|
| Re: Proper usage of diagram node selection and subsequent command processing [message #64750 is a reply to message #64725] |
Tue, 17 October 2006 22:39  |
Eclipse User |
|
|
|
Originally posted by: vcciubot.uwaterloo.ca
> (1)
> Why get methods like e.g. PackageEditHelper.getMoveCommand(MoveRequest
> req) executed 3-5 times in a row when I do a drag'n'drop operation (with
> a single diagram element!) in my diagram?
> Not only I wonder why that method is called multiple times, but also the
> fact that the number of subsequent calls varies.
Tools/Actions etc call getCommand() on an EditPart to update the mouse
pointer or their icon (enabled/disabled). As such, when you drag the tool
updates the target edit part and calls getCommand() on that edit part
which in turn calls the semantic policy which calls your edit helper.
>
> Further, I do expect that method to be called not earlier then the
> "release mouse button event". Instead, it's called on mouse hover (with
> dragged elements glued to the mouse pointer).
Those commands are checked for being executable but are only executed when
the dragging ends.
>
> (2)
> I feel that there are some things to know concerning proper selection of
> multiple diagram nodes. In particular, how do I access the set of
> currently active diagram elements?
EditPartViewer#getSelection
You can get the viewer from an edit part.
>
> (3, as a consequence of 1+2)
> When will the runtime documentation and related docs be reviewed to
> cover these nasty things?
It's hard at the beginning :)
Some of these topics have already been covered in the GEF programmer's
manual.
|
|
|
Powered by
FUDForum. Page generated in 0.03656 seconds
] 
Search
Help
Register
Login
Home
