Eclipse Community Forums
Forum Search:

Search      Help    Register    Login    Home
Home » Modeling » Epsilon » Odd bug in generated graphical editor
Odd bug in generated graphical editor [message #12507] Thu, 12 February 2009 10:15 Go to next message
Michael Moser is currently offline Michael MoserFriend
Messages: 914
Registered: July 2009
Senior Member
This is a multi-part message in MIME format.

------=_NextPart_000_0011_01C98D03.323719E0
Content-Type: text/plain;
format=flowed;
charset="iso-8859-1";
reply-type=original
Content-Transfer-Encoding: 7bit

I generated a simple editor from the attached .emf file. Oddly, when
running the generated graphical editor, one can nicely draw nodes and
connections. But whenever one creates a new node, then suddenly all
connections drawn so far are removed again from the diagram.

Can others reproduce this? Any idea why? Can one fix this somehow? And
is this now an Epsilon/Eugenia bug or a genuine GMF bug? I presume the
latter, but just wanted to double check. Before reporting transferring
this to the GMF NG.

Michael

------=_NextPart_000_0011_01C98D03.323719E0
Content-Type: application/octet-stream;
name="leonardo2.emf"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: attachment;
filename="leonardo2.emf"

@namespace(uri=3D"leonardo2", prefix=3D"leonardo2")=0A=
=0A=
@gmf (foo=3D"bar")=0A=
package leonardo2;=0A=
=0A=
@gmf.diagram (foo=3D"bar")=0A=
class Leonardo2 {=0A=
attr String title =3D "Diagram";=0A=
attr String description;=0A=
val Artifact[*] artifacts;=0A=
val Process[*] processes;=0A=
val ArtifactRelation[*] artifactRelations;=0A=
val ProcessRelation[*] processRelations;=0A=
}=0A=
=0A=
@gmf.node(label =3D "name", label.icon=3D"true", figure=3D"rectangle", =
color=3D"255,255,200")=0A=
class Artifact {=0A=
id attr String name;=0A=
attr String description;=0A=
attr String location;=0A=
}=0A=
=0A=
@gmf.node(label =3D "name", label.icon=3D"true")=0A=
class ArtifactContainer extends Artifact {=0A=
val Artifact[*] contents;=0A=
}=0A=
=0A=
@gmf.node(label =3D "name", label.icon=3D"true", color=3D"200,255,200")=0A=
class Process {=0A=
id attr String name;=0A=
attr String description;=0A=
val ArtifactRelation[*]#process1 artifactRelations;=0A=
val ProcessRelation[*]#process2 processRelations;=0A=
}=0A=
=0A=
@gmf.link(target=3D"artifact", source=3D"process1", label=3D"name", =
target.decoration=3D"arrow", style=3D"dash", width=3D"2", =
incoming=3D"true")=0A=
class ArtifactRelation {=0A=
attr String name;=0A=
attr String description;=0A=
attr ArtifactRelationType relationType =3D "0";=0A=
ref Artifact artifact;=0A=
ref Process#artifactRelations process1;=0A=
}=0A=
=0A=
enum ArtifactRelationType {=0A=
CREATES =3D 0;=0A=
READS =3D 1;=0A=
UPDATES =3D 2;=0A=
DELETES =3D 3;=0A=
IS_INSTANTIATED_FROM =3D 4;=0A=
}=0A=
=0A=
@gmf.link(target=3D"target", source=3D"process2", label=3D"name", =
target.decoration=3D"arrow", style=3D"dot", width=3D"2", =
incoming=3D"true")=0A=
class ProcessRelation {=0A=
attr String name;=0A=
attr String description;=0A=
attr ProcessRelationType relationType =3D "0";=0A=
ref Process target;=0A=
ref Process#processRelations process2;=0A=
}=0A=
=0A=
enum ProcessRelationType {=0A=
INSTANTIATES =3D 0;=0A=
CALLS =3D 1;=0A=
SENDS_DATA =3D 2;=0A=
RECEIVES_DATA =3D 3;=0A=
TERMINATES =3D 4;=0A=
}=0A=
=0A=

------=_NextPart_000_0011_01C98D03.323719E0--
Re: Odd bug in generated graphical editor [message #12559 is a reply to message #12507] Thu, 12 February 2009 11:10 Go to previous messageGo to next message
Dimitrios Kolovos is currently offline Dimitrios KolovosFriend
Messages: 1776
Registered: July 2009
Senior Member
Hi Michael,

Apparently the artificial containment workaround I proposed was not a
good one. However, there is no need for it any more since I've fixed it
properly in Eugenia. I've uploaded the new version under
http://download.eclipse.org/modeling/gmt/epsilon/interim/ so once the
download server has picked up the changes you can update from there
(I'll send a message to the newsgroup when it is ready).

Once you have updated, you can remove the artificial containment
references from class Leonardo2:

val ArtifactRelation[*] artifactRelations;
val ProcessRelation[*] processRelations;

and everything will work as expected. Again, apologies for the previous
bad workaround!

Cheers,
Dimitris

Michael Moser wrote:
> I generated a simple editor from the attached .emf file. Oddly, when
> running the generated graphical editor, one can nicely draw nodes and
> connections. But whenever one creates a new node, then suddenly all
> connections drawn so far are removed again from the diagram.
>
> Can others reproduce this? Any idea why? Can one fix this somehow? And
> is this now an Epsilon/Eugenia bug or a genuine GMF bug? I presume the
> latter, but just wanted to double check. Before reporting transferring
> this to the GMF NG.
>
> Michael
Re: Odd bug in generated graphical editor [message #12580 is a reply to message #12559] Thu, 12 February 2009 13:03 Go to previous messageGo to next message
Dimitrios Kolovos is currently offline Dimitrios KolovosFriend
Messages: 1776
Registered: July 2009
Senior Member
On an update to this, the interim update site
(http://download.eclipse.org/modeling/gmt/epsilon/interim/) is now ready.

Cheers,
Dimitris

Dimitris Kolovos wrote:
> Hi Michael,
>
> Apparently the artificial containment workaround I proposed was not a
> good one. However, there is no need for it any more since I've fixed it
> properly in Eugenia. I've uploaded the new version under
> http://download.eclipse.org/modeling/gmt/epsilon/interim/ so once the
> download server has picked up the changes you can update from there
> (I'll send a message to the newsgroup when it is ready).
>
> Once you have updated, you can remove the artificial containment
> references from class Leonardo2:
>
> val ArtifactRelation[*] artifactRelations;
> val ProcessRelation[*] processRelations;
>
> and everything will work as expected. Again, apologies for the previous
> bad workaround!
>
> Cheers,
> Dimitris
>
> Michael Moser wrote:
>> I generated a simple editor from the attached .emf file. Oddly, when
>> running the generated graphical editor, one can nicely draw nodes and
>> connections. But whenever one creates a new node, then suddenly all
>> connections drawn so far are removed again from the diagram.
>>
>> Can others reproduce this? Any idea why? Can one fix this somehow? And
>> is this now an Epsilon/Eugenia bug or a genuine GMF bug? I presume the
>> latter, but just wanted to double check. Before reporting transferring
>> this to the GMF NG.
>>
>> Michael
Re: Odd bug in generated graphical editor [message #12599 is a reply to message #12580] Thu, 12 February 2009 13:27 Go to previous messageGo to next message
Dimitrios Kolovos is currently offline Dimitrios KolovosFriend
Messages: 1776
Registered: July 2009
Senior Member
.... and the bug report for tracking purposes:
https://bugs.eclipse.org/bugs/show_bug.cgi?id=264677

Dimitris Kolovos wrote:
> On an update to this, the interim update site
> (http://download.eclipse.org/modeling/gmt/epsilon/interim/) is now ready.
>
> Cheers,
> Dimitris
>
> Dimitris Kolovos wrote:
>> Hi Michael,
>>
>> Apparently the artificial containment workaround I proposed was not a
>> good one. However, there is no need for it any more since I've fixed
>> it properly in Eugenia. I've uploaded the new version under
>> http://download.eclipse.org/modeling/gmt/epsilon/interim/ so once the
>> download server has picked up the changes you can update from there
>> (I'll send a message to the newsgroup when it is ready).
>>
>> Once you have updated, you can remove the artificial containment
>> references from class Leonardo2:
>>
>> val ArtifactRelation[*] artifactRelations;
>> val ProcessRelation[*] processRelations;
>>
>> and everything will work as expected. Again, apologies for the
>> previous bad workaround!
>>
>> Cheers,
>> Dimitris
>>
>> Michael Moser wrote:
>>> I generated a simple editor from the attached .emf file. Oddly, when
>>> running the generated graphical editor, one can nicely draw nodes and
>>> connections. But whenever one creates a new node, then suddenly all
>>> connections drawn so far are removed again from the diagram.
>>>
>>> Can others reproduce this? Any idea why? Can one fix this somehow?
>>> And is this now an Epsilon/Eugenia bug or a genuine GMF bug? I
>>> presume the latter, but just wanted to double check. Before reporting
>>> transferring this to the GMF NG.
>>>
>>> Michael
Re: Odd bug in generated graphical editor [message #12641 is a reply to message #12599] Thu, 12 February 2009 16:32 Go to previous message
Michael Moser is currently offline Michael MoserFriend
Messages: 914
Registered: July 2009
Senior Member
Dimitris Kolovos wrote:
> ... and the bug report for tracking purposes:
> https://bugs.eclipse.org/bugs/show_bug.cgi?id=264677

Great! No more complaints! For now... ;-)

Michael
Re: Odd bug in generated graphical editor [message #565240 is a reply to message #12507] Thu, 12 February 2009 11:10 Go to previous message
Dimitrios Kolovos is currently offline Dimitrios KolovosFriend
Messages: 1776
Registered: July 2009
Senior Member
Hi Michael,

Apparently the artificial containment workaround I proposed was not a
good one. However, there is no need for it any more since I've fixed it
properly in Eugenia. I've uploaded the new version under
http://download.eclipse.org/modeling/gmt/epsilon/interim/ so once the
download server has picked up the changes you can update from there
(I'll send a message to the newsgroup when it is ready).

Once you have updated, you can remove the artificial containment
references from class Leonardo2:

val ArtifactRelation[*] artifactRelations;
val ProcessRelation[*] processRelations;

and everything will work as expected. Again, apologies for the previous
bad workaround!

Cheers,
Dimitris

Michael Moser wrote:
> I generated a simple editor from the attached .emf file. Oddly, when
> running the generated graphical editor, one can nicely draw nodes and
> connections. But whenever one creates a new node, then suddenly all
> connections drawn so far are removed again from the diagram.
>
> Can others reproduce this? Any idea why? Can one fix this somehow? And
> is this now an Epsilon/Eugenia bug or a genuine GMF bug? I presume the
> latter, but just wanted to double check. Before reporting transferring
> this to the GMF NG.
>
> Michael
Re: Odd bug in generated graphical editor [message #565265 is a reply to message #12559] Thu, 12 February 2009 13:03 Go to previous message
Dimitrios Kolovos is currently offline Dimitrios KolovosFriend
Messages: 1776
Registered: July 2009
Senior Member
On an update to this, the interim update site
(http://download.eclipse.org/modeling/gmt/epsilon/interim/) is now ready.

Cheers,
Dimitris

Dimitris Kolovos wrote:
> Hi Michael,
>
> Apparently the artificial containment workaround I proposed was not a
> good one. However, there is no need for it any more since I've fixed it
> properly in Eugenia. I've uploaded the new version under
> http://download.eclipse.org/modeling/gmt/epsilon/interim/ so once the
> download server has picked up the changes you can update from there
> (I'll send a message to the newsgroup when it is ready).
>
> Once you have updated, you can remove the artificial containment
> references from class Leonardo2:
>
> val ArtifactRelation[*] artifactRelations;
> val ProcessRelation[*] processRelations;
>
> and everything will work as expected. Again, apologies for the previous
> bad workaround!
>
> Cheers,
> Dimitris
>
> Michael Moser wrote:
>> I generated a simple editor from the attached .emf file. Oddly, when
>> running the generated graphical editor, one can nicely draw nodes and
>> connections. But whenever one creates a new node, then suddenly all
>> connections drawn so far are removed again from the diagram.
>>
>> Can others reproduce this? Any idea why? Can one fix this somehow? And
>> is this now an Epsilon/Eugenia bug or a genuine GMF bug? I presume the
>> latter, but just wanted to double check. Before reporting transferring
>> this to the GMF NG.
>>
>> Michael
Re: Odd bug in generated graphical editor [message #565291 is a reply to message #12580] Thu, 12 February 2009 13:27 Go to previous message
Dimitrios Kolovos is currently offline Dimitrios KolovosFriend
Messages: 1776
Registered: July 2009
Senior Member
.... and the bug report for tracking purposes:
https://bugs.eclipse.org/bugs/show_bug.cgi?id=264677

Dimitris Kolovos wrote:
> On an update to this, the interim update site
> (http://download.eclipse.org/modeling/gmt/epsilon/interim/) is now ready.
>
> Cheers,
> Dimitris
>
> Dimitris Kolovos wrote:
>> Hi Michael,
>>
>> Apparently the artificial containment workaround I proposed was not a
>> good one. However, there is no need for it any more since I've fixed
>> it properly in Eugenia. I've uploaded the new version under
>> http://download.eclipse.org/modeling/gmt/epsilon/interim/ so once the
>> download server has picked up the changes you can update from there
>> (I'll send a message to the newsgroup when it is ready).
>>
>> Once you have updated, you can remove the artificial containment
>> references from class Leonardo2:
>>
>> val ArtifactRelation[*] artifactRelations;
>> val ProcessRelation[*] processRelations;
>>
>> and everything will work as expected. Again, apologies for the
>> previous bad workaround!
>>
>> Cheers,
>> Dimitris
>>
>> Michael Moser wrote:
>>> I generated a simple editor from the attached .emf file. Oddly, when
>>> running the generated graphical editor, one can nicely draw nodes and
>>> connections. But whenever one creates a new node, then suddenly all
>>> connections drawn so far are removed again from the diagram.
>>>
>>> Can others reproduce this? Any idea why? Can one fix this somehow?
>>> And is this now an Epsilon/Eugenia bug or a genuine GMF bug? I
>>> presume the latter, but just wanted to double check. Before reporting
>>> transferring this to the GMF NG.
>>>
>>> Michael
Re: Odd bug in generated graphical editor [message #565331 is a reply to message #12599] Thu, 12 February 2009 16:32 Go to previous message
Michael Moser is currently offline Michael MoserFriend
Messages: 914
Registered: July 2009
Senior Member
Dimitris Kolovos wrote:
> ... and the bug report for tracking purposes:
> https://bugs.eclipse.org/bugs/show_bug.cgi?id=264677

Great! No more complaints! For now... ;-)

Michael
Previous Topic:up-to-date list of supported annotations?
Next Topic:// Comments cause remainder of EGL document to be ignored
Goto Forum:
  


Current Time: Sun Dec 21 13:42:31 GMT 2014

Powered by FUDForum. Page generated in 0.03936 seconds
.:: Contact :: Home ::.

Powered by: FUDforum 3.0.2.
Copyright ©2001-2010 FUDforum Bulletin Board Software