I have an NSTableView with a set of rows. There are two types of rows - titles and data - which are displayed differently.
I am dragging an object from within my app onto the data rows. Drag and drop works fully and only data rows accept the drop - but there is a display issue.
I am using a custom NSTableRowView, and what I expect to see is that if I drag my object over a data row, then DrawDraggingDestinationFeedback should be called (it is) to draw my custom background used during drag operations. If I then drag my object out of the data row, I expect DrawBackground to be called (it is SOMETIMES) to redraw my standard background.
Heres the detail:
If I drag my object onto a data row and then completely away from the NSTableView, DrawDraggingDestinationFeedback is called on entry and DrawBackground is called on exit.
If I drag my object onto a row, and then down to the next row,
DrawDraggingDestinationFeedback is called for the top row, then when I move down DrawDraggingDestinationFeedback is called for the lower row, but DrawBackground is not called for the first row (i.e. its background color is not reset). This leaves BOTH rows looking like they are active drop destinations.
Weirdly, the final DrawBackground on exiting the table is called.
Here is my custom NSTableRowView:
public class CustomSelectionRowView : NSTableRowView
{
/// The log object
private static Logger log = LogManager.GetCurrentClassLogger();
private string type;
private DataItem ditem;
public CustomSelectionRowView(string type, DataItem ditem = null) : base()
{
this.type = type;
this.ditem = ditem;
DraggingDestinationFeedbackStyle = NSTableViewDraggingDestinationFeedbackStyle.Regular;
SelectionHighlightStyle = NSTableViewSelectionHighlightStyle.Regular;
}
public override void DrawDraggingDestinationFeedback(CGRect dirtyRect)
{
log.Debug(ditem.title);
if (type == "Data")
{
// This can accept a drag-drop
var color = NSColor.Gray;
var p = NSGraphicsContext.CurrentContext.GraphicsPort;
p.SetFillColor(color.CGColor);
p.FillRect(dirtyRect);
}
else
base.DrawDraggingDestinationFeedback(dirtyRect);
}
public override void DrawBackground(CGRect dirtyRect)
{
log.Debug(ditem.title);
if (type == "Data")
{
// We definitely have a Data block
var color = NSColor.FromName("MainPaneDataBackground");
var p = NSGraphicsContext.CurrentContext.GraphicsPort;
p.SetFillColor(color.CGColor);
p.FillRect(dirtyRect);
}
else
base.DrawBackground(dirtyRect);
}
}
Anyone have any ideas of how to work around this?
Many thanks.
[p.s. note that while my code here is C#/Xamarin... any swift help would be gratefully accepted - I speak both]
Related
I have a JFace TableViewer with an ILazyContentProvider and a StyledCellLabelProvider for each column, which I mostly grabbed from https://wiki.eclipse.org/JFaceSnippets#Snippet006TableMultiLineCells to enable multiline rows. When I open the table, all rows have the height of the row which takes up the most space, as intended. As I scroll down the table, the row heights will change as intended if a row takes up more space. However, this does not currently work in the other direction, i.e., if I scroll so that the current rows showing take up less space, all rows still have the height of the largest row in the whole table.
Is there a way to solve this? Somehow there seems to be a memory of the content that the lazy content provider should be forgetting?
This is my measure method in the StyledCellLabelProvider:
#Override
protected void measure(Event event, Object element) {
event.width = viewer.getTable().getColumn(event.index).getWidth();
if (event.width == 0) {
return;
}
TableEntryData rowData = (TableEntryData) element;
TableCellData cellData = getCellData(rowData, event.index);
int height = event.gc.textExtent(SOME_STRING).y; // Height of a written string on one line.
int lines = cellData.getPoints().size();
event.height = height * lines;
event.gc.dispose();
}
and this is most of my ILazyContentProvider:
#Override
public void updateElement(int index) {
viewer.replace(entries.get(index), index);
}
#SuppressWarnings("unchecked") // TODO:
#Override
public void inputChanged(Viewer viewer, Object oldInput, Object newInput) {
this.entries = (List<TableEntryData>) newInput;
}
The measure item code in the Table class (used by TableViewer) will never reduce the size of rows once it has grown bigger, so there is no way to change this.
The measure code is Table.sendMeasureItem but it can't be overridden. The code here is platform specific, but I checked both the Windows and macOS version (not sure about the Linux/GTK version).
I do have a hack to work around this, but it is platform dependent and I only have it for macOS.
I wrote some code to add a button to my custom editor in Unity. What I'd like is the following. When I click the button once, a component is added, and when I click again the component is removed. In the simple code below, I just try to print "Add" or "Remove" when clicking the button. I noticed that the variable toggleRigidBody takes the value true but then take the value false right after. It doesn't stay "true" even though I never explicitly change it in the code. The "else if" is never fired. I'm not sure why.
using UnityEditor;
using UnityEngine;
using SS3D.Engine.Inventory;
[CustomEditor(typeof(ContainerController))]
public class ContainerControllerEditor : Editor
{
private bool toggleRigidBody = false;
public override void OnInspectorGUI()
{
DrawDefaultInspector();
ContainerController containerController = (ContainerController)target;
Debug.Log(toggleRigidBody);
bool buttonPressed = GUILayout.Button("AddAttachedContainer");
if (buttonPressed && toggleRigidBody == false)
{
Debug.Log("Add");
toggleRigidBody = true;
}
else if (buttonPressed && toggleRigidBody == true)
{
Debug.Log("Remove");
toggleRigidBody = false;
}
}
}
My code only print "Add" when I click the button. What's happening here ?
The main problem here is that the editor instance is created when the object is clicked and the Inspector loaded. And then it is destroyed as soon as the object loses focus and the Inspector not shown for this object anymore.
=> Your flag toggleRigidBody is not persistent!
What you rather want to do is serialize the flag inside your object or even better: Serialize the reference itself.
This way you
Have already access to the reference in your script in case you need it on runtime
Have the reference in the editor for a) checking if it exists and b) being able to remove it directly
So having your class like
public class ContainerController : MonoBehaviour
{
// Store the reference in a field
[SerializeField] private Rigidbody _rigidbody;
...
}
The editor could look like
[CustomEditor(typeof(ContainerController))]
public class ContainerControllerEditor : Editor
{
private SerializedProperty _rigidbody;
ContainerController containerController;
private void OnEnable()
{
_rigidbody = serializedObject.FindProperty("_rigidbody");
containerController = (ContainerController)target;
}
public override void OnInspectorGUI()
{
DrawDefaultInspector();
// Loads all current serialized values from the target into the serialized properties
serializedObject.Update();
// If the _rigidbody field is not assigned
// try GetComponent as fallback
if(!_rigidbody.objectReferenceValue) _rigidbody.objectReferenceValue = containerController.GetComponent<Rigidbody>();
// simply put everything that belongs to one button click inside one if block
// this is easier to maintain and read
// Of course alternatively you could also simply have two completely different buttons to display
// depending on the value of "_rigidbody.objectReferenceValue"
if(GUILayout.Button(_rigidbody.objectReferenceValue ? "Remove Rigidbody" : "Add Rigidbody")
{
// Is there a Rigidbody?
if(_rigidbody.objectReferenceValue)
{
// Yes -> destroy it
// There are two different destroy methods depending whether you are
// in Play mode or Edit mode
if(Application.isPlaying)
{
Destroy(_rigidbody.objectReferenceValue);
}
else
{
DestroyImmediate(_rigidbody.objectReferenceValue);
}
}
// Otherwise the field is currently not set and no component was found using GetComponent
else
{
// Add the component via the ObjectFactory
// this enabled undo/redo and marks the scene dirty etc
// and assign it to the serialized property
_rigidbody.objectReferenceValue = ObjectFactory.AddComponent<Rigidbody>(target.gameObject);
}
}
// Writes back all modified properties to the target and takes care of Undo/Redo and marking dirty
serializedObject.ApplyModifiedProperties ();
}
}
The editor object is created when it's being displayed, and destroyed when it's not, so for your data to persist you will need to store the values somewhere else. So that's definitely what is happening. The easiest way to have a value of a variable to be persistent between sessions you would use EditorPrefs to save the variable values. Thats the easiest way you can go about it, so you would use this to save the toggleRigidBody value.
https://docs.unity3d.com/ScriptReference/EditorPrefs.SetBool.html
Here I create two objects form my Custom control and add them to the form but it show the first one only, why ?
private void frmWelcome_Load(object sender, EventArgs e)
{
// Count number of Contacts in XML
XmlDocument doc = new XmlDocument();
doc.Load(#"C:\Users\Taha\Documents\Visual Studio 2013\Projects\ContactsMangementSystem\ContactsMangementSystem\ContactsData.xml");
int numOfContacts = doc.GetElementsByTagName("Contact").Count;
XmlNodeList nodelist = doc.GetElementsByTagName("Contact");
foreach (XmlNode item in nodelist)
{
// Create Control for each Item
ContactControl.ContactControl userControl = new ContactControl.ContactControl();
// Fill Control With Data
userControl.labelOrganizationText = item.ChildNodes.Item(5).InnerText;
userControl.labelTitleText = item.ChildNodes.Item(0).InnerText;
userControl.labelWorkAddressText = item.ChildNodes.Item(12).InnerText;
userControl.labelWorkPhoneText = item.ChildNodes.Item(8).InnerText;
userControl.labelEmailText = item.ChildNodes.Item(7).InnerText;
Image img = Image.FromFile(item.ChildNodes.Item(9).InnerText);
userControl.ContactImage = img;
// Add item to the form
this.Controls.Add(userControl);
}
ContactControl.ContactControl userControl2 = new ContactControl.ContactControl();
this.Controls.Add(userControl2);
}
If you add controls all by yourself, aka programmaticaly, to the Control collection of a form, you become responsible for handling the position, bounds, size etc. I'll demonstrate the root cause of your issue with a trimmed down code example that is similar to yours.
First I create a UserControl, with one label on it. Set the label property Dock to Fill, Font to Point 28, TextAlign to MiddelCentre and most important Modifiers to Public:
Now we create a new Form in the Designer and add a Timer, a Button, two checkboxes and a FlowLayoutPanel to it. The FlowLayoutPanel has its Anchor property set to Top,Left,Right,Bottom so it will resize if the form resizes. Our design looks like this:
In the Timer_Tick event a new instance of the UserControl is created and either added to the Forms Control collection or the FlowLayoutPanel. There are two variables that keep track of the number of controls created and on which position the control needs to be.
Here is the code in the codebehind of the form:
int position = 0; // what position the control should be
int controlsCreated = 0; // how many controls did we create
private void timer1_Tick(object sender, EventArgs e)
{
// create the control
var ucnext = new UserControl1();
controlsCreated++;
ucnext.label1.Text = controlsCreated.ToString();
// where are we adding the control?
if (!chkFlow.Checked)
{
// do we want to use the Location?
// if not Top will be 0
if (chkLocation.Checked)
{
// Ok, position times height should gives us
// a free spot
ucnext.Top = position * ucnext.Height;
position++;
}
this.Controls.Add(ucnext);
ucnext.BringToFront(); // otherwise we always see control 1
} else
{
// now the FlowLayout takes over all Layout logic
this.flowLayoutPanel1.Controls.Add(ucnext);
}
}
private void button1_Click(object sender, EventArgs e)
{
this.timer1.Enabled = !this.timer1.Enabled;
this.btnStart.Text = this.timer1.Enabled ? "Pause" : "Continue";
}
If we run this we can check and uncheck the options for Location and Adding to the flowlayout to see the various effects of adding controls to either the form directly or the FlowLayout panel.
So keep in mind if you add a control yourself to the Control collection of a Form make sure to set its Location properties (Top, Left) to a suitable value. If you do this wrong your control is going to end up hidden behind another control or is placed outside the visible bounds of the form. If you don't know before hand how many controls you will have consider using one of the container controls, like a FlowLayoutPanel.
I'm learn as3, and building some exercises from the adobe online tutorials. I'm trying to do a very simple task and I think my code is exactly the same as the tutoriala nd it's not working! I just can't work it out...
I'm trying to change the textfield in a button from that buttons class.
The button is called GameButton, the textfield is called blabel and is classic text and dynamic text. Here is the code for the GameButton class. Instead of displaying "Click" as below it just changes the label to the number 1. The trace statement is working etc it is going there, but the text isn't passing through or something. Please help!!!
package {
import flash.display.MovieClip;
public class GameButton extends MovieClip {
public function GameButton() {
trace("Gamebutton has been created");
this.blabel.text = "Click";
stop();
}
}
}
The long and short of it is you can create the button in code, or else you can try listening for added to stage events coming from the parent object you're adding the children to (maybe the stage, maybe another DisplayObjectContainer). The problem with the listening method is I'm not sure how you would know which child just dispatched the event without making some messy code. I think the first option is generally easier and makes more sense, the only caveat is that you have to place the instances using x/y coordinates or apply scaleX, scaleY to stretch or shrink objects instead of doing it using the authoring tool. You can still use the drag and drop parts of flash to figure out coordinates and build individual movie clips etc.
Enough talk on to some code:
package
{
import flash.display.MovieClip;
public class GameButton extends MovieClip {
private var blabel:TextField; //This can be any display object class or a class that extends from a display object class (Sprite, MovieClip, MyCustomButton etc.)
public function GameButton() {
blabel = new TextField(); //The type after new, should be the same, or a sub-class (extension) of the type used in the variable declaration above
addChild(blabel);
//blabel.x=10; //optional over ten pixels from left
//blabel.y=10; //optional down ten pixels from top
//blabel.scaleX=.5; //optional half the width
//blabel.scaleY=2; //optional 2 times taller
trace("Gamebutton has been created");
blabel.text = "Click";
stop();
}
}
}
I'm a beginner of ActionScript 3.0. I'm making a simple memory game, the tool I'm using is Eclipse with flexsdk plugin. Right now I've done the shuffle and display images, and the cover of the images as well.
My idea is when clicking on the image, the cover will remove and show the image behind of it. After 2 covers are gone, the game will compare and check whether the selected images are match or not, if match both of the images will remain, otherwise the cover will reappear and the game keeps going on. If all of the images are match, a winning line will appear.
The problem I'm facing is I got no idea on how to deal with the images comparison part. I wanted to compare with the index number of array or the name of the images, but I really don't have any idea. I've refer some examples but all of them are develop in CS3 and uses the timeframe which is not exist in pure ActionScript.
package {
import flash.display.Sprite;
import flash.events.MouseEvent;
import myclasses.Cover;
public class Prototype extends Sprite {
protected static const WIDTH:int=3;
protected static const HEIGHT:int=2;
protected static const SPACINGX:int=100;
protected static const SPACINGY:int=74;
protected static const OFFSETX:int=96;
protected static const OFFSETY:int=100;
protected static const SIZE:Number=100;
protected static const COLOUR:uint=0x999999;
[Embed(source="images/pic1.jpg")]
protected static const PIC1:Class;
[Embed(source="images/pic2.jpg")]
protected static const PIC2:Class;
[Embed(source="images/pic3.jpg")]
protected static const PIC3:Class;
protected var imagesList:Array=[PIC1,PIC1,PIC2,PIC2,PIC3,PIC3];
protected var X:int;
protected var Y:int=27;
protected var count:int;
//protected var firstTap:Class;
//protected var secondTap:Class;
public function Prototype() {
var shuffled:Array = shuffleList(imagesList.length);
for(var i:int; i<imagesList.length; i++) {
//var colour:uint;
//colour=0x999999;
var j:int = shuffled[i];
var thing:Sprite=new Sprite();
thing.addChild(new imagesList[j]());
thing.x=X;
thing.y=Y;
addChild(thing);
new Cover(this,X,Y,SIZE,COLOUR);
X=X+SPACINGX+OFFSETX;
count++;
if(count == WIDTH){
Y=Y+SPACINGY+OFFSETY;
X=0;
}
addEventListener(MouseEvent.MOUSE_DOWN,selectImages);
}
}
public function selectImages(event:MouseEvent):void {
//(P/S: this is not the actual code)
var target:Sprite = Sprite(event.target)
if(firstTap == null){
firstTap = ; //1st selected image
removeChild(target);
}else if(secondTap == null){
secondTap = ; //2nd selected image
if(firstTap == secondTap){
firstTap = null;
secondTap = null;
}else{
//firstTap = ; //cover back
//secondTap = ; //cover back
secondTap = null;
firstTap = ; //selected another image
}
}
}
protected function shuffleList(n:Number):Array {
var startList:Array = new Array(n);
var endList:Array = new Array(n);
var i:Number, j:Number;
for (i=0; i<n; i++) {
startList[i] = i;
}
for (i=0; i<n; i++) {
j = Math.floor(Math.random()*startList.length);
endList[i] = startList.splice(j,1)[0];
}
return(endList);
}
}
}
Please help me figure out. Thanks.
Well I didn't really go through your code in details, but from looking at it, it seems you are making things really complex. So let me give you a hint in how I would do it...
1) First I would make a 'card' class, which extends sprite or bitmap, mainly containing the image of that card, image of the cover/mask, some animtion methods like hide/show e.t.c, and possibly an 'id' variable to recognize it later, though that is not needed. You can also get away with the array you have, and in that case, skip step 2.
2) Now push two copies of each card in an array.
example [new card('c1'),new card('c1'),new card('c2'),new card('c2'),.....].
3) Now comes the part where you made it most complex, that is, the shuffling of the array. Let's write a custom function for it, shall we?
function shuffleArr(arr):Array
{
var len:int = arr.length();
for(var i:int=0; i<len; i++)
{
//Swap the value at i with a random value within the array;
var tmp_pos:int = Math.floor(Math.random()*len);
while(tmp_pos==i)
{
tmp_pos = Math.floor(Math.random()*len);
}
var tmp_var:card = arr[tmp_pos];
arr[tmp_pos] = arr[i];
arr[i] = tmp_var;
}
}
4) Now that the array is shuffled, you simply have to lay them out in a grid.
for(var row:int=0; row<6; row++)
{
for(var col:int=0; col<6; col++)
{
card_arr[i].x = card_arr[i].width*cols+5;
card_arr[i].y = card_arr[i].height*row+5;
stage.addChild(card[i]);
}
}
5) Now you have to check for user click and take action, and there are many ways to do it, but I will tell one of them...
a) Give the cards a click event handler, this becomes easy if the cards are a class, or you can look into event.target property and use a general click handler. it's up to you.
b) On click, push the card's id in an array. If you did make them into a class, their ids should now be, c1, c2, e.t.c, and you can do
holder_arr.push(this);
this.removeCover();
6) Now you have to make sure, that the holder array can only hold two values at a time. Then do the checking. I am writing a semi-pseudo code with a lot of assumed functions and values:
//Insert
if(holder_arr.length()==2)
{
//flip back the cards and empty the array
holder_arr[0].showCover();
holder_arr[1].showCover();
holder_arr = [];
}
holder_arr.push(this);
...
..
7) For checking make a function and run it every time a card is clicked, and also when a timer ends, to flip back the cards.
function checkCards()
{
if(holder_arr.length==2)
{
if(holder_arr[0].id==holder_arr[1].id)
{
//the cards match
holder_arr[0].vanishAnim();
holder_arr[1].vanishAnim();
holder_arr=[];
}
else
{
holder_arr[0].showCover();
holder_arr[1].showCover();
holder_arr=[];
}
}
}
Obviously you will remove the cards from the actual card_arr too, but when to do that is up to you to figure out ;)
Personally , I find the problem has more to do with the way you shuffle your cards as it makes it a bit more difficult to identify what card is what.
Let's say for instance that instead of shuffling your embed assets , you were first creating your cards , naming them after the image they add , then creating an Array ( or Vector ) of cards and shuffle that Array ( or Vector ), identifying the cards would be fairly easy since you'd only have to compare their name property.
//PSEUDO CODE
- create Array ( Vector ) of embed assets [PIC1, PIC2 , PIC3]
- create cards and use Array ( Vector ) index to identify
each card with the name property
- create a set of six cards
- shuffle set