| Zoom / routing issue in GEF? [message #205570] |
Mon, 19 December 2005 17:58  |
Eclipse User |
|
|
|
Originally posted by: cricard.businessobjects.com
In ShortestPathRouter, algorithm.solve() returns a strange List of paths
when the zoom factor is not 1.0 (attached snapshot there:
http://img480.imageshack.us/img480/3008/zoombug6lg.png. Look at the dotted
dummy connection that's being routed while the mouse moves. Two bendpoints
are computed that are really far far away from the cursor's actual
location).
I wondered if this might be a GEF bug, or if some special code has to be
written to avoid this.
Thanks for any hint,
Christophe.
|
|
|
| Re: Zoom / routing issue in GEF? [message #205591 is a reply to message #205570] |
Mon, 19 December 2005 21:54  |
| Eclipse User |
|
|
|
Originally posted by: none.us.ibm.com
It's working in the logic example. See what it does different
"cricri" <cricard@businessobjects.com> wrote in message
news:do6sbo$h28$1@utils.eclipse.org...
> In ShortestPathRouter, algorithm.solve() returns a strange List of paths
> when the zoom factor is not 1.0 (attached snapshot there:
> http://img480.imageshack.us/img480/3008/zoombug6lg.png. Look at the dotted
> dummy connection that's being routed while the mouse moves. Two bendpoints
> are computed that are really far far away from the cursor's actual
> location).
>
> I wondered if this might be a GEF bug, or if some special code has to be
> written to avoid this.
>
> Thanks for any hint,
>
> Christophe.
>
>
|
|
|
| Re: Zoom / routing issue in GEF? [message #205768 is a reply to message #205591] |
Wed, 21 December 2005 17:39 |
| Eclipse User |
|
|
|
Originally posted by: cricard.businessobjects.com
Oooh my, you're right, as usual...
The problem is, it does a lot of things differently ;)
Thanks for taking the time to have a peek at this issue, anyway.
"Randy Hudson" <none@us.ibm.com> wrote in message
news:do7a60$1he$1@utils.eclipse.org...
> It's working in the logic example. See what it does different
>
> "cricri" <cricard@businessobjects.com> wrote in message
> news:do6sbo$h28$1@utils.eclipse.org...
> > In ShortestPathRouter, algorithm.solve() returns a strange List of paths
> > when the zoom factor is not 1.0 (attached snapshot there:
> > http://img480.imageshack.us/img480/3008/zoombug6lg.png. Look at the
dotted
> > dummy connection that's being routed while the mouse moves. Two
bendpoints
> > are computed that are really far far away from the cursor's actual
> > location).
> >
> > I wondered if this might be a GEF bug, or if some special code has to be
> > written to avoid this.
> >
> > Thanks for any hint,
> >
> > Christophe.
> >
> >
>
>
|
|
|
| Re: Zoom / routing issue in GEF? [message #223531 is a reply to message #205768] |
Thu, 28 September 2006 15:22 |
| Eclipse User |
|
|
|
Originally posted by: cricard.businessobjects.com
Well, it took me some time to be in the proper mood, but yesterday evening I
felt I could find out what I was doing wrong, and eventually fix that darn
bug.
After overriding ShortestPathRouter (in order to scale obstacles) to no
avail, I thought to myself: "How come dummy connections are badly routed
when the scaling is not 1.0, while "final" connections are okay?"
Sounded like a layer issue...
And, guess what?
The "final" connections are usually added to the
LayerConstants.CONNECTION_LAYER layer, which is scaled.
But the dummy connections are drawn using GraphicalNodeEditPolicy's
getFeedbackLayer(), which returns LayerConstants.FEEDBACK_LAYER (not
scaled).
So, in my implementation of GraphicalNodeEditPolicy, I now return
LayerConstants.SCALED_FEEDBACK_LAYER instead, and everything works
perfectly!
"cricri" <cricard@businessobjects.com> wrote in message
news:doc40g$um9$1@utils.eclipse.org...
> Oooh my, you're right, as usual...
> The problem is, it does a lot of things differently ;)
> Thanks for taking the time to have a peek at this issue, anyway.
>
> "Randy Hudson" <none@us.ibm.com> wrote in message
> news:do7a60$1he$1@utils.eclipse.org...
>> It's working in the logic example. See what it does different
>>
>> "cricri" <cricard@businessobjects.com> wrote in message
>> news:do6sbo$h28$1@utils.eclipse.org...
>> > In ShortestPathRouter, algorithm.solve() returns a strange List of
>> > paths
>> > when the zoom factor is not 1.0 (attached snapshot there:
>> > http://img480.imageshack.us/img480/3008/zoombug6lg.png. Look at the
> dotted
>> > dummy connection that's being routed while the mouse moves. Two
> bendpoints
>> > are computed that are really far far away from the cursor's actual
>> > location).
>> >
>> > I wondered if this might be a GEF bug, or if some special code has to
>> > be
>> > written to avoid this.
>> >
>> > Thanks for any hint,
>> >
>> > Christophe.
>> >
>> >
>>
>>
>
>
|
|
|
Powered by
FUDForum. Page generated in 0.03339 seconds
] 
Search
Help
Register
Login
Home
