I need help with my VS Code extension. I've written a custom view which works just fine, however I'd like to activate / focus / bring into view that view by using a keyboard shortcut or a context menu command. I am unable to find how to use the VS code API to achieve that.
context.subscriptions.push(vscode.commands.registerCommand('extensionId.showView', () =>
{
// how to do that?
}));
I know this can be done, because one can display the file explorer by using this code snppet:
vscode.commands.executeCommand('workbench.view.search');
But how would you do that for a custom tree view?
You should be able to use the new focus option that was added to TreeView.reveal() in 1.25 for that. The method requires you to pass a tree item to be revealed, so it's more of a workaround for not being able to focus the view itself directly, but you could simply pass the first / root node.
treeView.reveal(item, {focus: true});
Note that focus in this case means keyboard focus. If you just want to bring it into view, calling reveal() without the focus option is good enough.
To obtain a TreeView instance, you need to call vscode.window.createTreeView() with your view ID and provider.
I think, it is also possible to use
vscode.commands.executeCommand("exampleView.focus")
using the exampleView declared in package.json:
...
"views": {
"exampleView": [
{
"id": "exampleView",
"name": "Example View"
}
]
},
...
as #Empiire said, focus command in the form of : vscode.commands.executeCommand
and adding the '.focus' at the end of the id of the view that was declared in the package.json works!
You can also provide an object as parameter such as { preserveFocus: true } to show without disturbing the current focus.
eamodio does it like that in gitlens : https://github.com/gitkraken/vscode-gitlens/blob/417587d0dfcda89e9e2d723f8b662d7cf9008c8f/src/webviews/webviewViewBase.ts#L85
Related
I faced an issue that is combined with UI testing of the view.
I have an issue with an incorrect path to the element that is the text field, which is Search. I would like to achieve the state when the code for app.navigationBars will be resistant for different languages. That's why right now it looks the following way:
var airportsSearchTextField: XCUIElement { app.navigationBars["Airports"].searchFields["Search"] }
This won't work properly for another language like English. I know that it could be possible to add accessibilityIdentifier, but I'm not sure where and maybe for navigationBar is there any other way to implement a better solution to grab this element within UI tests?
func selectFlight() { airportsSearchTextField.tap() }
I have created multipage HTML editor in which one tab has text editor. I have set the global action handler to the action bar for undo / redo actions in the source editor. Whenever I am adding something in the source editor then undo it, it is not returning the same code in the first attempt. It is completing in the second attempt. Can anyone help me to solve this issue.
In the main editor override the setFocus() method and in this method call the following method of source editor.
public void setUndoRedoActionHandlers() {
final IActionBars actionBars = getEditorSite().getActionBars();
actionBars.setGlobalActionHandler(ActionFactory.UNDO.getId(),
mUndoAction);
actionBars.setGlobalActionHandler(ActionFactory.REDO.getId(),
mRedoAction);
actionBars.setGlobalActionHandler(ActionFactory.DELETE.getId(),
mDeleteAction);
actionBars.setGlobalActionHandler(
ITextEditorActionConstants.DELETE_LINE, mDeleteLineAction);
actionBars.updateActionBars();
}
Thanks
Are you sure that you set global actions by overriding setFocus()? Look at this : https://wiki.eclipse.org/FAQ_How_do_I_enable_global_actions_such_as_Cut,_Paste,_and_Print_in_my_editor%3F
It says they need to be set inside method setActiveEditor()
The reason I guess why you are seeing it working second time is - When the first time you click your editor gets focus and setFocus is called and only then the global actions are set. Then second time it will work because the actions are now set.
I'm trying to configure the width for Alfresco.util.PopupManager.displayPrompt() but I don't see how to do it.
YUI Dialog has a width property, but the only config that I've managed to see, defaultDisplayPromptConfig object, doesn't seem to pay much attention to my messing with it.
Specifically, I tried setting Alfresco.util.PopupManager.defaultDisplayPromptConfig.width but it didn't work :)
I'm also trying to style up the panel I'm loading (create-site style injected panel), but it does not work for now.
Is there a way to configure the PopupManager Prompt object?
TIA
If you look at the source for Alfresco.util.PopupManager.displayPrompt() contained in alfresco.js then you'll see that the method creates an instance of YAHOO.widget.SimpleDialog, which does support a width property.
The problem is that displayPrompt() takes only specific configuration options which it passes through to the SimpleDialog, so adding an additional width parameter to your config will not have any effect, as you can see.
// Create the SimpleDialog that will display the text
var prompt = new YAHOO.widget.SimpleDialog("prompt",
{
close: c.close,
constraintoviewport: c.constraintoviewport,
draggable: c.draggable,
effect: c.effect,
modal: c.modal,
visible: c.visible,
zIndex: this.zIndex++
});
// Show the title if it exists
if (c.title)
{
prompt.setHeader($html(c.title));
}
// Show the prompt text
prompt.setBody(c.noEscape ? c.text : $html(c.text));
// Show the icon if it exists
if (c.icon)
{
prompt.cfg.setProperty("icon", c.icon);
}
// Add the buttons to the dialog
if (c.buttons)
{
prompt.cfg.queueProperty("buttons", c.buttons);
}
// Add the dialog to the dom, center it and show it.
prompt.render(parent);
prompt.center();
prompt.show();
I like the idea of enhancing the function to support a width property and possibly others, but in the meantime you are best off using SimpleDialog directly in your own code, modifying the above to add a width parameter.
I'm using a CellTree with the KeyboardSelectionPolicy.BOUND_TO_SELECTION.
I'd like to open the tree with a given path selected.
The code opening the child path and selecting the node is working fine when KeyboardSelectionPolicy is ENABLED/DISABLED but when BOUND_TO_SELECTION I can see that the keyboard-selected-node in the tree is never updated from : cellTree.selectionModel.setSelected( ... )
So I'm wondering if setSelected can work with BOUND_TO_SELECTION and how to do it.
See http://code.google.com/p/google-web-toolkit/issues/detail?id=6310
I suppose you could find a way to call setKeyboardSelectedRow in response to SelectionChangeEvents.
I've been through the Dojo docs as well as the API and tried Google but can't find a solution to my problem, I hope anybody here can help me out.
I'm trying to create a dijit.form.select programmatically (using Dojo 1.4) and connect to the "onClick"-event of the widget.
Here's a part of my code:
var dataSelect = new dijit.form.Select({
id : "myselect",
name : "myselect",
labelAttr: "label",
labelType: "html"
},
"selectid");
dataSelect.addOption({value: "value", label: "first item label"});
dojo.connect(dataSelect, "onClick", function() {
alert("clicked!");
});
What it does: A select-box is created replacing an input-field with the ID "selectid", an option "first item label" is created. Everythings all right until here.
Then I connect to the "onClick"-event of the select, which is supposed to load more options via AJAX (but will just display an alert for testing purposes in this example).
The problem: When I click on the little arrow next to the dropdown, the event is fired (OK). But when I click on the select box itself (the area containing the option), the event is NOT fired the first time I click it (unless I clicked on the arrow before).
When I click the select box a second time (and every time after that), the event will fire!
I've tried to use "onFocus" instead of "onClick" which does work, but then the dropdown will not open when first clicked, even if I use the "openDropDown"-function (which does work when connecting to "onClick"!).
Is it me, did I run into a Dojo bug or is it a strange feature I just don't get?
Any help is appreciated.
Greetings,
Select0r
Here is a cross-browser solution:
var sizeSelect = new dijit.form.Select({
id: "sizeSelect",
name: "state",
options: size,
onChange: function(val) {
dojo.style(dojo.byId("textInput"), {"fontSize":val});
}
}, "sizeSelect");
Try to connect not to the widget itself but to it's dom node:
dojo.connect(dataSelect.domNode, "onclick", function() {
alert("clicked!");
});
Select (which extends _HasDropDown) has fancy code to handle:
mouse down on select widget
mouse move to one of the options
mouse up
Maybe that's canceling the click event.
Maybe you can connect to _loadChildren() instead.
The root cause for this issue is that the _onDropDownMouseDown method of dijit._HasDropDown will manipulate the dom node, which cause the e.target of onmousedown and onmouseup changes for the first initialization.
As we know, the onclick event will be triggered only when the target of onmousedown and onmouseup are the same target.
So in this case, the onclick event is not triggered.
It seems that in Dojo 1.5, the problem still remains.