I would like to know if we can let the scroll box open after having selected an item. In other words, the same behavior of if we hold the CTRL key and a selected item.
I don't think it's possible without chosen script modification.
In jquery.chosen.js look for something like this (line ~955):
if (!((evt.metaKey || evt.ctrlKey) && this.is_multiple)) {
this.results_hide();
}
Modify it for your needs, maybe something like:
Add new option named stay_opened (line ~155):
this.stay_opened = this.options.stay_opened != null ? this.options.stay_opened : true;
And modify mentioned code:
if (!((evt.metaKey || evt.ctrlKey || this.stay_opened) && this.is_multiple)) {
this.results_hide();
}
Note that it will work for multiple mode only (but you can change that too).
Related
In the onChange of React-Bootstrap-Typeahead, I need to manually set a custom display value. My first thought was to use a ref and do something similar to the .clear() in this example.
But although .clear() works, inputNode.value = 'abc' does not work, and I'm left with the old selected value from the menu.
onChange={option => {
typeaheadRef.current.blur(); // This works
typeaheadRef.current.inputNode.value = 'abc'; // This does not work (old value is retained)
}}
I also tried directly accessing the DOM input element, whose ID I know, and doing
var inputElement = document.querySelector('input[id=myTypeahead]');
inputElement.value = 'abc';
But that didn't work either. For a brief second, right after my changed value = , I do see the new display label, but then it's quickly lost. I think the component saves or retains the menu-selected value.
Note: I cannot use selected, I use defaultSelected. I have some Formik-related behavior that I've introduced, and it didn't work with selected, so I'm stuck with defaultSelected.
The only workaround I found is to re-render the Typeahead component (hide and re-show, from a blank state) with a new defaultSelected="abc" which is a one-time Mount-time value specification for the control.
I couldn't get selected=.. to work, I have a wrapper around the component which makes it fit into Formik with custom onChange and onInputChange and selected wasn't working with that.
So the simple workaround that works is, if the visibility of the Typeahead depends on some condition (otherwise it won't be rendered), use that to momentarily hide and re-show the component (a brand new repaint) with a new defaultSelected, e.g.
/* Conditions controlling the visibility of the Typeahead */
!isEmptyObject(values) &&
(values.approverId === null || (values.approverId !== null && detailedApproverUserInfo)
)
&&
<AsyncTypehead defaultSelected={{...whatever is needed to build the string, or the literal string itself...}}
..
// Given the above visibility condition, we'll hide/re-show the component
// The below will first hide the control in React's renders
setFieldValue("approver", someId);
setDetailedUserInfo(null);
// The below will re-show the control in React's renders, after a small delay (a fetch)
setDetailedUserInfo(fetchDetailedUserInfo());
I am using the area select plugin. By default it responds to a ctrlKey box drag. And by default Leaflet boxZoom responds to a shiftKey box drag. All good so far. However a ctrlKey + shiftKey box drag triggers the Leaflet boxZoom and the area select plugin. I would like it to trigger just the area select plugin instead. Any suggestions?
If you have a look at Leaflet's source code for the BoxZoom map handler, you can see the line where it checks for a pressed shift key plus primary (""left"") mouse/pointer button to start the box zoom:
_onMouseDown: function (e) {
if (!e.shiftKey || ((e.which !== 1) && (e.button !== 1))) { return false; }
And you want to change that to check for ctrlKey, so that the box zoom doesn't start if it's set to true, something like:
if (!e.shiftKey || e.ctrlKey || ((e.which !== 1) && (e.button !== 1))) { return false; }
The question is how to do this without rewriting or breaking up everything. An approach is to monkey-patch that method from the BoxZoom handler's prototype while keeping a reference to the old one, e.g. something like:
var oldBoxZoomMouseDown = L.Map.BoxZoom.prototype._onMouseDown;
L.Map.BoxZoom.prototype._onMouseDown = function(ev) {
// Worry only about ctrlKey...
if (ev.ctrlKey) { return false; }
// ...and let the previous event handler worry about shift and primary button
oldBoxZoomMouseDown.call(this, ev);
}
Note that it'll work only when done before the map has been instantiated. There are other approaches, such as replacing the method of the BoxZoom instance after the map has been instantiated, and creating a subclass of the BoxZoom handler. Reading about javascript's prototypal inheritance and Leaflet's way of dealing with OOP is recommended at this point.
I want to hide the "top" part of all connector labels of a diagram. For this, I tried to set up a script, but it currently hides ALL labels (also the "bottom" labels which I want to preserve):
// Get a reference to the current diagram
var currentDiagram as EA.Diagram;
currentDiagram = Repository.GetCurrentDiagram();
if (currentDiagram != null)
{
for (var i = 0; i < currentDiagram.DiagramLinks.Count; i++)
{
var currentDiagramLink as EA.DiagramLink;
currentDiagramLink = currentDiagram.DiagramLinks.GetAt(i);
currentDiagramLink.Geometry = currentDiagramLink.Geometry
.replace(/HDN=0/g, "HDN=1")
.replace(/LLT=;/, "LLT=HDN=1;")
.replace(/LRT=;/, "LRT=HDN=1;");
if (!currentDiagramLink.Update())
{
Session.Output(currentDiagramLink.GetLastError());
}
}
}
When I hide only the top labels manually (context menu of a connector/Visibility/Set Label Visibility), the Geometry property of the DiagramLinks remains unchanged, so I guess the detailed label visibility information must be contained somewhere else in the model.
Does anyone know how to change my script?
Thanks in advance!
EDIT:
The dialog for editing the detailed label visibility looks as follows:
My goal is unchecking the "top label" checkboxes programmatically.
In the Geometry attribute you will find a partial string like
LLT=CX=36:CY=13:OX=0:OY=0:HDN=0:BLD=0:ITA=0:UND=0:CLR=-1:ALN=1:DIR=0:ROT=0;
So in between LLT and the next semi-colon you need to locate the HDN=0 and replace that with HDN=1. A simple global change like above wont work. You need a wild card like in the regex LLT=([^;]+); to work correctly.
I am trying to programmatically change the "highlighted" item in my CellList, based on keyboard events in another control (a textbox). I can get the keyboard events just fine, but when I use CellList...
int row = getKeyboardSelectedRow() + 1;
setKeyboardSelected(row,selected,stealFocus)
It doesn't have any affect. If I put focus on my cellList indeed, I can move up and down and highlight a row. If I put true for "stealFocus" on the setKeyboardSelected() method, it "works", except I really want to keep focus on the textbox, rather than the cellList. How do I do this? I've attempted fireEvents, sending the keystrokes to the CellList, but that doesn't help either.
J
I have no immediate solution, but if I were you I'd step through the code of setKeyBoardSelected in com.google.gwt.user.cellview.client.AbstractCellTable in debug mode.
Maybe this is where it's going wrong?
if (KeyboardSelectionPolicy.DISABLED == getKeyboardSelectionPolicy()
|| !isRowWithinBounds(index) || columns.size() == 0) {
return;
}
I am using a TableViewer with a content provider, label provider, a ICellModifier and TextCellEditors for each column.
How can I add arrow key navigation and cell editing when the user selects the cell? I would like this to be as natural a behavior as possible.
After looking at some of the online examples, there seems to be an old way (with a TableCursor) and a new way (TableCursor does not mix with CellEditors??).
Currently, my TableViewer without a cursor will scroll in the first column only. The underlying SWT table is showing cursor as null.
Is there a good example of TableViewer using CellEditors and cell navigation via keyboard?
Thanks!
I don't know if there is a good example. I use a cluster of custom code to get what I would consider to be basic table behaviors for my application working on top of TableViewer. (Note that we are still targetting 3.2.2 at this point, so maybe things have gotten better or have otherwise changed.) Some highlights:
I do setCellEditors() on my TableViewer.
On each CellEditor's control, I establish what I consider to be an appropriate TraverseListener. For example, for text cells:
cellEditor = new TextCellEditor(table, SWT.SINGLE | getAlignment());
cellEditor.getControl().addTraverseListener(new TraverseListener() {
public void keyTraversed(TraverseEvent e) {
switch (e.detail) {
case SWT.TRAVERSE_TAB_NEXT:
// edit next column
e.doit = true;
e.detail = SWT.TRAVERSE_NONE;
break;
case SWT.TRAVERSE_TAB_PREVIOUS:
// edit previous column
e.doit = true;
e.detail = SWT.TRAVERSE_NONE;
break;
case SWT.TRAVERSE_ARROW_NEXT:
// Differentiate arrow right from down (they both produce the same traversal #*$&#%^)
if (e.keyCode == SWT.ARROW_DOWN) {
// edit same column next row
e.doit = true;
e.detail = SWT.TRAVERSE_NONE;
}
break;
case SWT.TRAVERSE_ARROW_PREVIOUS:
// Differentiate arrow left from up (they both produce the same traversal #*$&#%^)
if (e.keyCode == SWT.ARROW_UP) {
// edit same column previous row
e.doit = true;
e.detail = SWT.TRAVERSE_NONE;
}
break;
}
}
});
(For drop-down table cells, I catch left and right arrow instead of up and down.)
I also add a TraverseListener to the TableViewer's control whose job it is to begin cell editing if someone hits "return" while an entire row is selected.
// This really just gets the traverse events for the TABLE itself. If there is an active cell editor, this doesn't see anything.
tableViewer.getControl().addTraverseListener(new TraverseListener() {
public void keyTraversed(TraverseEvent e) {
if (e.detail == SWT.TRAVERSE_RETURN) {
// edit first column of selected row
}
}
});
Now, how exactly I control the editing is another story. In my case, my whole TableViewer (and a representation of each column therein) is loosely wrapped up in a custom object with methods to do what the comments above say. The implementations of those methods ultimately end up calling tableViewer.editElement() and then checking tableViewer.isCellEditorActive() to see if the cell was actually editable (so we can skip to the next editable one if not).
I also found it useful to be able to programmatically "relinquish editing" (e.g. when tabbing out of the last cell in a row). Unfortunately the only way I could come up with to do that is a terrible hack determined to work with my particular version by spelunking through the source for things that would produce the desired "side effects":
private void relinquishEditing() {
// OMG this is the only way I could find to relinquish editing without aborting.
tableViewer.refresh("some element you don't have", false);
}
Sorry I can't give a more complete chunk of code, but really, I'd have to release a whole mini-project of stuff, and I'm not prepared to do that now. Hopefully this is enough of a "jumpstart" to get you going.
Here is what has worked for me:
TableViewerFocusCellManager focusCellManager = new TableViewerFocusCellManager(tableViewer,new FocusCellOwnerDrawHighlighter(tableViewer));
ColumnViewerEditorActivationStrategy actSupport = new ColumnViewerEditorActivationStrategy(tableViewer) {
protected boolean isEditorActivationEvent(ColumnViewerEditorActivationEvent event) {
return event.eventType == ColumnViewerEditorActivationEvent.TRAVERSAL
|| event.eventType == ColumnViewerEditorActivationEvent.MOUSE_DOUBLE_CLICK_SELECTION
|| (event.eventType == ColumnViewerEditorActivationEvent.KEY_PRESSED && event.keyCode == SWT.CR)
|| event.eventType == ColumnViewerEditorActivationEvent.PROGRAMMATIC;
}
};
I can navigate in all directions with tab while editing, and arrow around when not in edit mode.
I got it working based on this JFace Snippet, but I had to copy a couple of related classes also:
org.eclipse.jface.snippets.viewers.TableCursor
org.eclipse.jface.snippets.viewers.CursorCellHighlighter
org.eclipse.jface.snippets.viewers.AbstractCellCursor
and I don't remember exactly where I found them. The is also a org.eclipse.swt.custom.TableCursor, but I couldn't get that to work.
Have a look at
Example of enabling Editor Activation on a Double Click.
The stuff between lines [ 110 - 128 ] add a ColumnViewerEditorActivationStrategy and TableViewerEditor. In my case the I wanted a single click to begin editing so i changed line 115 from:
ColumnViewerEditorActivationEvent.MOUSE_DOUBLE_CLICK_SELECTION
to ColumnViewerEditorActivationEvent.MOUSE_CLICK_SELECTION. After adding this to my TableViewer, the tab key would go from field to field with the editor enabled.