I would like to bind my EMF model to a TreeViewer. My Model looks like this:
Facility : Name
- Part : Name
- SubPart : SubPartName
- SubSubPart : SubSubPartName
- SubSubPart : SubSubPartName
- SubSubPart : SubSubPartName
- SubPart : SubPartName
- Part : Another Name
- SubPart : SubPartName
I followed the Tutorial from Tom Schindl but I still do not understand how to implement the TreeFactoryImpl. So far it looks like this:
#Override
public IObservable createObservable(Object target) {
if (target instanceof IObservableList) {
return (IObservable) target;
} else if (target instanceof Facility) {
IEMFListProperty multi = EMFProperties.multiList(ModelPackage.Literals.FACILITY__NAME);
return multi.observe(target);
} else if (target instanceof Part) {
IEMFListProperty multi = EMFProperties.multiList(ModelPackage.Literals.PART__NAME);
return multi.observe(target);
}
return null;
}
The problem is that only the facility is displayed. Which is now wonder since the TreeStructureAdvisor only seems to receive strings.
What I would like o accomplish is:
1.) Update the TreeViewrt if new nodes are added
2.) Update the name label if the name is changed
Taken from here: some googlecode this method looks different to what you are using.
private static class TreeFactoryImpl implements IObservableFactory {
private IEMFListProperty multi = EMFProperties.multiList(
WorkspacePackage.Literals.PROJECT_SPACE__PROJECT,
MetamodelPackage.Literals.PROJECT__MODEL_ELEMENTS);
public IObservable createObservable(final Object target) {
if (target instanceof IObservableList) {
return (IObservable) target;
} else if (target instanceof ProjectSpace) {
return multi.observe(target);
} else if (target instanceof Project) {
return multi.observe(target);
}
return null;
}
}
From what I have seen there most example calls to the multilist are made with two lists. Is there a reason you are using that?
The following works:
#Override
public IObservable createObservable(Object target) {
if (target instanceof IObservableList) {
return (IObservable) target;
} else if (target instanceof Facility) {
IEMFListProperty list = EMFProperties.list(ModelPackage.Literals.FACILITY__SUBPARTS);
return list.observe(target);
} else if (target instanceof SUBPART) {
IEMFListProperty list = EMFProperties.list(ModelPackage.Literals.SUBPART__SUBPARTS);
return list.observe(target);
} else {
return null;
}
}
The key here is to listen on the collection. This solves requirement 1 (Update the TreeView if new nodes are added)
After that I added a TreeLabelProvider as described in the blog Post from Tom Schindl.
This solves requirement 2 (Update the name label if the name is changed).
Related
I wrote an Eclipse Plugin that basically allow a programmer to select a Java source from the Project Explorer and by selecting the corresponding DropDown menu option it will creates an interface .java file based on the one selected.
Everything works fine, but now I need to program the update part of the job.
The update requierement is simple, I need to listen for changes and identify that the sources that have the interface generated have been modified and recreate the interface file.
To do this I wrote a class that implements IResourceChangeListener interface.
That class looks like:
public class DTOChangeListener implements IResourceChangeListener {
private List<UpdatedUnit> updatedUnits;
public DTOChangeListener() {
super();
this.updatedUnits=new ArrayList<UpdatedUnit>();
}
#Override
public void resourceChanged(IResourceChangeEvent event) {
try{
if(event.getType() == IResourceChangeEvent.POST_CHANGE){
event.getDelta().accept(this.buildVisitor());
}
}catch(CoreException ex){
ex.printStackTrace();
}
}
protected IResourceDeltaVisitor buildVisitor(){
IResourceDeltaVisitor result=new IResourceDeltaVisitor() {
#Override
public boolean visit(IResourceDelta resDelta) throws CoreException {
String resName=resDelta.getResource().getName();
if(resName==null || resName.equals("")){
return true;
}
String[] splits=resName.split("\\.");
String name = splits[0];
if(name.contains("PropertyAccess")){
return false;
}
String interfaceName=name + "PropertyAccess";
String interfaceFile=interfaceName + ".java";
IResource res=resDelta.getResource();
if((res instanceof IFolder) || (res instanceof IProject)){
// Avoid Folder & Project Nodes
return true;
}
IProject project=res.getProject();
if(project!=null){
if(project.isNatureEnabled("org.eclipse.jdt.core.javanature")){
IJavaElement element=JavaCore.create(res);
if(element instanceof ICompilationUnit){
ICompilationUnit unit=(ICompilationUnit)element;
IPath path=res.getProjectRelativePath().removeLastSegments(1);
IResource propertyAccess=project.findMember(path.append(interfaceFile));
if(propertyAccess!=null){
UpdatedUnit updatedUnit=new UpdatedUnit(project, path, unit);
updatedUnits.add(updatedUnit);
return false;
}
}
}
}
return true;
}
};
return result;
}
public List<UpdatedUnit> getUpdatedUnits() {
return updatedUnits;
}
}
I add the Listener to the Workspace, now the question I have is:
How can I know when the updatedUnits List is completed in order to proccess the list with my own code?
One posible answer to this question would be, don't worry, the:
event.getData().accept(this.buildVisitor());
will block until proccessing of the visitor finish.
but at least is not documented like it would.
Any ideas would be appreciated.
Thanks in Advance.
Daniel
Unless it's documented to not block, it blocks.
I can't understand that part, neither trying the showcase examples.
I'm using an extension of AsyncDataProvider to bind my tree to RPC service. Here's my method:
public <T> NodeInfo<?> getNodeInfo(T value) {
/*
if (value instanceof Categoria) {
dataProvider.setCurrentParent((Categoria)value);
}
*/
return new DefaultNodeInfo<Categoria>(dataProvider, new CategoriaCell());
}
"currentParent" is my stuff: except for (null => root) values, I set the parent to pass via RPC to my service. Actually, in my widget code:
dataProvider = new CategorieTreeDataProvider() {
#Override
protected void onRangeChanged(HasData<Categoria> display) {
updateTree(getCurrentParent());
}
};
private void updateTree(Categoria categoria) {
rpcService.getCategorie(categoria, new AsyncCallback<Categoria[]>() {
#Override
public void onSuccess(Categoria[] result) {
dataProvider.updateRowCount(result.length, true);
dataProvider.updateRowData(0, Arrays.asList(result));
}
#Override
public void onFailure(Throwable caught) {
Window.alert(caught.toString());
}
});
}
My rpc-server code, however, is working as expected:
#Override
public Categoria[] getCategorie(Categoria parent) {
List<Categoria> categoryList = categorieDao.listByProperty("parent", parent);
for (Categoria c : categoryList) {
if (categorieDao.listByProperty("parent", c).size() == 0) {
c.setLeaf(true);
}
}
return categoryList.toArray(new Categoria[0]);
}
**Then I add some data to my Categories: 'GrandFather', 'Father' and 'Son'.
Unfortunately, after loading my widget, I see:
The grandfather correctly, with his "+" how expected;
Then I click it and...
The grandfather disappear and I see 'Father' with his '+'
same for father -> son
I suspect the bug is in updateRowCount / updateRowData usage.**
Any ideas?
The getNodeInfo is called whenever you open a node so you have to create distinct DataProvider for each of the nodes's childs.
public <T> NodeInfo<?> getNodeInfo(T value) {
if (value == null) {
return new DefaultNodeInfo<Category>(dataProvider, new CategoriaCell());
}
else if (value instanceof Categoria) {
Category category = (Category)value;
return new DefaultNodeInfo<Grandfather>(new ListDataProvider<Grandfather>(category.getGrandFathers()),new GrandFatherCell());
}
else if (value instanceof Grandfather) {
Grandfather grandfather = (Grandfather)value;
return new DefaultNodeInfo<Father>(new ListDataProvider<Father>(granfather.getFathers()),new FatherCell());
}
else if (value instanceof Father) {
//same as above but with fathers.
}
}
The category.getGrandFathers() function can for example do a RPC request to the server or just return the list if you retrieve everything in one RPC request.
UPDATE based on comment:
So in case you have only one class and want to achieve a dynamic CellTree (number of levels are not pre-determined) you could take following approach.
public <T> NodeInfo<?> getNodeInfo(T value) {
if (value == null) {
return new DefaultNodeInfo<Category>(dataProvider, new CategoriaCell());
}
else {
Category category = (Category)value;
return new DefaultNodeInfo<Category>(new ListDataProvider<Category>(category.getSubCategories()),new CategoryCell());
}
}
category.getSubCategories() is either an RPC call which retrieves the subcategories for the current category or if the Category class is a linked list type datastructure it could just return the list of subcategories.
Each data provider updates a given "list" (child nodes of a given parent node), so you have to use a distinct data provider instance for each parent node, or your calls will update some random list.
I'm just getting to grips with custom validation attributes, and I'm trying to write a custom validation attirbute which will be placed at class level to validate against multiple properties of my model.
I can access all properties on my model, and I want to be able to check for multiple conditions in my IsValid overload, and report on them, having different error messages as follows (simplistic example).
public override bool IsValid(object value)
{
var model = (MyObject) value;
//if this value is set, I don't want to do anything other checks
if (model.Prop3)
{
return true;
}
if (model.Prop1 == "blah" && model.Prop2 == 1)
{
ErrorMessage = "you can't enter blah if prop 2 equals 1";
return false;
}
if(model.Prop1 == "blah blah" && model.Prop2 == 2)
{
ErrorMessage = "you can't enter blah blah if prop 2 equals 2";
return false;
}
return true;
}
But when I do this I get an exception on the first time ErrorMessage is referenced "Cannot set property more than once.
Now I could split up my custom attribute into multiple custom attributes, but hoped there would be a way to do it in one, otherwise, I'll be repeating my "catch all" in each
//if this value is set, I don't want to do anything other checks
if (model.Prop3)
{
return true;
}
I've had a search already, but couldn't find anything, so apologies if I am missing anything obvious.
thanks in advance!
In MVC4 you can override IsValid to return different messages as the ValidationResult
public class StrongPasswordAttribute : ValidationAttribute
{
protected override ValidationResult IsValid(object value, ValidationContext context)
{
if (value == null)
return new ValidationResult("Password is required");
var val = value.ToString();
if (!Regex.Match(val, #"^(?=.*[a-z]).{0,}$").Success)
{
return new ValidationResult("Password must contain at least one lower case letter");
}
if (!Regex.Match(val, #"^(?=.*[A-Z]).{0,}$").Success)
{
return new ValidationResult("Password must contain at least one UPPER case letter");
}
if (!Regex.Match(val, #"^(?=.*\d).{0,}$").Success)
{
return new ValidationResult("Password must contain at least one number");
}
return ValidationResult.Success;
}
}
Interesting question! I can think of two work-arounds to this. So not proper solutions based on what you want but they might help to re-use your code. Cant you create a CustomAttribute abstract class called MyCustomAttribute (or something) that overrides IsValid in the following way:
public override bool IsValid(object value)
{
var model = (MyObject) value;
//if this value is set, I don't want to do anything other checks
if (model.Prop3)
{
return true;
}
CustomValidate(model);
}
CustomValidate(MyObject model) is your abstract method then, you can write multiple custom attribute classes that extend MyCustomAttribute and purely need to implement the validation logic for A particular scenario.
So you can have two classes:
public class BlahCustomAttribute : MyCustomAttribute
{
public override Boolean CustomValidate(MyObject obj)
{
if (model.Prop1 == "blah" && model.Prop2 == 1)
{
ErrorMessage = "you can't enter blah if prop 2 equals 1";
return false;
}
}
}
public class BlahBlahCustomAttribute : MyCustomAttribute
{
public override Boolean CustomValidate(MyObject obj)
{
if (model.Prop1 == "blah" && model.Prop2 == 1)
{
ErrorMessage = "you can't enter blah blah if prop 2 equals 1";
return false;
}
}
}
Hope this helps - not exactly what you wanted but will do the job and its clean as well.
The other solution is to comma-separate the error messages in the ErrorMessage property and handle it in the front-end (but I would go with the first approach).
Does the ASP.NET MVC 2 Default View Model Binding support binding a multi-value cookie to a custom object? Before I write a custom Value Provider, I would like to be sure that the functionality didn't already exist.
So given an action like:
public ActionResult SomeAction(CustomObject foo)
where CustomObject is something like:
public class CustomObject
{
public string Name { get; set; }
public int Rank { get; set; }
}
and a cookie that is part of the request like:
foo=Name=John&Rank=10
Could I get the Default View Model Binding to map the cookie to the parameter with some clever tweaks to the naming of the cookie or cookie values like posting "foo.Name=John" and "foo.Rank=10" would do?
Well, there's one way to do it would be to implement IModelBinder
public class CustomObjectModelBinder : IModelBinder {
public object BindModel(ControllerContext controllerContext, ModelBindingContext bindingContext) {
HttpCookie c = controllerContext.HttpContext.Request.Cookies["foo"]
CustomObject value = new CustomObject() {
foo.Name = c.Values["Name"],
foo.Rank = c.Values["Rank"]
}
return CustomObject
}
}
Then just add this to your Application_Start()
ModelBinders.Binders.Add(typeof(CustomObject), new CustomObjectModelBinder());
you can add the cookie object to any action as far as i know and it will attempt to bind it for you
In the end I created something to do this. Based on the work posted by Mehdi Golchin, I created a value provider that allows this kind of binding to happen.
For those interrested, the following are the custom changes I made to Mehdi's work linked above. See the link for full details on implementation. This doesn't support binding to nested objects (e.g., Foo.Cell.X) because I didn't need that level of complexity, but it would be possible to implement with a bit of recursion.
protected virtual bool ContainsPrefix(string prefix)
{
try
{
var parts = prefix.Split(new char[] { '.' }, 2, StringSplitOptions.RemoveEmptyEntries);
switch (parts.Length)
{
case 0:
return false;
case 1:
return this._context.HttpContext.Request.Cookies.AllKeys.Contains(parts[0]);
default:
var cookie = this._context.HttpContext.Request.Cookies[parts[0]];
if (cookie == null) { return false; }
return cookie.Values.AllKeys.Contains(parts[1]);
}
}
catch (Exception ex)
{
ExceptionPolicy.HandleException(ex, "Controller Policy");
return false;
}
}
protected virtual ValueProviderResult GetValue(string key)
{
try
{
var parts = key.Split(new char[] { '.' }, 2, StringSplitOptions.RemoveEmptyEntries);
if (parts.Length < 2) { return null; }
var cookie = this._context.HttpContext.Request.Cookies[parts[0]];
if (cookie == null) { return null; }
var value = cookie.Values[parts[1]];
if (value == null) { return null; }
return new ValueProviderResult(value, value, CultureInfo.CurrentCulture);
}
catch (Exception ex)
{
ExceptionPolicy.HandleException(ex, "Controller Policy");
return null;
}
}
Is there a proper way to access other form fields from inside a validator?
Is there another solution than:
context.getViewRoot().findComponent("formid:exampleField:example")?
f.e I want to validate a city field inside a custom validator and checking if country is US.
Re-read your question and I am going to interpret it as this:
"You would like to write a custom validator that checks that if a city field exists, the country field is equal to 'US'"
So, I would look at going about this in the following fashion:
First create a validator interface:
#Documented
#ValidatorClass(value=CountryEqualsUSValidator.class)
#Target(ElementType.TYPE)
#Retention(RetentionPolicy.RUNTIME)
public #interface CountryEqualsUS {
String message() default "The country should be US for the city provided";
}
Then create a validator class:
public class CountryEqualsUSValidator implements Validator<CountryEqualsUS> {
public void initialize(CountryEqualsUS arg0) {
}
public boolean isValid(Object value) {
if(value != null && value instanceof YourBeanClass) {
YourBeanClass yourBeanClass = (YourBeanClass) value;
if(/*some test logic here*/) {
return true;
else {
return false;
}
}
return false;
}
}
Then on the class that you want to validate:
#CountryEqualsUS
public class YourBeanClass {
...
}
Then, finally, on your controller/action class, when the form is submitted, the city is a value for which you want to check the country, add this method and call it:
public boolean doValidation(YourBeanClass yourBeanClass) {
ClassValidator requestValidator = new ClassValidator(yourBeanClass.getClass());
InvalidValue[] validationMessages = requestValidator.getInvalidValues(yourBeanClass);
if (validationMessages != null && validationMessages.length > 0) {
for (int i = 0; i < validationMessages.length; i++) {
//Add a validation message to be displayed to the user
}
return false;
}
return true;
}