I want to validate a form that consists of a parent component and a child component. The user can dynamically add another child component on click of a button. I am unable to check the invalid state of the the child components and therefore also the invalid state of the parent component.
Previously, this was not an issue since Vuelidate had $each function.
Related
When program fails to find a component in a test case, a failure trace is shown. It shows attributes of components in the hierarchy. For example:
[name=null, text='Close', enabled=true, visible=true, showing=false]
I wonder what value of visible and showing tell us about the component.
This are properties from the java.awt.Component class.
The JavaDoc for these properties (https://docs.oracle.com/javase/8/docs/api/java/awt/Component.html) has the following information:
visible:
Determines whether this component should be visible when its parent is visible.
showing:
Determines whether this component is showing on screen. This means that the component must be visible, and it must be in a container that is visible and showing.
I have a parent component where the view model contains a property prop1.
In its view, a custom element has a view-model.ref="prop1".
export class Parent {
public prop1;
}
<template>
<custom-element view-model.ref="prop1"></custom-element>
</template>
This works like a charm and I get a reference to the view model of my custom element in my parent.
Now, I add a child router to parent with a child component. In child's view model, I add a property prop1. and in its view, a custom element has a view-model.ref="prop1". So exactly like I did in parent...
As soon as I navigate to parent/child, the parent's container prop1 stops referencing the custom element of parent and starts referencing the one from child.
If I name the properties differently, there is no problem.
Any idea why this happens? And how could I avoid this behavior without worrying about the naming of the properties?
EDIT
I chanced upon some more information! If the properties are initialized in the view model, I seem to be able to retain the references in the view models. Note that I'm using Typescript so I think the compiled code for an unassigned class property doesn't mention the property at all until it is assigned.
I still don't really understand where the problem comes from exactly...
And I remain with the same problem if I use a view-model.ref directly in the template without mapping it to an explicit property from the view model like this:
<template>
<custom-element view-model.ref="custom"></custom-element>
<custom-element2 opened.call="custom.opened()"></custom-element2>
</template>
when you create a property in a class and don't assign anything to it, babel/typescript will remove that property as if it was not even declared.. because it's really not doing anything.
typescript is interested in your definitions in compile time only.
now that your property in the child is emitted, you have a binding in the child to an undeclared property.
in that case, aurelia creates a property for you and bind to it..
but in your case, aurelia finds this property in the parent scope (just like regular JS scoping rules), and therefor do not create a "new property" in the child scope but rather binds to the parent property.
this is why when you initialize the property in the child (even with undefined) it "works". because you have 2 properties in 2 scopes.
and when you change the name, aurelia will create a new property for you..
and here come another rule - wen aurelia create a property for you - it creates it in the lowest scope available.. in your case - the child scope. and everything works again.
you can notice this behavior a lot with repeate.for, because the repeater creates an invisible scope for each repeate loop..
if you bind anything in the repeate to a property that dont exists, you will have that property in each child, and not once in the parent.
I have a parent component that contains a table, and a child component that contains a bootstrap modal.
When a row in the table is clicked the modal should pop up with many functions and filtered select-drop-boxes that matches the clicked row.
From my understanding, these three are my options:
A. Opening the modal with a function inside the child component (the modal) via #Viewchild. When closed, another function in the child component should clear the properties and fields in the modal.
pros: The DOM will not have to recreate the child component each time a row is clicked.
cons: If the modal is not needed it will be created unnecessarily. + I have to reset the fields on close.
B. Use *ngIf on the child component (the modal). When a row is clicked, the if statement becomes true and in the ngOnInit the modal will call a method to pop up the modal.
pros: No need to reset the fields, everything will be set in the ngOnInit.
cons: will be created each time a row is clicked.
C. use a message service between the parent and the child that will trigger show/hide on the modal.
cons: a service needs to be built for this.
I would like to understand what is considered a good practice and what's a bad practice in these approaches. Or alternative ways to achieve this. I'm working on a big project and need to do this a lot.
I'm trying to understand host and target (and element) in the following context taken from the Polymer Path and Polymer Data Flow documentation.
Consider the following diagram:
Now consider the following statement (from the same documentation):
"When two elements in the local DOM are bound to the same property data appears to flow from one element to the other, but this flow is mediated by the host."
So far, so good. Then it goes on to say:
"A change made by one element propagates up to the host, then the host propagates the change down to the second element."
The first part: "A change made by one element propagates up to the host..." Does this mean that a change to the first element propagates to its own host first? And does "element" actually mean the element's data properties?
The second part "then the host propagates the change down to the second element." Are we propagating down to the second element's data properties? It's extra confusing here as there is only one element or data object that is shared between the two ehhh elements??
I'm thinking that the change made in the first element's data property goes to its own host first and then the first host propagates the change back down to the second element's data element (which so happens to be the first element's data object as well).
<parent-el>
<user-profile primary-address="{{addr}}"></user-profile>
<address-card address="{{addr}}"></address-card>
</parent-el>
If either element changes addr (the child elements can use whatever name they want), the change will be propagated to the parent and then to the other element.
If either binding used [[addr]] instead, changes would only propagate from parent to child.
Note that both child elements should have notify: true set on the relevant property (primaryAddress or address) so that the parent is notified of changes and the two-way binding is fully setup.
Also note that this listens for the object to change as a whole only. To listen for changes to sub-properties e.g. addr.street the parent should add an observer. For more info on that see complex observers.
Can parent-child directive communication using 'require' be used if there are multiple instances of both parent and child directives? (i.e. both directives are inside ng-repeat blocks)
I have a parent child directive relationship, but the parent directive is included in an ng-repeat block. When I use 'require' on the child directive it always binds to the first instance of the parent directive and not the first one above the child directive in the DOM.
Yes it is possible; as nice visual example (I'm a visual person) angular-gridster has parent and children directives. If we just throw them in a ng-repeat everything works. This is because ng-repeat has its own scope in which it creates new child scopes for each item.
See this plunker