Home » Modeling » GMF (Graphical Modeling Framework) » Rotation of EditParts possible?
|
| Re: Rotation of EditParts possible? [message #82793 is a reply to message #82215] |
Mon, 04 December 2006 22:06   |
 Cherie Revells
Messages: 299
Registered: July 2009 |
Senior Member |
|
|
Sven,
I do not know of any working examples, but I have seen it implemented.
I believe the code you are looking at in GMF will provide a client with
the tool and feedback required to rotate a shape, but the drawing of
the shape would be left for the implementor of the editpart/figure.
If your editpart returns the RotatableShapeEditPolicy in the
getPrimaryDragEditPolicy() method and implements IRotatableEditPart,
then this should (I haven't tried it) provide you with the handles on
the corner of your shape to rotate the shape.
The RotatableShapeEditPolicy will set the resizeDirection on the
ChangeBoundsRequest. Now, all you have to do, is respond to this
request, by returning a command that will not only change the bounds of
your rotatable shape, but also somehow sets its direction. :-)
I would be interested in knowing if this works out for you.
- Cherie
Sven Wende wrote:
> Hi,
>
> I just have discovered
>
> RotatableShapeEditPolicy
> ( http://help.eclipse.org/help32/index.jsp?topic=/org.eclipse. gmf.doc/reference/api/runtime/org/eclipse/gmf/runtime/diagra m/ui/editpolicies/RotatableShapeEditPolicy.html)
>
> and
>
> IRotatableEditPart
> ( http://help.eclipse.org/help32/index.jsp?topic=/org.eclipse. gmf.doc/reference/api/runtime/org/eclipse/gmf/runtime/diagra m/ui/editparts/IRotatableEditPart.html)
>
> in the documentation.
>
> I wonder, if these can really be used to implement rotatable EditParts. Are
> there any working examples?
>
> I already tried to apply them to a simple GEF editor, but without success.
>
> Kind regards
>
> Sven
>
>
|
|
|
| Re: Rotation of EditParts possible? [message #82848 is a reply to message #82793] |
Mon, 04 December 2006 23:43 |
Hallvard Traetteberg
Messages: 673
Registered: July 2009
Location: Trondheim, Norway |
Senior Member |
|
|
Hi,
I need the ability to rotate an edit part, too! In my case, I want to
change a custom style attribute to the direction the edit part's
figure should point (the figure is a triangle).
Based on your information, I tried to track what was happening during
a rotate. As far as I can tell, a rotate uses a RotateShapeRequest,
which for most practical purposes is a change bounds request, e.g. it
results in a commans that changes the bounds. A RotateShapeRequest
does provide an int telling which handle is used, e.g. NORTH_WEST, but
I could not find any information telling whether the rotate was
clockwise or counter-clockwise! For the rectangle this doesn't matter,
a rotate only swaps the width/height, but for me it does matter, as I
need to tell whether the triange figure should point up or down, left
or right.
So, how can I find the rotation direction (as opposed to the direction
of the handle used for rotating)? Do I need to make my own tracker (to
get the new direction), policy (to create the appropriate tracker) and
request (to record the final direction)?
Hallvard
On Mon, 04 Dec 2006 17:06:27 -0500, Cherie Revells
<crevells@ca.ibm.com> wrote:
>Sven,
>
>I do not know of any working examples, but I have seen it implemented.
>
>I believe the code you are looking at in GMF will provide a client with
> the tool and feedback required to rotate a shape, but the drawing of
>the shape would be left for the implementor of the editpart/figure.
>
>If your editpart returns the RotatableShapeEditPolicy in the
>getPrimaryDragEditPolicy() method and implements IRotatableEditPart,
>then this should (I haven't tried it) provide you with the handles on
>the corner of your shape to rotate the shape,
>
>The RotatableShapeEditPolicy will set the resizeDirection on the
>ChangeBoundsRequest. Now, all you have to do, is respond to this
>request, by returning a command that will not only change the bounds of
>your rotatable shape, but also somehow sets its direction. :-)
>
>I would be interested in knowing if this works out for you.
>
>- Cherie
>
>Sven Wende wrote:
>> Hi,
>>
>> I just have discovered
>>
>> RotatableShapeEditPolicy
>> ( http://help.eclipse.org/help32/index.jsp?topic=/org.eclipse. gmf.doc/reference/api/runtime/org/eclipse/gmf/runtime/diagra m/ui/editpolicies/RotatableShapeEditPolicy.html)
>>
>> and
>>
>> IRotatableEditPart
>> ( http://help.eclipse.org/help32/index.jsp?topic=/org.eclipse. gmf.doc/reference/api/runtime/org/eclipse/gmf/runtime/diagra m/ui/editparts/IRotatableEditPart.html)
>>
>> in the documentation.
>>
>> I wonder, if these can really be used to implement rotatable EditParts. Are
>> there any working examples?
>>
>> I already tried to apply them to a simple GEF editor, but without success.
>>
>> Kind regards
>>
>> Sven
>>
>>
|
|
|
| Re: Rotation of EditParts possible? [message #84238 is a reply to message #82848] |
Thu, 07 December 2006 21:03 |
Cherie Revells
Messages: 299
Registered: July 2009 |
Senior Member |
|
|
Hallvard,
Yes, the current infrastructure in GMF seems to be quite restricted.
You are right on track with what needs to be done. I think this
behavior could be beneficial in GMF so you could even make the changes
with the existing RotateShape classes and submit it as a patch. :-)
Regards,
Cherie
Hallvard Trætteberg wrote:
> Hi,
>
> I need the ability to rotate an edit part, too! In my case, I want to
> change a custom style attribute to the direction the edit part's
> figure should point (the figure is a triangle).
>
> Based on your information, I tried to track what was happening during
> a rotate. As far as I can tell, a rotate uses a RotateShapeRequest,
> which for most practical purposes is a change bounds request, e.g. it
> results in a commans that changes the bounds. A RotateShapeRequest
> does provide an int telling which handle is used, e.g. NORTH_WEST, but
> I could not find any information telling whether the rotate was
> clockwise or counter-clockwise! For the rectangle this doesn't matter,
> a rotate only swaps the width/height, but for me it does matter, as I
> need to tell whether the triange figure should point up or down, left
> or right.
>
> So, how can I find the rotation direction (as opposed to the direction
> of the handle used for rotating)? Do I need to make my own tracker (to
> get the new direction), policy (to create the appropriate tracker) and
> request (to record the final direction)?
>
> Hallvard
>
> On Mon, 04 Dec 2006 17:06:27 -0500, Cherie Revells
> <crevells@ca.ibm.com> wrote:
>
>> Sven,
>>
>> I do not know of any working examples, but I have seen it implemented.
>>
>> I believe the code you are looking at in GMF will provide a client with
>> the tool and feedback required to rotate a shape, but the drawing of
>> the shape would be left for the implementor of the editpart/figure.
>>
>> If your editpart returns the RotatableShapeEditPolicy in the
>> getPrimaryDragEditPolicy() method and implements IRotatableEditPart,
>> then this should (I haven't tried it) provide you with the handles on
>> the corner of your shape to rotate the shape,
>>
>> The RotatableShapeEditPolicy will set the resizeDirection on the
>> ChangeBoundsRequest. Now, all you have to do, is respond to this
>> request, by returning a command that will not only change the bounds of
>> your rotatable shape, but also somehow sets its direction. :-)
>>
>> I would be interested in knowing if this works out for you.
>>
>> - Cherie
>>
>> Sven Wende wrote:
>>> Hi,
>>>
>>> I just have discovered
>>>
>>> RotatableShapeEditPolicy
>>> ( http://help.eclipse.org/help32/index.jsp?topic=/org.eclipse. gmf.doc/reference/api/runtime/org/eclipse/gmf/runtime/diagra m/ui/editpolicies/RotatableShapeEditPolicy.html)
>>>
>>> and
>>>
>>> IRotatableEditPart
>>> ( http://help.eclipse.org/help32/index.jsp?topic=/org.eclipse. gmf.doc/reference/api/runtime/org/eclipse/gmf/runtime/diagra m/ui/editparts/IRotatableEditPart.html)
>>>
>>> in the documentation.
>>>
>>> I wonder, if these can really be used to implement rotatable EditParts. Are
>>> there any working examples?
>>>
>>> I already tried to apply them to a simple GEF editor, but without success.
>>>
>>> Kind regards
>>>
>>> Sven
>>>
>>>
>
|
|
|
| Re: Rotation of EditParts possible? [message #84360 is a reply to message #84238] |
Fri, 08 December 2006 14:31 |
Hallvard Traetteberg
Messages: 673
Registered: July 2009
Location: Trondheim, Norway |
Senior Member |
|
|
Cherie,
On Thu, 07 Dec 2006 16:03:42 -0500, Cherie Revells
<crevells@ca.ibm.com> wrote:
>Hallvard,
>
>Yes, the current infrastructure in GMF seems to be quite restricted.
>You are right on track with what needs to be done. I think this
>behavior could be beneficial in GMF so you could even make the changes
>with the existing RotateShape classes and submit it as a patch. :-)
I've now implemented new RotatableShapeEditPolicy and
RotateShapeRequest classes, that support rotating through the 8
directions defined in PositionConstants. The feedback is as before,
but the RotateShapeRequest will hold both the direction of the handle
(initial direction) and the final direction, so it's easy to define
another edit policy that generates a command defined in terms of
either absolute or relative direction. E.g. I have implemented my own
ChangeDirectionEditPolicy that based on the original direction of a
rotateable object and the difference between initial and final
direction, returns a command for setting a new direction. I have,
however, not been able to make the ChangeDirectionEditPolicy turn off
the setting of the bounds by the SetBoundsCommand returned by the
XYLayoutEditPolicy. Hence, both are executed, which is not ideal.
Due to class loading issues, I was not able to write drop-in
replacements for the existing classes and make a patch. I had to
change their names to RotatableShapeEditPolicyEx and
RotateShapeRequestEx. What is the best way to get them into GMF?
Register a request for enhancement and attach the files?
A side comment: It's fairly confusing and difficult to debug code that
involve tools, edit parts, edit policies and commands. Since commands
are build both during dragging and mouse button release it's
cumbersome to use the debugger. There really should be a way of
logging/tracing how a command is built, i.e. which edit parts and edit
policies participated in the creation of a specific command.
Hallvard
>
>Regards,
>Cherie
>
>Hallvard Trætteberg wrote:
>> Hi,
>>
>> I need the ability to rotate an edit part, too! In my case, I want to
>> change a custom style attribute to the direction the edit part's
>> figure should point (the figure is a triangle).
>>
>> Based on your information, I tried to track what was happening during
>> a rotate. As far as I can tell, a rotate uses a RotateShapeRequest,
>> which for most practical purposes is a change bounds request, e.g. it
>> results in a commans that changes the bounds. A RotateShapeRequest
>> does provide an int telling which handle is used, e.g. NORTH_WEST, but
>> I could not find any information telling whether the rotate was
>> clockwise or counter-clockwise! For the rectangle this doesn't matter,
>> a rotate only swaps the width/height, but for me it does matter, as I
>> need to tell whether the triange figure should point up or down, left
>> or right.
>>
>> So, how can I find the rotation direction (as opposed to the direction
>> of the handle used for rotating)? Do I need to make my own tracker (to
>> get the new direction), policy (to create the appropriate tracker) and
>> request (to record the final direction)?
>>
>> Hallvard
>>
>> On Mon, 04 Dec 2006 17:06:27 -0500, Cherie Revells
>> <crevells@ca.ibm.com> wrote:
>>
>>> Sven,
>>>
>>> I do not know of any working examples, but I have seen it implemented.
>>>
>>> I believe the code you are looking at in GMF will provide a client with
>>> the tool and feedback required to rotate a shape, but the drawing of
>>> the shape would be left for the implementor of the editpart/figure.
>>>
>>> If your editpart returns the RotatableShapeEditPolicy in the
>>> getPrimaryDragEditPolicy() method and implements IRotatableEditPart,
>>> then this should (I haven't tried it) provide you with the handles on
>>> the corner of your shape to rotate the shape,
>>>
>>> The RotatableShapeEditPolicy will set the resizeDirection on the
>>> ChangeBoundsRequest. Now, all you have to do, is respond to this
>>> request, by returning a command that will not only change the bounds of
>>> your rotatable shape, but also somehow sets its direction. :-)
>>>
>>> I would be interested in knowing if this works out for you.
>>>
>>> - Cherie
>>>
>>> Sven Wende wrote:
>>>> Hi,
>>>>
>>>> I just have discovered
>>>>
>>>> RotatableShapeEditPolicy
>>>> ( http://help.eclipse.org/help32/index.jsp?topic=/org.eclipse. gmf.doc/reference/api/runtime/org/eclipse/gmf/runtime/diagra m/ui/editpolicies/RotatableShapeEditPolicy.html)
>>>>
>>>> and
>>>>
>>>> IRotatableEditPart
>>>> ( http://help.eclipse.org/help32/index.jsp?topic=/org.eclipse. gmf.doc/reference/api/runtime/org/eclipse/gmf/runtime/diagra m/ui/editparts/IRotatableEditPart.html)
>>>>
>>>> in the documentation.
>>>>
>>>> I wonder, if these can really be used to implement rotatable EditParts. Are
>>>> there any working examples?
>>>>
>>>> I already tried to apply them to a simple GEF editor, but without success.
>>>>
>>>> Kind regards
>>>>
>>>> Sven
>>>>
>>>>
>>
|
|
|
| Re: Rotation of EditParts possible? [message #84435 is a reply to message #84360] |
Fri, 08 December 2006 17:57 |
Cherie Revells
Messages: 299
Registered: July 2009 |
Senior Member |
|
|
Hallvard,
That's great! Yes, you should create an enhancement request and attach
your new files.
Thanks,
Cherie
Hallvard Trætteberg wrote:
> Cherie,
>
> On Thu, 07 Dec 2006 16:03:42 -0500, Cherie Revells
> <crevells@ca.ibm.com> wrote:
>
>> Hallvard,
>>
>> Yes, the current infrastructure in GMF seems to be quite restricted.
>> You are right on track with what needs to be done. I think this
>> behavior could be beneficial in GMF so you could even make the changes
>> with the existing RotateShape classes and submit it as a patch. :-)
>
> I've now implemented new RotatableShapeEditPolicy and
> RotateShapeRequest classes, that support rotating through the 8
> directions defined in PositionConstants. The feedback is as before,
> but the RotateShapeRequest will hold both the direction of the handle
> (initial direction) and the final direction, so it's easy to define
> another edit policy that generates a command defined in terms of
> either absolute or relative direction. E.g. I have implemented my own
> ChangeDirectionEditPolicy that based on the original direction of a
> rotateable object and the difference between initial and final
> direction, returns a command for setting a new direction. I have,
> however, not been able to make the ChangeDirectionEditPolicy turn off
> the setting of the bounds by the SetBoundsCommand returned by the
> XYLayoutEditPolicy. Hence, both are executed, which is not ideal.
>
> Due to class loading issues, I was not able to write drop-in
> replacements for the existing classes and make a patch. I had to
> change their names to RotatableShapeEditPolicyEx and
> RotateShapeRequestEx. What is the best way to get them into GMF?
> Register a request for enhancement and attach the files?
>
> A side comment: It's fairly confusing and difficult to debug code that
> involve tools, edit parts, edit policies and commands. Since commands
> are build both during dragging and mouse button release it's
> cumbersome to use the debugger. There really should be a way of
> logging/tracing how a command is built, i.e. which edit parts and edit
> policies participated in the creation of a specific command.
>
> Hallvard
>
>> Regards,
>> Cherie
>>
>> Hallvard Trætteberg wrote:
>>> Hi,
>>>
>>> I need the ability to rotate an edit part, too! In my case, I want to
>>> change a custom style attribute to the direction the edit part's
>>> figure should point (the figure is a triangle).
>>>
>>> Based on your information, I tried to track what was happening during
>>> a rotate. As far as I can tell, a rotate uses a RotateShapeRequest,
>>> which for most practical purposes is a change bounds request, e.g. it
>>> results in a commans that changes the bounds. A RotateShapeRequest
>>> does provide an int telling which handle is used, e.g. NORTH_WEST, but
>>> I could not find any information telling whether the rotate was
>>> clockwise or counter-clockwise! For the rectangle this doesn't matter,
>>> a rotate only swaps the width/height, but for me it does matter, as I
>>> need to tell whether the triange figure should point up or down, left
>>> or right.
>>>
>>> So, how can I find the rotation direction (as opposed to the direction
>>> of the handle used for rotating)? Do I need to make my own tracker (to
>>> get the new direction), policy (to create the appropriate tracker) and
>>> request (to record the final direction)?
>>>
>>> Hallvard
>>>
>>> On Mon, 04 Dec 2006 17:06:27 -0500, Cherie Revells
>>> <crevells@ca.ibm.com> wrote:
>>>
>>>> Sven,
>>>>
>>>> I do not know of any working examples, but I have seen it implemented.
>>>>
>>>> I believe the code you are looking at in GMF will provide a client with
>>>> the tool and feedback required to rotate a shape, but the drawing of
>>>> the shape would be left for the implementor of the editpart/figure.
>>>>
>>>> If your editpart returns the RotatableShapeEditPolicy in the
>>>> getPrimaryDragEditPolicy() method and implements IRotatableEditPart,
>>>> then this should (I haven't tried it) provide you with the handles on
>>>> the corner of your shape to rotate the shape,
>>>>
>>>> The RotatableShapeEditPolicy will set the resizeDirection on the
>>>> ChangeBoundsRequest. Now, all you have to do, is respond to this
>>>> request, by returning a command that will not only change the bounds of
>>>> your rotatable shape, but also somehow sets its direction. :-)
>>>>
>>>> I would be interested in knowing if this works out for you.
>>>>
>>>> - Cherie
>>>>
>>>> Sven Wende wrote:
>>>>> Hi,
>>>>>
>>>>> I just have discovered
>>>>>
>>>>> RotatableShapeEditPolicy
>>>>> ( http://help.eclipse.org/help32/index.jsp?topic=/org.eclipse. gmf.doc/reference/api/runtime/org/eclipse/gmf/runtime/diagra m/ui/editpolicies/RotatableShapeEditPolicy.html)
>>>>>
>>>>> and
>>>>>
>>>>> IRotatableEditPart
>>>>> ( http://help.eclipse.org/help32/index.jsp?topic=/org.eclipse. gmf.doc/reference/api/runtime/org/eclipse/gmf/runtime/diagra m/ui/editparts/IRotatableEditPart.html)
>>>>>
>>>>> in the documentation.
>>>>>
>>>>> I wonder, if these can really be used to implement rotatable EditParts. Are
>>>>> there any working examples?
>>>>>
>>>>> I already tried to apply them to a simple GEF editor, but without success.
>>>>>
>>>>> Kind regards
>>>>>
>>>>> Sven
>>>>>
>>>>>
>
|
|
| | | |
Goto Forum:
Current Time: Wed Jan 15 11:42:59 GMT 2025
Powered by FUDForum. Page generated in 0.02632 seconds |
] 
Search
Help
Register
Login
Home
