beneficiary: Beneficiary]
}
I've read in several articles, including SO answers, that one should be using Command Objects for this if one wishes but I'm failing to understand just how.
I've wrote this:
class DependentCommand {
List<Dependent> dependents = ListUtils.lazyList([], {new Dependent()} as Factory)
}
but I'm not sure how to use it in my Beneficiary class.
Also, I wish to have it all under a single view (beneficiary/create)
Any help would be greatly appreciated.

I don't think you should use them in the Beneficary class, use them in BeneficaryController.
Command objects give you a standardized way of encapsulating, converting and validating request parameters. As such the main use for them is in a controller, not a domain class which can already do most of a command object's functions natively.
You could rewrite your command like this if you wanted to accept a request containing parameters along the lines of dependents=1&dependents=2:
class DependentCommand {
List<Dependent> dependents
}

Related

Grails createCriteria() used inside Domain classes

Just got to know about the capability of createCriteria() method. Just wanna know that other than applying it on the Controller, is there a way to apply onto the domain classes as well? Probably on its own mapping to a property like:
static mapping = {
additionalInfo: Page.createCriteria().list()
}
Perhaps you may want to simply create a new field based the target field like the below example:
class myInfo {
String additionalInfo
String[] moreInfo // a transient field
getMoreInfo(){
def myresultmap = createCriteria.list{
// insert any other criteria shenanigans
}
return myresultmap
}
static transients = ['moreInfo']
}
In a controller return a view like normal with the Domain instance of class MyInfo
Then use in view like:
<g:each in="${domaininstancefromcontroller}">
${it.moreInfo[0]
</g:each>
see docs.
Hope this helps.
Just wanna know that other than applying it on the Controller, is
there a way to apply onto the domain classes as well?
Criteria queries are not limited to controllers, you can apply them elsewhere using the same syntax that you would in a controller. However, the particular example you show there is probably going to be problematic because you are trying to use GORM inside of the mapping block which is used to configure GORM.

Grails binding one to one associations

When you generate grails views, grails looks at your relationships and generates the right html for your form data to be automatically binded to the back end domain. For one to one associations grails creates a drop down list.
However, you might not want to present that property as a drop down list but something more custom (for example a text field with autocomplete). As soon as you do that the value that comes to the controller from that field, comes in as a String and you have to first:
Clear errors
Perform a findBy based on a given param and assign it to the property of the domain
I really want to avoid doing findBys in the controller as much as possible because it seems like I am doing logic/things that should not go there. The controller should delegate to the Service layer. It is not clear to me from the grails documentation how would I do that by using bindData which seems to work really well with String, date, Integer properties etc.. but I do not see how bindData is used for properties that are other domains.
I also really want to avoid passing the params object to the Service layer as it seems less reusable (or maybe not, correct me if I am wrong). I guess that I do not like how it looks semantically. I would prefer the first over the second:
#Transactional
class WithdrawService {
def addWithdraw(Withdraw withdraw) {
//perform business logic here
}
def createWithdraw(Map params){
//perform business logic here
}
}
Let's take the following example:
class Withdraw {
Person person
Date withdrawDate
}
and the parent lookup table
class Person {
String name
String lastName
static constraints = {
}
#Override
public String toString() {
return "$name $lastName"
}
}
In order for the bind to happen automatically without any extra work grails passes in the following request params to automatically bind the one to one:
person.id
a person map with the id.
[person.id:2, person:[id:2], withdrawDate:date.struct, withdrawDate_month:11, create:Create, withdrawDate_year:2015, withdrawDate_day:10, action:save, format:null, controller:withdraw]
What is the best way to go about this?
Pass two hidden fields that look exactly like this: person.id:2, person:[id:2] that get populated as a result of the Ajax call that populates the autocomplete?
In the controller do a Person.findBySomeKnownProperty(params.someKnownValue)
Or any other approach?

Laravel 4: how to inject another class in a eloquent model

I'm trying to use the built-in laravel's Ioc container to inject a PageManager class inside a Page model and I'm a little lost.
What I'm trying to achieve is something like that:
class Pages extends Eloquent {
public function __construct(PagesManagerInterface $manager, array $attributes = array())
{
parent::__construct($attributes);
$this->manager = new $manager;
}
public function saveToDisk()
{
$this->manager->writeToFile();
}
But I obtain this error:
ErrorException: Argument 1 passed to Pages::__construct() must be an instance of PagesManagerInterface, none given.
I tried to add this in app/start/global.php:
App::bind('Pages',function(){
return new Pages(new PagesManager);
});
But is seems ignored by the framework, and also i don't know how to insert the $attribute array into this declaration.
I'm a little lost so any help is appreciated!
It's not a good idea to overload a model's constructor because new instances can be spawned behind the scenes through various methods, like Model::find().
When that happens, the dependencies you're asking for in your custom constructor aren't being passed in because the Model class isn't aware of them. So, you get that error message.
See the find() method here: http://laravel.com/api/source-class-Illuminate.Database.Eloquent.Model.html#380-397
See this post by Jason Lewis: http://forums.laravel.io/viewtopic.php?pid=47124#p47124
I think that what you need is:
App::bind('PagesManagerInterface',function(){
return new Pages(new PagesManager);
});
This tells Laravel to inject a new Page object everytime it needs an instance of your PagesManagerInterface wich wasn't passed while creating the model.
In Laravel you can use the IoC Container:
public function saveToDisk(){
$managerObject = app()->make('path\to\class\PagesManagerInterface');
$managerObject->writeToFile();
}

Is it legal to extend an entity model with functionality in ASP.NET MVC

first of all here is my situation. I am programming an intranet application using ASP.NET MVC 3 with Entity Framework 4.1. My application has been developed using the "Unit of Work" and "Repository" design patterns.
How ever in my opinion it should go the way that my application has an unit of work that provides a central access to all the repositories which further provide access to the entities.
Lets say I have a entity called "ProductApprovalDocument" with the properties "id", "creationDate" and "approvalDecission" stored in the database. Now I want the user to be able to access a PDF file of the document thats shortly described by the entity. Because the files are stored in a central directory on a file server using the URL format "[fileServerDirectoryPath]/[ProductApprovalDocument.id].pdf", I do not want to save an extra property for that filepath on the database. What I would like to do, is give the entity an extra property called "filepath" that automatically constructs the path with the given information and returns it.
Now the Problem:
I use an interface called FileService to abstract file access from the rest of the application. Now in my case I would have to access the UnitOfWork object out of the entity model, to retrieve the current FileService implementetion and get the preconfigured filepath. I think that's the totaly wrong way because to me an entity model should only be used as a data container not more or less.
Now the Question:
How do I handle such a situation. I would not like to always set the filepath property through the controller because ist more or less static and therefore could be done somehow automatic by the model.
Edit (final solution):
Thanks to the answer of Andre Loker I gained another point of view to my problem.
What was the central target I wanted to reach?
I wanted the user to gain access to a file stored on a fileserver.
Do I have to provide every displayed entity with the total filepath?
No! Think about the principle of MVC! User actions get processed by the controller just in time. You don't have to provide information untill it really get's used.
So the solution is just to render all data as usual but instead of displaying a static html link to the files, you have to include an ActionLink to the Controller which calculates the filepath on the fly and automatically redirects the user to the file.
In the View do this:
#Html.ActionLink(Model.ID.ToString(), "ShowProductApprovalDocumentFile", "ProductApprovalDocument", new { ProductApprovalDocumentID = Model.ID }, null)
instead of this:
#Model.ID
And add an corresponding Action to the controller:
public ActionResult ShowProductApprovalDocumentFile(int ProductApprovalDocumentID )
{
return Redirect(_unitOfWork.FileService.GetFilePathForProductApprovalDocument(ProductApprovalDocumentID));
}
Thanks to the guys that took the time to give me an answer and special thanks to Andre who lead me to the satisfying answer! :)
If I understand the property correctly, there are several options:
1) Make the FilePath property use a service locator to find the FileService:
public string FilePath {
get {
FileService fileService = DependencyResolver.Current.GetService<FileService>();
return fileService.GetFilePathForDocument(this);
}
}
While I'm not a hugh fan of static service locators as they make testing more difficult, this could be a viable option. To make it more easily testable you can make the file service locator injectable:
private static readonly Func<FileService> defaultFileServiceLocator = ()=>DependencyResolver.Current.GetService<FileService>():
private Func<FileService> fileServiceLocator = defaultFileServiceLocator;
public Func<FileService> FileServiceLocator {
get { return fileServiceLocator; }
set { fileServiceLocator = value ?? defaultFileServiceLocator; }
}
And then use this in FilePath
public string FilePath {
get {
FileService fileService = fileServiceLocator();
return fileService.GetFilePathForDocument(this);
}
}
This way you can inject your own file service locator during testing.
2) Explicitly require the FileService when retrieving the file path. Instead of a FilePath property you'd have:
public string GetFilePath(FileService service){
service.GetFilePathForDocument(this);
}
The problem with this is of course that now the caller of GetFilePath needs to have a FileService. This isn't much of a problem for controllers, because if you use an IoC you can inject a FileService into the controller constructor. This approach is the cleaner one as it doesn't depend on service locators, but as you see it is slightly more inconvenient for the caller.
3) Inject the FileService into the document class itself.
Instead of using a file service locator you'd inject the file service itself when you construct your ProductApprovalDocument. With this approach you can use a simple FilePath property again. The main problem is that this often doesn't play too well with ORMs, as they often construct the objects using a default constructor and you'd have to somehow hook into the object construction process to inject the dependencies. Also, I'm not a big fan of injection services into domain objects.
4) You set the FilePath from outside the entity. As you said this should be done somewhat automatically as you don't want to do it manually every time. This would require some layer through which all entities need to pass which sets up the FilePath property.
5) Don't make FilePath a property of ProductApprovalDocument at all. This would be a reasonable choice, too. ProductApprovalDocument doesn't know anything about its FilePath, so why should it be a property? Its the FileService that calculates the value. You can still have a distinct view model version of ProductApprovalDocument which does have a FilePath property. You'd set the property when you create your view model:
var model = new ProductApprovalDocumentViewModel();
mapper.Map(realDocument, model); // map common properties with AutoMapper or so
model.FilePath = fileService.GetFilePathForDocument(realDocument);
However, if ProductApprovalDocument needs to do something with its FilePath (why would it?) this approach doesn't work anymore.
Personally I'd go with solution 5, 2 or 1 in that order of precedence, where applicable.
Whilst I would be hesitant to rely on being able to calculate the filepath and I would prefer to store it as part of the entity (in case it ever needs to change for some reason), in your situation if I was adamant I wanted to do it the way you've said, I think I would extend the FileService/ViewModel to have a Filepath property which was derived in the fashion you have stated.
e.g. if I wanted to create a download link I'd do this in the ViewModel
public string FilePath
{
get
{
return String.Format(#"thehardcodedbit{0}.pdf",ID);
}
}
EDIT: If you have an Entity generated by EF4.x then it will have been generated as a partial class so you could always extend it like this (I have done this sort of thing and it works okay):
Say the generated entity looks like this:
Namespace Da_Wolf.Model.Entities.File
{
public partial class UploadedFile
{....}
}
Then you could create a partial class like this:
Namespace Da_Wolf.Model.Entities.File
{
public partial class UploadedFile
{
public string FilePath
{
get
{
return String.Format(#"thehardcodedbit{0}.pdf",ID);
}
}
}
}
Now you have the property you desire available everywhere without adding anything to the ViewModels.

How to override the DomainClass.list() in GORM (Grails)

People, I'm facing a problem with grails GORM, my Application is totally dependent of the DomainClass.list() method, it is in all of my create/edit GSPs, but now I need a particular behavior for listing objects. Being more specific I need to filter these lists (All of them) by one attribute.
The problem is I'm hoping not to change all the appearances of these methods calling, so is there a way to customize the behavior of the default list() method ? I need it to function just the way it does, but adding an ending filter.
Maybe you can use hibernate filter plugin (see here). This will allow you to filter all finder methods (including list()) based on a property:
static hibernateFilters = {
enabledFilter(condition: 'deleted=0', default: true)
}
Have you considered using names queries? You could always do something like this:
class DomainClass {
// ... class members
static namedQueries = {
myList { params->
// put your complicated logic here
}
}
}
Then you can just replace your calls to DomainClass.list() with DomainClass.myList.list().

Resources