Skip to main content


Eclipse Community Forums
Forum Search:

Search      Help    Register    Login    Home
Home » Modeling » UML2 » Re: Problem using UML Genmodel
Re: Problem using UML Genmodel [message #468510] Wed, 17 January 2007 15:53 Go to next message
Andreas Morf is currently offline Andreas MorfFriend
Messages: 14
Registered: July 2009
Junior Member
This is a multi-part message in MIME format.

------=_NextPart_000_000E_01C73A58.07A75B90
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Hi James
Thanks for your answer!
Fortunately the problem disappeared unexpectedly when I changed the =
Aggregation attribute of the memberEnd to "none". After switching back =
to 'composite' everything remained ok...

But after some experiences with the Genmodel there are some further =
questions:

- For every implementation class there is a getCacheAdapter() method =
generated (in the uml2.uml-impl-classes there is no such method) - this =
leads to a stack overflow using the generated editor plugin. Deleting =
the method manually resolves the problem. What am I doing wrong with the =
UML-Model/Genmodel?
- In the UML.merged.uml-model used as metamodel, what is the difference =
beween memberEnd Property with <eReference> Stereotype and the ones =
without Stereotype?
- Using an UML-model as metamodel is there a special behaviour in =
Genmodel when using ownedEnd's in associations (as in =
A_general_generalization)?

And finally a question which is maybe not just related with uml2:
Getting all the sources from CVS (emf: 200701110200 uml2: 200701111331) =
on 3.3M4 there are several errors in the UML2 projects even after =
updating the classpathes (org.eclipse.uml2, .uml2.importer.rose, . Is =
that just a strange behaviour of my environment?

Greets, Andreas


"James Bruck" <jbruck@ca.ibm.com> wrote in message =
news:eo93md$g96$2@utils.eclipse.org...
Hi Andreas,

I just tried this on my machine with a contrived example and it seemed =
to work fine ( If I understood this correctly ;) ).
Would it be possible for you to send more information or a snippet of =
your metamodel so I can inspect it.
What errors are you seeing in the impl class?

Regards,

- James.

"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eo7rou$ob0$1@utils.eclipse.org...
Hi all
I'm trying to make a new language metamodel on the basis on UML.uml =
and using UML-Genmodel. In principle, the process is working but with =
the following aspect I get into trouble:

- In my metamodel I have two classes with an 1 -- 0..* association
- the creation of the genmodel works well
- after model-code generation, there are errors in the impl-class of =
the 0..* -side: in the following code-snipped the varable ownedrules is =
not declared in the class
- if I change the Multiplicity to 1, the resulting code works
public EList getOwnedrules() {

if (ownedrules =3D=3D null) {

ownedrules =3D new EObjectWithInverseResolvingEList(

MappingRule.class, this,

UMLTPackage.TRANSFORMATION_GROUP__OWNEDRULES,

UMLTPackage.MAPPING_RULE__TGROUP);

}

return ownedrules;

}


In the attachments there are some pictures of the .uml and .genmodel =
to get the context.

Anyone an idea
Regards, Andreas

------=_NextPart_000_000E_01C73A58.07A75B90
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 6.00.5730.11" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>Hi James</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Thanks for your answer!</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Fortunately the problem disappeared =
unexpectedly=20
when I changed the Aggregation attribute of the memberEnd to "none". =
After=20
switching back to 'composite' everything remained ok...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>But after some experiences with the =
Genmodel there=20
are some further questions:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- For every implementation class there =
is a <FONT=20
size=3D2>getCacheAdapter() method generated (in the =
uml2.uml-impl-classes there is=20
no such method) - this leads to a stack overflow using the generated =
editor=20
plugin. Deleting the method manually resolves the problem. What am I =
doing wrong=20
with the UML-Model/Genmodel?</FONT></FONT></DIV>
<DIV><FONT face=3DArial size=3D2>- In the UML.merged.uml-model used as =
metamodel,=20
what is the difference beween memberEnd Property with &lt;eReference&gt; =

Stereotype and the ones without Stereotype?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>- Using an UML-model as metamodel is =
there a=20
special behaviour in Genmodel when using ownedEnd's in associations (as =
in=20
A_general_generalization)?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>And finally a question which is maybe =
not just=20
related with uml2:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Getting all the sources from CVS (emf: =
200701110200=20
uml2: 200701111331) on 3.3M4 there are several errors in the UML2 =
projects even=20
after updating the classpathes (org.eclipse.uml2, .uml2.importer.rose, . =
Is that=20
just a strange behaviour of my environment?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Greets, Andreas</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"James Bruck" &lt;<A=20
href=3D"mailto:jbruck@ca.ibm.com">jbruck@ca.ibm.com</A>&gt; wrote in =
message <A=20
=
href=3D"news:eo93md$g96$2@utils.eclipse.org">news:eo93md$g96$2@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>I just tried this on my machine with =
a contrived=20
example and it seemed to work fine ( If I understood this correctly =
;)&nbsp;=20
).</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Would it be possible for you to send =
more=20
information or a snippet of your metamodel so I can inspect =
it.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>What errors are you seeing in the =
impl=20
class?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Regards,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- James.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt; =
wrote=20
in message <A=20
=
href=3D"news:eo7rou$ob0$1@utils.eclipse.org">news:eo7rou$ob0$1@utils.ecli=
pse.org</A>...</DIV>
<DIV>Hi all</DIV>
<DIV>I'm trying to make a new language metamodel on the basis on =
UML.uml and=20
using UML-Genmodel. In principle, the process is working but with =
the=20
following aspect I get into trouble:</DIV>
<DIV>&nbsp;</DIV>
<DIV>- In my metamodel I have two classes with an 1 -- 0..*=20
association</DIV>
<DIV>- the creation of the genmodel works well</DIV>
<DIV>- after model-code generation, there are errors in the =
impl-class of=20
the 0..* -side: in the following code-snipped the varable ownedrules =
is not=20
declared in the class</DIV>
<DIV>- if I change the Multiplicity to 1, the resulting code =
works</DIV>
<DIV><FONT face=3DArial size=3D2>
<P align=3Dleft><B><FONT color=3D#7f0055 =
size=3D2>public</B></FONT><FONT size=3D2>=20
EList getOwnedrules() {</P>
<P align=3Dleft></FONT><B><FONT color=3D#7f0055 =
size=3D2>if</B></FONT><FONT=20
size=3D2> (ownedrules =3D=3D </FONT><B><FONT color=3D#7f0055=20
size=3D2>null</B></FONT><FONT size=3D2>) {</P>
<P align=3Dleft>ownedrules =3D </FONT><B><FONT color=3D#7f0055=20
size=3D2>new</B></FONT><FONT size=3D2> =
EObjectWithInverseResolvingEList(</P>
<P align=3Dleft>MappingRule.</FONT><B><FONT color=3D#7f0055=20
size=3D2>class</B></FONT><FONT size=3D2>, </FONT><B><FONT =
color=3D#7f0055=20
size=3D2>this</B></FONT><FONT size=3D2>,</P>
<P align=3Dleft>UMLTPackage.</FONT><I><FONT color=3D#0000c0=20
size=3D2>TRANSFORMATION_GROUP__OWNEDRULES</I></FONT><FONT =
size=3D2>,</P>
<P align=3Dleft>UMLTPackage.</FONT><I><FONT color=3D#0000c0=20
size=3D2>MAPPING_RULE__TGROUP</I></FONT><FONT size=3D2>);</P>
<P align=3Dleft>}</P>
<P align=3Dleft></FONT><B><FONT color=3D#7f0055 =
size=3D2>return</B></FONT><FONT=20
size=3D2> ownedrules;</P>
<P align=3Dleft>}</FONT></FONT></P></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV>In the attachments there are some pictures of the .uml and =
..genmodel to=20
get the context.</DIV>
<DIV>&nbsp;</DIV>
<DIV>Anyone an idea</DIV>
<DIV>Regards, Andreas</DIV>
<DIV><FONT face=3DArial=20
size=3D2></FONT>&nbsp;</DIV></BLOCKQUOTE></BLOCKQUOTE ></BODY></HTML>

------=_NextPart_000_000E_01C73A58.07A75B90--
Re: Problem using UML Genmodel [message #468524 is a reply to message #468510] Fri, 19 January 2007 02:34 Go to previous messageGo to next message
james bruck is currently offline james bruckFriend
Messages: 1724
Registered: July 2009
Senior Member
This is a multi-part message in MIME format.

------=_NextPart_000_0080_01C73B48.610BDD00
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Hi Andreas,

some comments below ...

- James.
"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eolgq1$q5s$1@utils.eclipse.org...
Hi James
Thanks for your answer!
Fortunately the problem disappeared unexpectedly when I changed the =
Aggregation attribute of the memberEnd to "none". After switching back =
to 'composite' everything remained ok...

But after some experiences with the Genmodel there are some further =
questions:

- For every implementation class there is a getCacheAdapter() method =
generated (in the uml2.uml-impl-classes there is no such method) - this =
leads to a stack overflow using the generated editor plugin. Deleting =
the method manually resolves the problem. What am I doing wrong with the =
UML-Model/Genmodel?

+ Not sure about this one I'll have to investigate and get back to =
you. =20

- In the UML.merged.uml-model used as metamodel, what is the =
difference beween memberEnd Property with <eReference> Stereotype and =
the ones without Stereotype?

+ In general the Ecore stereotype provides information when converting =
from UMLto Ecore. Specific code generation details might be necessary =
for those items with stereotypes applied. Sometimes information is =
provided in the stereotype to allow conversion from Ecore to UML and =
back to Ecore and preserve the integrity of the original .ecore model ( =
ie. round tripping ).

- Using an UML-model as metamodel is there a special behaviour in =
Genmodel when using ownedEnd's in associations (as in =
A_general_generalization)?

+ I'm not sure what you mean about "special behavior". I'm not aware =
of unique behavior for this. I'll discuss it with Kenn and get back to =
you if there is something unique here.

And finally a question which is maybe not just related with uml2:
Getting all the sources from CVS (emf: 200701110200 uml2: =
200701111331) on 3.3M4 there are several errors in the UML2 projects =
even after updating the classpathes (org.eclipse.uml2, =
..uml2.importer.rose, . Is that just a strange behaviour of my =
environment?

+ I'm not seeing any build issues for the uml2 project. I do however =
have some issues with org.eclipse.emf.importer.rose. .. may not be the =
same issue you are having but these problems are a result of .jars =
needed for ant.
Sometimes there are .readme's that explain how to resolve these =
issues.

eg. Before running your build scripts or the examples provided here, =
make sure you have the 2 jar files required by this plugin =
("importer.rose.jar" and "importer.rose.tasks.jar") and also the 2 jar =
files required by the org.eclipse.emf.ant plugin ("emf.ant.jar" and =
"emf.ant.tasks.jar"). If necessary, Eclipse can generate them for you. =
All you need to do is to run the Ant script created by PDE when you =
right click on the manifest file and select "PDE Tools>Create Ant Build =
File".=20

but again, I don't see any issues with uml.


Greets, Andreas


"James Bruck" <jbruck@ca.ibm.com> wrote in message =
news:eo93md$g96$2@utils.eclipse.org...
Hi Andreas,

I just tried this on my machine with a contrived example and it =
seemed to work fine ( If I understood this correctly ;) ).
Would it be possible for you to send more information or a snippet =
of your metamodel so I can inspect it.
What errors are you seeing in the impl class?

Regards,

- James.

"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eo7rou$ob0$1@utils.eclipse.org...
Hi all
I'm trying to make a new language metamodel on the basis on =
UML.uml and using UML-Genmodel. In principle, the process is working but =
with the following aspect I get into trouble:

- In my metamodel I have two classes with an 1 -- 0..* association
- the creation of the genmodel works well
- after model-code generation, there are errors in the impl-class =
of the 0..* -side: in the following code-snipped the varable ownedrules =
is not declared in the class
- if I change the Multiplicity to 1, the resulting code works
public EList getOwnedrules() {

if (ownedrules =3D=3D null) {

ownedrules =3D new EObjectWithInverseResolvingEList(

MappingRule.class, this,

UMLTPackage.TRANSFORMATION_GROUP__OWNEDRULES,

UMLTPackage.MAPPING_RULE__TGROUP);

}

return ownedrules;

}


In the attachments there are some pictures of the .uml and =
..genmodel to get the context.

Anyone an idea
Regards, Andreas

------=_NextPart_000_0080_01C73B48.610BDD00
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 6.00.2900.2995" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>Hi Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>some comments below ...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- James.</FONT></DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt; =
wrote in=20
message <A=20
=
href=3D"news:eolgq1$q5s$1@utils.eclipse.org">news:eolgq1$q5s$1@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi James</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Thanks for your answer!</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Fortunately the problem disappeared =
unexpectedly=20
when I changed the Aggregation attribute of the memberEnd to "none". =
After=20
switching back to 'composite' everything remained ok...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>But after some experiences with the =
Genmodel=20
there are some further questions:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- For every implementation class =
there is a <FONT=20
size=3D2>getCacheAdapter() method generated (in the =
uml2.uml-impl-classes there=20
is no such method) - this leads to a stack overflow using the =
generated editor=20
plugin. Deleting the method manually resolves the problem. What am I =
doing=20
wrong with the UML-Model/Genmodel?</FONT></FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ Not sure about this one&nbsp; I'll =
have to=20
investigate and get back to you.&nbsp; </FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- In the UML.merged.uml-model used as =
metamodel,=20
what is the difference beween memberEnd Property with =
&lt;eReference&gt;=20
Stereotype and the ones without Stereotype?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ In general the Ecore stereotype =
provides=20
information when converting from UMLto Ecore.&nbsp; Specific code =
generation=20
details might be necessary for those items with stereotypes =
applied.&nbsp;=20
Sometimes information is provided in the stereotype to allow =
conversion from=20
Ecore to UML and back to Ecore and preserve the integrity of the =
original=20
.ecore model ( ie. round tripping ).</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- Using an UML-model as metamodel is =
there a=20
special behaviour in Genmodel when using ownedEnd's in associations =
(as in=20
A_general_generalization)?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ I'm not sure what you mean about =
"special=20
behavior".&nbsp;&nbsp; I'm not aware of unique behavior for =
this.&nbsp; I'll=20
discuss it with Kenn and get back to you if there is something unique=20
here.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>And finally a question which is maybe =
not just=20
related with uml2:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Getting all the sources from CVS =
(emf:=20
200701110200 uml2: 200701111331) on 3.3M4 there are several errors in =
the UML2=20
projects even after updating the classpathes (org.eclipse.uml2,=20
.uml2.importer.rose, . Is that just a strange behaviour of my=20
environment?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ I'm not seeing any build issues for =
the uml2=20
project.&nbsp; I do however have some issues with=20
org.eclipse.emf.importer.rose.&nbsp;&nbsp; .. may not be the same =
issue you=20
are having but these problems are a result of .jars needed for=20
ant.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Sometimes there are .readme's that =
explain how to=20
resolve these issues.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>eg.&nbsp;&nbsp; <FONT size=3D3><FONT=20
face=3D"Times New Roman">Before running your build scripts or the =
examples=20
provided here, make sure you have the 2 jar files required by this =
plugin=20
("importer.rose.jar" and "importer.rose.tasks.jar") and also the 2 jar =
files=20
required by the </FONT><TT>org.eclipse.emf.ant</TT><FONT=20
face=3D"Times New Roman"> plugin ("emf.ant.jar" and =
"emf.ant.tasks.jar"). If=20
necessary, Eclipse can generate them for you. All you need to do is to =
run the=20
Ant script created by PDE when you right click on the manifest file =
and select=20
"PDE Tools&gt;Create Ant Build File". </FONT></FONT></FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>but again, I don't see any issues =
with=20
uml.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Greets, Andreas</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"James Bruck" &lt;<A=20
href=3D"mailto:jbruck@ca.ibm.com">jbruck@ca.ibm.com</A>&gt; wrote in =
message=20
<A=20
=
href=3D"news:eo93md$g96$2@utils.eclipse.org">news:eo93md$g96$2@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>I just tried this on my machine =
with a=20
contrived example and it seemed to work fine ( If I understood this=20
correctly ;)&nbsp; ).</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Would it be possible for you to =
send more=20
information or a snippet of your metamodel so I can inspect =
it.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>What errors are you seeing in the =
impl=20
class?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Regards,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- James.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt;=20
wrote in message <A=20
=
href=3D"news:eo7rou$ob0$1@utils.eclipse.org">news:eo7rou$ob0$1@utils.ecli=
pse.org</A>...</DIV>
<DIV>Hi all</DIV>
<DIV>I'm trying to make a new language metamodel on the basis on =
UML.uml=20
and using UML-Genmodel. In principle, the process is working but =
with the=20
following aspect I get into trouble:</DIV>
<DIV>&nbsp;</DIV>
<DIV>- In my metamodel I have two classes with an 1 -- 0..*=20
association</DIV>
<DIV>- the creation of the genmodel works well</DIV>
<DIV>- after model-code generation, there are errors in the =
impl-class of=20
the 0..* -side: in the following code-snipped the varable =
ownedrules is=20
not declared in the class</DIV>
<DIV>- if I change the Multiplicity to 1, the resulting code =
works</DIV>
<DIV><FONT face=3DArial size=3D2>
<P align=3Dleft><B><FONT color=3D#7f0055 =
size=3D2>public</B></FONT><FONT size=3D2>=20
EList getOwnedrules() {</P>
<P align=3Dleft></FONT><B><FONT color=3D#7f0055 =
size=3D2>if</B></FONT><FONT=20
size=3D2> (ownedrules =3D=3D </FONT><B><FONT color=3D#7f0055=20
size=3D2>null</B></FONT><FONT size=3D2>) {</P>
<P align=3Dleft>ownedrules =3D </FONT><B><FONT color=3D#7f0055=20
size=3D2>new</B></FONT><FONT size=3D2> =
EObjectWithInverseResolvingEList(</P>
<P align=3Dleft>MappingRule.</FONT><B><FONT color=3D#7f0055=20
size=3D2>class</B></FONT><FONT size=3D2>, </FONT><B><FONT =
color=3D#7f0055=20
size=3D2>this</B></FONT><FONT size=3D2>,</P>
<P align=3Dleft>UMLTPackage.</FONT><I><FONT color=3D#0000c0=20
size=3D2>TRANSFORMATION_GROUP__OWNEDRULES</I></FONT><FONT =
size=3D2>,</P>
<P align=3Dleft>UMLTPackage.</FONT><I><FONT color=3D#0000c0=20
size=3D2>MAPPING_RULE__TGROUP</I></FONT><FONT size=3D2>);</P>
<P align=3Dleft>}</P>
<P align=3Dleft></FONT><B><FONT color=3D#7f0055 =
size=3D2>return</B></FONT><FONT=20
size=3D2> ownedrules;</P>
<P align=3Dleft>}</FONT></FONT></P></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV>In the attachments there are some pictures of the .uml and =
..genmodel=20
to get the context.</DIV>
<DIV>&nbsp;</DIV>
<DIV>Anyone an idea</DIV>
<DIV>Regards, Andreas</DIV>
<DIV><FONT face=3DArial=20
size=3D2></FONT>&nbsp;</DIV></BLOCKQUOTE></BLOCKQUOTE ></BLOCKQUOTE></BODY=
></HTML>

------=_NextPart_000_0080_01C73B48.610BDD00--
Re: Problem using UML Genmodel [message #468525 is a reply to message #468510] Fri, 19 January 2007 03:01 Go to previous messageGo to next message
Kenn Hussey is currently offline Kenn HusseyFriend
Messages: 1620
Registered: July 2009
Senior Member
This is a multi-part message in MIME format.

------=_NextPart_000_078A_01C73B4C.33147340
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Andreas,

Regarding the errors, please ensure that you only have the following =
projects checked out (from HEAD):

org.eclipse.uml2-feature
org.eclipse.uml2-plugin
org.eclipse.uml2.codegen.ecore
org.eclipse.uml2.codegen.ecore.ui
org.eclipse.uml2.common
org.eclipse.uml2.common.edit
org.eclipse.uml2.doc-feature
org.eclipse.uml2.doc-plugin
org.eclipse.uml2.examples-feature
org.eclipse.uml2.examples-plugin
org.eclipse.uml2.examples.uml.ui
org.eclipse.uml2.sdk-feature
org.eclipse.uml2.tests-feature
org.eclipse.uml2.tests-plugin
org.eclipse.uml2.uml
org.eclipse.uml2.uml.ecore.exporter
org.eclipse.uml2.uml.ecore.importer
org.eclipse.uml2.uml.edit
org.eclipse.uml2.uml.editor
org.eclipse.uml2.uml.resources
org.eclipse.uml2.uml.tests

Kenn

"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eolgq1$q5s$1@utils.eclipse.org...
Hi James
Thanks for your answer!
Fortunately the problem disappeared unexpectedly when I changed the =
Aggregation attribute of the memberEnd to "none". After switching back =
to 'composite' everything remained ok...

But after some experiences with the Genmodel there are some further =
questions:

- For every implementation class there is a getCacheAdapter() method =
generated (in the uml2.uml-impl-classes there is no such method) - this =
leads to a stack overflow using the generated editor plugin. Deleting =
the method manually resolves the problem. What am I doing wrong with the =
UML-Model/Genmodel?
- In the UML.merged.uml-model used as metamodel, what is the =
difference beween memberEnd Property with <eReference> Stereotype and =
the ones without Stereotype?
- Using an UML-model as metamodel is there a special behaviour in =
Genmodel when using ownedEnd's in associations (as in =
A_general_generalization)?

And finally a question which is maybe not just related with uml2:
Getting all the sources from CVS (emf: 200701110200 uml2: =
200701111331) on 3.3M4 there are several errors in the UML2 projects =
even after updating the classpathes (org.eclipse.uml2, =
..uml2.importer.rose, . Is that just a strange behaviour of my =
environment?

Greets, Andreas


"James Bruck" <jbruck@ca.ibm.com> wrote in message =
news:eo93md$g96$2@utils.eclipse.org...
Hi Andreas,

I just tried this on my machine with a contrived example and it =
seemed to work fine ( If I understood this correctly ;) ).
Would it be possible for you to send more information or a snippet =
of your metamodel so I can inspect it.
What errors are you seeing in the impl class?

Regards,

- James.

"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eo7rou$ob0$1@utils.eclipse.org...
Hi all
I'm trying to make a new language metamodel on the basis on =
UML.uml and using UML-Genmodel. In principle, the process is working but =
with the following aspect I get into trouble:

- In my metamodel I have two classes with an 1 -- 0..* association
- the creation of the genmodel works well
- after model-code generation, there are errors in the impl-class =
of the 0..* -side: in the following code-snipped the varable ownedrules =
is not declared in the class
- if I change the Multiplicity to 1, the resulting code works
public EList getOwnedrules() {

if (ownedrules =3D=3D null) {

ownedrules =3D new EObjectWithInverseResolvingEList(

MappingRule.class, this,

UMLTPackage.TRANSFORMATION_GROUP__OWNEDRULES,

UMLTPackage.MAPPING_RULE__TGROUP);

}

return ownedrules;

}


In the attachments there are some pictures of the .uml and =
..genmodel to get the context.

Anyone an idea
Regards, Andreas

------=_NextPart_000_078A_01C73B4C.33147340
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 6.00.2900.3020" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Regarding the errors, please ensure =
that you only=20
have the following projects checked out (from HEAD):</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV>org.eclipse.uml2-feature<BR>org.eclipse.uml2-plugin<BR >org.eclipse.u=
ml2.codegen.ecore<BR>org.eclipse.uml2.codegen.ecore.ui<BR >org.eclipse.uml=
2.common<BR>org.eclipse.uml2.common.edit<BR>org.eclipse.uml2.doc-feature <=
BR>org.eclipse.uml2.doc-plugin<BR>org.eclipse.uml2.examples-feature <BR>or=
g.eclipse.uml2.examples-plugin<BR>org.eclipse.uml2.examples.uml.ui <BR>org=
..eclipse.uml2.sdk-feature<BR>org.eclipse.uml2.tests-feature <BR>org.eclips=
e.uml2.tests-plugin<BR>org.eclipse.uml2.uml<BR>org.eclipse.uml2.uml.ecore=
..exporter<BR>org.eclipse.uml2.uml.ecore.importer<BR>org.eclipse.uml2.uml.=
edit<BR>org.eclipse.uml2.uml.editor<BR>org.eclipse.uml2.uml.resources <BR>=
org.eclipse.uml2.uml.tests<BR></DIV>
<DIV><FONT face=3DArial size=3D2>Kenn</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt; =
wrote in=20
message <A=20
=
href=3D"news:eolgq1$q5s$1@utils.eclipse.org">news:eolgq1$q5s$1@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi James</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Thanks for your answer!</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Fortunately the problem disappeared =
unexpectedly=20
when I changed the Aggregation attribute of the memberEnd to "none". =
After=20
switching back to 'composite' everything remained ok...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>But after some experiences with the =
Genmodel=20
there are some further questions:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- For every implementation class =
there is a <FONT=20
size=3D2>getCacheAdapter() method generated (in the =
uml2.uml-impl-classes there=20
is no such method) - this leads to a stack overflow using the =
generated editor=20
plugin. Deleting the method manually resolves the problem. What am I =
doing=20
wrong with the UML-Model/Genmodel?</FONT></FONT></DIV>
<DIV><FONT face=3DArial size=3D2>- In the UML.merged.uml-model used as =
metamodel,=20
what is the difference beween memberEnd Property with =
&lt;eReference&gt;=20
Stereotype and the ones without Stereotype?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>- Using an UML-model as metamodel is =
there a=20
special behaviour in Genmodel when using ownedEnd's in associations =
(as in=20
A_general_generalization)?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>And finally a question which is maybe =
not just=20
related with uml2:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Getting all the sources from CVS =
(emf:=20
200701110200 uml2: 200701111331) on 3.3M4 there are several errors in =
the UML2=20
projects even after updating the classpathes (org.eclipse.uml2,=20
.uml2.importer.rose, . Is that just a strange behaviour of my=20
environment?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Greets, Andreas</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"James Bruck" &lt;<A=20
href=3D"mailto:jbruck@ca.ibm.com">jbruck@ca.ibm.com</A>&gt; wrote in =
message=20
<A=20
=
href=3D"news:eo93md$g96$2@utils.eclipse.org">news:eo93md$g96$2@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>I just tried this on my machine =
with a=20
contrived example and it seemed to work fine ( If I understood this=20
correctly ;)&nbsp; ).</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Would it be possible for you to =
send more=20
information or a snippet of your metamodel so I can inspect =
it.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>What errors are you seeing in the =
impl=20
class?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Regards,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- James.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt;=20
wrote in message <A=20
=
href=3D"news:eo7rou$ob0$1@utils.eclipse.org">news:eo7rou$ob0$1@utils.ecli=
pse.org</A>...</DIV>
<DIV>Hi all</DIV>
<DIV>I'm trying to make a new language metamodel on the basis on =
UML.uml=20
and using UML-Genmodel. In principle, the process is working but =
with the=20
following aspect I get into trouble:</DIV>
<DIV>&nbsp;</DIV>
<DIV>- In my metamodel I have two classes with an 1 -- 0..*=20
association</DIV>
<DIV>- the creation of the genmodel works well</DIV>
<DIV>- after model-code generation, there are errors in the =
impl-class of=20
the 0..* -side: in the following code-snipped the varable =
ownedrules is=20
not declared in the class</DIV>
<DIV>- if I change the Multiplicity to 1, the resulting code =
works</DIV>
<DIV><FONT face=3DArial size=3D2>
<P align=3Dleft><B><FONT color=3D#7f0055 =
size=3D2>public</B></FONT><FONT size=3D2>=20
EList getOwnedrules() {</P>
<P align=3Dleft></FONT><B><FONT color=3D#7f0055 =
size=3D2>if</B></FONT><FONT=20
size=3D2> (ownedrules =3D=3D </FONT><B><FONT color=3D#7f0055=20
size=3D2>null</B></FONT><FONT size=3D2>) {</P>
<P align=3Dleft>ownedrules =3D </FONT><B><FONT color=3D#7f0055=20
size=3D2>new</B></FONT><FONT size=3D2> =
EObjectWithInverseResolvingEList(</P>
<P align=3Dleft>MappingRule.</FONT><B><FONT color=3D#7f0055=20
size=3D2>class</B></FONT><FONT size=3D2>, </FONT><B><FONT =
color=3D#7f0055=20
size=3D2>this</B></FONT><FONT size=3D2>,</P>
<P align=3Dleft>UMLTPackage.</FONT><I><FONT color=3D#0000c0=20
size=3D2>TRANSFORMATION_GROUP__OWNEDRULES</I></FONT><FONT =
size=3D2>,</P>
<P align=3Dleft>UMLTPackage.</FONT><I><FONT color=3D#0000c0=20
size=3D2>MAPPING_RULE__TGROUP</I></FONT><FONT size=3D2>);</P>
<P align=3Dleft>}</P>
<P align=3Dleft></FONT><B><FONT color=3D#7f0055 =
size=3D2>return</B></FONT><FONT=20
size=3D2> ownedrules;</P>
<P align=3Dleft>}</FONT></FONT></P></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV>In the attachments there are some pictures of the .uml and =
..genmodel=20
to get the context.</DIV>
<DIV>&nbsp;</DIV>
<DIV>Anyone an idea</DIV>
<DIV>Regards, Andreas</DIV>
<DIV><FONT face=3DArial=20
size=3D2></FONT>&nbsp;</DIV></BLOCKQUOTE></BLOCKQUOTE ></BLOCKQUOTE></BODY=
></HTML>

------=_NextPart_000_078A_01C73B4C.33147340--
Re: Problem using UML Genmodel [message #468526 is a reply to message #468524] Fri, 19 January 2007 03:10 Go to previous messageGo to next message
Kenn Hussey is currently offline Kenn HusseyFriend
Messages: 1620
Registered: July 2009
Senior Member
This is a multi-part message in MIME format.

------=_NextPart_000_079D_01C73B4D.6CDA39B0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

- Could you provide more details on the stack overflow you are seeing? =
Every "model root" class will have this method in it, but it's not clear =
to me why this would result in a stack overflow...

- Association ends that are owned by the assocation are ignored when =
converting UML models to Ecore.

Kenn
"James Bruck" <jbruck@ca.ibm.com> wrote in message =
news:eopai8$174$1@utils.eclipse.org...
Hi Andreas,

some comments below ...

- James.
"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eolgq1$q5s$1@utils.eclipse.org...
Hi James
Thanks for your answer!
Fortunately the problem disappeared unexpectedly when I changed the =
Aggregation attribute of the memberEnd to "none". After switching back =
to 'composite' everything remained ok...

But after some experiences with the Genmodel there are some further =
questions:

- For every implementation class there is a getCacheAdapter() method =
generated (in the uml2.uml-impl-classes there is no such method) - this =
leads to a stack overflow using the generated editor plugin. Deleting =
the method manually resolves the problem. What am I doing wrong with the =
UML-Model/Genmodel?

+ Not sure about this one I'll have to investigate and get back to =
you. =20

- In the UML.merged.uml-model used as metamodel, what is the =
difference beween memberEnd Property with <eReference> Stereotype and =
the ones without Stereotype?

+ In general the Ecore stereotype provides information when =
converting from UMLto Ecore. Specific code generation details might be =
necessary for those items with stereotypes applied. Sometimes =
information is provided in the stereotype to allow conversion from Ecore =
to UML and back to Ecore and preserve the integrity of the original =
..ecore model ( ie. round tripping ).

- Using an UML-model as metamodel is there a special behaviour in =
Genmodel when using ownedEnd's in associations (as in =
A_general_generalization)?

+ I'm not sure what you mean about "special behavior". I'm not =
aware of unique behavior for this. I'll discuss it with Kenn and get =
back to you if there is something unique here.

And finally a question which is maybe not just related with uml2:
Getting all the sources from CVS (emf: 200701110200 uml2: =
200701111331) on 3.3M4 there are several errors in the UML2 projects =
even after updating the classpathes (org.eclipse.uml2, =
..uml2.importer.rose, . Is that just a strange behaviour of my =
environment?

+ I'm not seeing any build issues for the uml2 project. I do =
however have some issues with org.eclipse.emf.importer.rose. .. may =
not be the same issue you are having but these problems are a result of =
..jars needed for ant.
Sometimes there are .readme's that explain how to resolve these =
issues.

eg. Before running your build scripts or the examples provided =
here, make sure you have the 2 jar files required by this plugin =
("importer.rose.jar" and "importer.rose.tasks.jar") and also the 2 jar =
files required by the org.eclipse.emf.ant plugin ("emf.ant.jar" and =
"emf.ant.tasks.jar"). If necessary, Eclipse can generate them for you. =
All you need to do is to run the Ant script created by PDE when you =
right click on the manifest file and select "PDE Tools>Create Ant Build =
File".=20

but again, I don't see any issues with uml.


Greets, Andreas


"James Bruck" <jbruck@ca.ibm.com> wrote in message =
news:eo93md$g96$2@utils.eclipse.org...
Hi Andreas,

I just tried this on my machine with a contrived example and it =
seemed to work fine ( If I understood this correctly ;) ).
Would it be possible for you to send more information or a snippet =
of your metamodel so I can inspect it.
What errors are you seeing in the impl class?

Regards,

- James.

"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eo7rou$ob0$1@utils.eclipse.org...
Hi all
I'm trying to make a new language metamodel on the basis on =
UML.uml and using UML-Genmodel. In principle, the process is working but =
with the following aspect I get into trouble:

- In my metamodel I have two classes with an 1 -- 0..* =
association
- the creation of the genmodel works well
- after model-code generation, there are errors in the =
impl-class of the 0..* -side: in the following code-snipped the varable =
ownedrules is not declared in the class
- if I change the Multiplicity to 1, the resulting code works
public EList getOwnedrules() {

if (ownedrules =3D=3D null) {

ownedrules =3D new EObjectWithInverseResolvingEList(

MappingRule.class, this,

UMLTPackage.TRANSFORMATION_GROUP__OWNEDRULES,

UMLTPackage.MAPPING_RULE__TGROUP);

}

return ownedrules;

}


In the attachments there are some pictures of the .uml and =
..genmodel to get the context.

Anyone an idea
Regards, Andreas

------=_NextPart_000_079D_01C73B4D.6CDA39B0
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 6.00.2900.3020" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>- Could you provide more details on the =
stack=20
overflow you are seeing? Every "model root" class will have this method =
in it,=20
but it's not clear to me why this would result in a stack=20
overflow...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- Association ends that are owned by =
the assocation=20
are ignored when converting UML models to Ecore.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Kenn</FONT></DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"James Bruck" &lt;<A=20
href=3D"mailto:jbruck@ca.ibm.com">jbruck@ca.ibm.com</A>&gt; wrote in =
message <A=20
=
href=3D"news:eopai8$174$1@utils.eclipse.org">news:eopai8$174$1@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>some comments below ...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- James.</FONT></DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt; =
wrote=20
in message <A=20
=
href=3D"news:eolgq1$q5s$1@utils.eclipse.org">news:eolgq1$q5s$1@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi James</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Thanks for your =
answer!</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Fortunately the problem disappeared =

unexpectedly when I changed the Aggregation attribute of the =
memberEnd to=20
"none". After switching back to 'composite' everything remained=20
ok...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>But after some experiences with the =
Genmodel=20
there are some further questions:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- For every implementation class =
there is a=20
<FONT size=3D2>getCacheAdapter() method generated (in the=20
uml2.uml-impl-classes there is no such method) - this leads to a =
stack=20
overflow using the generated editor plugin. Deleting the method =
manually=20
resolves the problem. What am I doing wrong with the=20
UML-Model/Genmodel?</FONT></FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ Not sure about this one&nbsp; =
I'll have to=20
investigate and get back to you.&nbsp; </FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- In the UML.merged.uml-model used =
as=20
metamodel, what is the difference beween memberEnd Property with=20
&lt;eReference&gt; Stereotype and the ones without =
Stereotype?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ In general the Ecore stereotype =
provides=20
information when converting from UMLto Ecore.&nbsp; Specific code =
generation=20
details might be necessary for those items with stereotypes =
applied.&nbsp;=20
Sometimes information is provided in the stereotype to allow =
conversion from=20
Ecore to UML and back to Ecore and preserve the integrity of the =
original=20
.ecore model ( ie. round tripping ).</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- Using an UML-model as metamodel =
is there a=20
special behaviour in Genmodel when using ownedEnd's in associations =
(as in=20
A_general_generalization)?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ I'm not sure what you mean about =
"special=20
behavior".&nbsp;&nbsp; I'm not aware of unique behavior for =
this.&nbsp; I'll=20
discuss it with Kenn and get back to you if there is something =
unique=20
here.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>And finally a question which is =
maybe not just=20
related with uml2:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Getting all the sources from CVS =
(emf:=20
200701110200 uml2: 200701111331) on 3.3M4 there are several errors =
in the=20
UML2 projects even after updating the classpathes (org.eclipse.uml2, =

.uml2.importer.rose, . Is that just a strange behaviour of my=20
environment?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ I'm not seeing any build issues =
for the uml2=20
project.&nbsp; I do however have some issues with=20
org.eclipse.emf.importer.rose.&nbsp;&nbsp; .. may not be the same =
issue you=20
are having but these problems are a result of .jars needed for=20
ant.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Sometimes there are .readme's that =
explain how=20
to resolve these issues.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>eg.&nbsp;&nbsp; <FONT =
size=3D3><FONT=20
face=3D"Times New Roman">Before running your build scripts or the =
examples=20
provided here, make sure you have the 2 jar files required by this =
plugin=20
("importer.rose.jar" and "importer.rose.tasks.jar") and also the 2 =
jar files=20
required by the </FONT><TT>org.eclipse.emf.ant</TT><FONT=20
face=3D"Times New Roman"> plugin ("emf.ant.jar" and =
"emf.ant.tasks.jar"). If=20
necessary, Eclipse can generate them for you. All you need to do is =
to run=20
the Ant script created by PDE when you right click on the manifest =
file and=20
select "PDE Tools&gt;Create Ant Build File". =
</FONT></FONT></FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>but again, I don't see any issues =
with=20
uml.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Greets, Andreas</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"James Bruck" &lt;<A=20
href=3D"mailto:jbruck@ca.ibm.com">jbruck@ca.ibm.com</A>&gt; wrote =
in message=20
<A=20
=
href=3D"news:eo93md$g96$2@utils.eclipse.org">news:eo93md$g96$2@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>I just tried this on my machine =
with a=20
contrived example and it seemed to work fine ( If I understood =
this=20
correctly ;)&nbsp; ).</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Would it be possible for you to =
send more=20
information or a snippet of your metamodel so I can inspect=20
it.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>What errors are you seeing in the =
impl=20
class?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Regards,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- James.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt;=20
wrote in message <A=20
=
href=3D"news:eo7rou$ob0$1@utils.eclipse.org">news:eo7rou$ob0$1@utils.ecli=
pse.org</A>...</DIV>
<DIV>Hi all</DIV>
<DIV>I'm trying to make a new language metamodel on the basis on =
UML.uml=20
and using UML-Genmodel. In principle, the process is working but =
with=20
the following aspect I get into trouble:</DIV>
<DIV>&nbsp;</DIV>
<DIV>- In my metamodel I have two classes with an 1 -- 0..*=20
association</DIV>
<DIV>- the creation of the genmodel works well</DIV>
<DIV>- after model-code generation, there are errors in the =
impl-class=20
of the 0..* -side: in the following code-snipped the varable =
ownedrules=20
is not declared in the class</DIV>
<DIV>- if I change the Multiplicity to 1, the resulting code =
works</DIV>
<DIV><FONT face=3DArial size=3D2>
<P align=3Dleft><B><FONT color=3D#7f0055 =
size=3D2>public</B></FONT><FONT=20
size=3D2> EList getOwnedrules() {</P>
<P align=3Dleft></FONT><B><FONT color=3D#7f0055 =
size=3D2>if</B></FONT><FONT=20
size=3D2> (ownedrules =3D=3D </FONT><B><FONT color=3D#7f0055=20
size=3D2>null</B></FONT><FONT size=3D2>) {</P>
<P align=3Dleft>ownedrules =3D </FONT><B><FONT color=3D#7f0055=20
size=3D2>new</B></FONT><FONT size=3D2> =
EObjectWithInverseResolvingEList(</P>
<P align=3Dleft>MappingRule.</FONT><B><FONT color=3D#7f0055=20
size=3D2>class</B></FONT><FONT size=3D2>, </FONT><B><FONT =
color=3D#7f0055=20
size=3D2>this</B></FONT><FONT size=3D2>,</P>
<P align=3Dleft>UMLTPackage.</FONT><I><FONT color=3D#0000c0=20
size=3D2>TRANSFORMATION_GROUP__OWNEDRULES</I></FONT><FONT =
size=3D2>,</P>
<P align=3Dleft>UMLTPackage.</FONT><I><FONT color=3D#0000c0=20
size=3D2>MAPPING_RULE__TGROUP</I></FONT><FONT size=3D2>);</P>
<P align=3Dleft>}</P>
<P align=3Dleft></FONT><B><FONT color=3D#7f0055=20
size=3D2>return</B></FONT><FONT size=3D2> ownedrules;</P>
<P align=3Dleft>}</FONT></FONT></P></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV>In the attachments there are some pictures of the .uml and=20
.genmodel to get the context.</DIV>
<DIV>&nbsp;</DIV>
<DIV>Anyone an idea</DIV>
<DIV>Regards, Andreas</DIV>
<DIV><FONT face=3DArial=20
=
size=3D2></FONT>&nbsp;</DIV></BLOCKQUOTE></BLOCKQUOTE ></BLOCKQUOTE></BLOC=
KQUOTE></BODY></HTML>

------=_NextPart_000_079D_01C73B4D.6CDA39B0--
Re: Problem using UML Genmodel [message #468690 is a reply to message #468526] Fri, 19 January 2007 16:49 Go to previous messageGo to next message
Andreas Morf is currently offline Andreas MorfFriend
Messages: 14
Registered: July 2009
Junior Member
This is a multi-part message in MIME format.

------=_NextPart_000_0036_01C73BF2.2DD55AB0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Kenn,
I don't know how and why it comes, but now, there is no such method =
anymore when generating the model code - strange!
I will send you the details as soon as the problem occurs again...
So, how do you markup a class as a "model root" class in the genmodel?

Andreas
"Kenn Hussey" <khussey@ca.ibm.com> wrote in message =
news:eopcql$mcq$1@utils.eclipse.org...
- Could you provide more details on the stack overflow you are seeing? =
Every "model root" class will have this method in it, but it's not clear =
to me why this would result in a stack overflow...

- Association ends that are owned by the assocation are ignored when =
converting UML models to Ecore.

Kenn
"James Bruck" <jbruck@ca.ibm.com> wrote in message =
news:eopai8$174$1@utils.eclipse.org...
Hi Andreas,

some comments below ...

- James.
"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eolgq1$q5s$1@utils.eclipse.org...
Hi James
Thanks for your answer!
Fortunately the problem disappeared unexpectedly when I changed =
the Aggregation attribute of the memberEnd to "none". After switching =
back to 'composite' everything remained ok...

But after some experiences with the Genmodel there are some =
further questions:

- For every implementation class there is a getCacheAdapter() =
method generated (in the uml2.uml-impl-classes there is no such method) =
- this leads to a stack overflow using the generated editor plugin. =
Deleting the method manually resolves the problem. What am I doing wrong =
with the UML-Model/Genmodel?

+ Not sure about this one I'll have to investigate and get back =
to you. =20

- In the UML.merged.uml-model used as metamodel, what is the =
difference beween memberEnd Property with <eReference> Stereotype and =
the ones without Stereotype?

+ In general the Ecore stereotype provides information when =
converting from UMLto Ecore. Specific code generation details might be =
necessary for those items with stereotypes applied. Sometimes =
information is provided in the stereotype to allow conversion from Ecore =
to UML and back to Ecore and preserve the integrity of the original =
..ecore model ( ie. round tripping ).

- Using an UML-model as metamodel is there a special behaviour in =
Genmodel when using ownedEnd's in associations (as in =
A_general_generalization)?

+ I'm not sure what you mean about "special behavior". I'm not =
aware of unique behavior for this. I'll discuss it with Kenn and get =
back to you if there is something unique here.

And finally a question which is maybe not just related with uml2:
Getting all the sources from CVS (emf: 200701110200 uml2: =
200701111331) on 3.3M4 there are several errors in the UML2 projects =
even after updating the classpathes (org.eclipse.uml2, =
..uml2.importer.rose, . Is that just a strange behaviour of my =
environment?

+ I'm not seeing any build issues for the uml2 project. I do =
however have some issues with org.eclipse.emf.importer.rose. .. may =
not be the same issue you are having but these problems are a result of =
..jars needed for ant.
Sometimes there are .readme's that explain how to resolve these =
issues.

eg. Before running your build scripts or the examples provided =
here, make sure you have the 2 jar files required by this plugin =
("importer.rose.jar" and "importer.rose.tasks.jar") and also the 2 jar =
files required by the org.eclipse.emf.ant plugin ("emf.ant.jar" and =
"emf.ant.tasks.jar"). If necessary, Eclipse can generate them for you. =
All you need to do is to run the Ant script created by PDE when you =
right click on the manifest file and select "PDE Tools>Create Ant Build =
File".=20

but again, I don't see any issues with uml.


Greets, Andreas


"James Bruck" <jbruck@ca.ibm.com> wrote in message =
news:eo93md$g96$2@utils.eclipse.org...
Hi Andreas,

I just tried this on my machine with a contrived example and it =
seemed to work fine ( If I understood this correctly ;) ).
Would it be possible for you to send more information or a =
snippet of your metamodel so I can inspect it.
What errors are you seeing in the impl class?

Regards,

- James.

"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eo7rou$ob0$1@utils.eclipse.org...
Hi all
I'm trying to make a new language metamodel on the basis on =
UML.uml and using UML-Genmodel. In principle, the process is working but =
with the following aspect I get into trouble:

- In my metamodel I have two classes with an 1 -- 0..* =
association
- the creation of the genmodel works well
- after model-code generation, there are errors in the =
impl-class of the 0..* -side: in the following code-snipped the varable =
ownedrules is not declared in the class
- if I change the Multiplicity to 1, the resulting code works
public EList getOwnedrules() {

if (ownedrules =3D=3D null) {

ownedrules =3D new EObjectWithInverseResolvingEList(

MappingRule.class, this,

UMLTPackage.TRANSFORMATION_GROUP__OWNEDRULES,

UMLTPackage.MAPPING_RULE__TGROUP);

}

return ownedrules;

}


In the attachments there are some pictures of the .uml and =
..genmodel to get the context.

Anyone an idea
Regards, Andreas

------=_NextPart_000_0036_01C73BF2.2DD55AB0
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 6.00.5730.11" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>Kenn,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>I don't know how and why it comes, but =
now, there=20
is no such method anymore when generating the model code - =
strange!</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>I will send you the details as soon as =
the problem=20
occurs again...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>So, how do you markup a class as a =
"model root"=20
class in the genmodel?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Andreas</FONT></DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Kenn Hussey" &lt;<A=20
href=3D"mailto:khussey@ca.ibm.com">khussey@ca.ibm.com</A>&gt; wrote in =
message=20
<A=20
=
href=3D"news:eopcql$mcq$1@utils.eclipse.org">news:eopcql$mcq$1@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>- Could you provide more details on =
the stack=20
overflow you are seeing? Every "model root" class will have this =
method in it,=20
but it's not clear to me why this would result in a stack=20
overflow...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- Association ends that are owned by =
the=20
assocation are ignored when converting UML models to =
Ecore.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Kenn</FONT></DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"James Bruck" &lt;<A=20
href=3D"mailto:jbruck@ca.ibm.com">jbruck@ca.ibm.com</A>&gt; wrote in =
message=20
<A=20
=
href=3D"news:eopai8$174$1@utils.eclipse.org">news:eopai8$174$1@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>some comments below =
....</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- James.</FONT></DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt;=20
wrote in message <A=20
=
href=3D"news:eolgq1$q5s$1@utils.eclipse.org">news:eolgq1$q5s$1@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi James</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Thanks for your =
answer!</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Fortunately the problem =
disappeared=20
unexpectedly when I changed the Aggregation attribute of the =
memberEnd to=20
"none". After switching back to 'composite' everything remained=20
ok...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>But after some experiences with =
the Genmodel=20
there are some further questions:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- For every implementation class =
there is a=20
<FONT size=3D2>getCacheAdapter() method generated (in the=20
uml2.uml-impl-classes there is no such method) - this leads to a =
stack=20
overflow using the generated editor plugin. Deleting the method =
manually=20
resolves the problem. What am I doing wrong with the=20
UML-Model/Genmodel?</FONT></FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ Not sure about this one&nbsp; =
I'll have to=20
investigate and get back to you.&nbsp; </FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- In the UML.merged.uml-model =
used as=20
metamodel, what is the difference beween memberEnd Property with=20
&lt;eReference&gt; Stereotype and the ones without=20
Stereotype?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ In general the Ecore stereotype =
provides=20
information when converting from UMLto Ecore.&nbsp; Specific code=20
generation details might be necessary for those items with =
stereotypes=20
applied.&nbsp; Sometimes information is provided in the stereotype =
to=20
allow conversion from Ecore to UML and back to Ecore and preserve =
the=20
integrity of the original .ecore model ( ie. round tripping=20
).</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- Using an UML-model as metamodel =
is there a=20
special behaviour in Genmodel when using ownedEnd's in =
associations (as in=20
A_general_generalization)?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ I'm not sure what you mean =
about "special=20
behavior".&nbsp;&nbsp; I'm not aware of unique behavior for =
this.&nbsp;=20
I'll discuss it with Kenn and get back to you if there is =
something unique=20
here.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>And finally a question which is =
maybe not=20
just related with uml2:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Getting all the sources from CVS =
(emf:=20
200701110200 uml2: 200701111331) on 3.3M4 there are several errors =
in the=20
UML2 projects even after updating the classpathes =
(org.eclipse.uml2,=20
.uml2.importer.rose, . Is that just a strange behaviour of my=20
environment?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ I'm not seeing any build issues =
for the=20
uml2 project.&nbsp; I do however have some issues with=20
org.eclipse.emf.importer.rose.&nbsp;&nbsp; .. may not be the same =
issue=20
you are having but these problems are a result of .jars needed for =

ant.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Sometimes there are .readme's =
that explain=20
how to resolve these issues.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>eg.&nbsp;&nbsp; <FONT =
size=3D3><FONT=20
face=3D"Times New Roman">Before running your build scripts or the =
examples=20
provided here, make sure you have the 2 jar files required by this =
plugin=20
("importer.rose.jar" and "importer.rose.tasks.jar") and also the 2 =
jar=20
files required by the </FONT><TT>org.eclipse.emf.ant</TT><FONT=20
face=3D"Times New Roman"> plugin ("emf.ant.jar" and =
"emf.ant.tasks.jar"). If=20
necessary, Eclipse can generate them for you. All you need to do =
is to run=20
the Ant script created by PDE when you right click on the manifest =
file=20
and select "PDE Tools&gt;Create Ant Build File".=20
</FONT></FONT></FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>but again, I don't see any issues =
with=20
uml.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Greets, Andreas</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"James Bruck" &lt;<A=20
href=3D"mailto:jbruck@ca.ibm.com">jbruck@ca.ibm.com</A>&gt; =
wrote in=20
message <A=20
=
href=3D"news:eo93md$g96$2@utils.eclipse.org">news:eo93md$g96$2@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>I just tried this on my machine =
with a=20
contrived example and it seemed to work fine ( If I understood =
this=20
correctly ;)&nbsp; ).</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Would it be possible for you to =
send more=20
information or a snippet of your metamodel so I can inspect=20
it.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>What errors are you seeing in =
the impl=20
class?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Regards,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- James.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: =
5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt;=20
wrote in message <A=20
=
href=3D"news:eo7rou$ob0$1@utils.eclipse.org">news:eo7rou$ob0$1@utils.ecli=
pse.org</A>...</DIV>
<DIV>Hi all</DIV>
<DIV>I'm trying to make a new language metamodel on the basis =
on=20
UML.uml and using UML-Genmodel. In principle, the process is =
working=20
but with the following aspect I get into trouble:</DIV>
<DIV>&nbsp;</DIV>
<DIV>- In my metamodel I have two classes with an 1 -- 0..*=20
association</DIV>
<DIV>- the creation of the genmodel works well</DIV>
<DIV>- after model-code generation, there are errors in the =
impl-class=20
of the 0..* -side: in the following code-snipped the varable=20
ownedrules is not declared in the class</DIV>
<DIV>- if I change the Multiplicity to 1, the resulting code=20
works</DIV>
<DIV><FONT face=3DArial size=3D2>
<P align=3Dleft><B><FONT color=3D#7f0055 =
size=3D2>public</B></FONT><FONT=20
size=3D2> EList getOwnedrules() {</P>
<P align=3Dleft></FONT><B><FONT color=3D#7f0055 =
size=3D2>if</B></FONT><FONT=20
size=3D2> (ownedrules =3D=3D </FONT><B><FONT color=3D#7f0055=20
size=3D2>null</B></FONT><FONT size=3D2>) {</P>
<P align=3Dleft>ownedrules =3D </FONT><B><FONT color=3D#7f0055 =

size=3D2>new</B></FONT><FONT size=3D2>=20
EObjectWithInverseResolvingEList(</P>
<P align=3Dleft>MappingRule.</FONT><B><FONT color=3D#7f0055=20
size=3D2>class</B></FONT><FONT size=3D2>, </FONT><B><FONT =
color=3D#7f0055=20
size=3D2>this</B></FONT><FONT size=3D2>,</P>
<P align=3Dleft>UMLTPackage.</FONT><I><FONT color=3D#0000c0=20
size=3D2>TRANSFORMATION_GROUP__OWNEDRULES</I></FONT><FONT =
size=3D2>,</P>
<P align=3Dleft>UMLTPackage.</FONT><I><FONT color=3D#0000c0=20
size=3D2>MAPPING_RULE__TGROUP</I></FONT><FONT size=3D2>);</P>
<P align=3Dleft>}</P>
<P align=3Dleft></FONT><B><FONT color=3D#7f0055=20
size=3D2>return</B></FONT><FONT size=3D2> ownedrules;</P>
<P align=3Dleft>}</FONT></FONT></P></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV>In the attachments there are some pictures of the .uml =
and=20
.genmodel to get the context.</DIV>
<DIV>&nbsp;</DIV>
<DIV>Anyone an idea</DIV>
<DIV>Regards, Andreas</DIV>
<DIV><FONT face=3DArial=20
=
size=3D2></FONT>&nbsp;</DIV></BLOCKQUOTE></BLOCKQUOTE ></BLOCKQUOTE></BLOC=
KQUOTE></BLOCKQUOTE></BODY></HTML>

------=_NextPart_000_0036_01C73BF2.2DD55AB0--
Re: Problem using UML Genmodel [message #468693 is a reply to message #468690] Fri, 19 January 2007 18:05 Go to previous messageGo to next message
Kenn Hussey is currently offline Kenn HusseyFriend
Messages: 1620
Registered: July 2009
Senior Member
This is a multi-part message in MIME format.

------=_NextPart_000_07E8_01C73BCA.837935E0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Andreas,

Whether a class is considered a model root is determined in EMF like =
this:

public boolean isModelRoot()
{
return getClassExtendsGenClass() =3D=3D null || =
getClassExtendsGenClass().getGenModel() !=3D getGenModel();
}

Kenn

"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eoqsqu$i5o$1@utils.eclipse.org...
Kenn,
I don't know how and why it comes, but now, there is no such method =
anymore when generating the model code - strange!
I will send you the details as soon as the problem occurs again...
So, how do you markup a class as a "model root" class in the genmodel?

Andreas
"Kenn Hussey" <khussey@ca.ibm.com> wrote in message =
news:eopcql$mcq$1@utils.eclipse.org...
- Could you provide more details on the stack overflow you are =
seeing? Every "model root" class will have this method in it, but it's =
not clear to me why this would result in a stack overflow...

- Association ends that are owned by the assocation are ignored when =
converting UML models to Ecore.

Kenn
"James Bruck" <jbruck@ca.ibm.com> wrote in message =
news:eopai8$174$1@utils.eclipse.org...
Hi Andreas,

some comments below ...

- James.
"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eolgq1$q5s$1@utils.eclipse.org...
Hi James
Thanks for your answer!
Fortunately the problem disappeared unexpectedly when I changed =
the Aggregation attribute of the memberEnd to "none". After switching =
back to 'composite' everything remained ok...

But after some experiences with the Genmodel there are some =
further questions:

- For every implementation class there is a getCacheAdapter() =
method generated (in the uml2.uml-impl-classes there is no such method) =
- this leads to a stack overflow using the generated editor plugin. =
Deleting the method manually resolves the problem. What am I doing wrong =
with the UML-Model/Genmodel?

+ Not sure about this one I'll have to investigate and get back =
to you. =20

- In the UML.merged.uml-model used as metamodel, what is the =
difference beween memberEnd Property with <eReference> Stereotype and =
the ones without Stereotype?

+ In general the Ecore stereotype provides information when =
converting from UMLto Ecore. Specific code generation details might be =
necessary for those items with stereotypes applied. Sometimes =
information is provided in the stereotype to allow conversion from Ecore =
to UML and back to Ecore and preserve the integrity of the original =
..ecore model ( ie. round tripping ).

- Using an UML-model as metamodel is there a special behaviour =
in Genmodel when using ownedEnd's in associations (as in =
A_general_generalization)?

+ I'm not sure what you mean about "special behavior". I'm not =
aware of unique behavior for this. I'll discuss it with Kenn and get =
back to you if there is something unique here.

And finally a question which is maybe not just related with =
uml2:
Getting all the sources from CVS (emf: 200701110200 uml2: =
200701111331) on 3.3M4 there are several errors in the UML2 projects =
even after updating the classpathes (org.eclipse.uml2, =
..uml2.importer.rose, . Is that just a strange behaviour of my =
environment?

+ I'm not seeing any build issues for the uml2 project. I do =
however have some issues with org.eclipse.emf.importer.rose. .. may =
not be the same issue you are having but these problems are a result of =
..jars needed for ant.
Sometimes there are .readme's that explain how to resolve these =
issues.

eg. Before running your build scripts or the examples provided =
here, make sure you have the 2 jar files required by this plugin =
("importer.rose.jar" and "importer.rose.tasks.jar") and also the 2 jar =
files required by the org.eclipse.emf.ant plugin ("emf.ant.jar" and =
"emf.ant.tasks.jar"). If necessary, Eclipse can generate them for you. =
All you need to do is to run the Ant script created by PDE when you =
right click on the manifest file and select "PDE Tools>Create Ant Build =
File".=20

but again, I don't see any issues with uml.


Greets, Andreas


"James Bruck" <jbruck@ca.ibm.com> wrote in message =
news:eo93md$g96$2@utils.eclipse.org...
Hi Andreas,

I just tried this on my machine with a contrived example and =
it seemed to work fine ( If I understood this correctly ;) ).
Would it be possible for you to send more information or a =
snippet of your metamodel so I can inspect it.
What errors are you seeing in the impl class?

Regards,

- James.

"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eo7rou$ob0$1@utils.eclipse.org...
Hi all
I'm trying to make a new language metamodel on the basis on =
UML.uml and using UML-Genmodel. In principle, the process is working but =
with the following aspect I get into trouble:

- In my metamodel I have two classes with an 1 -- 0..* =
association
- the creation of the genmodel works well
- after model-code generation, there are errors in the =
impl-class of the 0..* -side: in the following code-snipped the varable =
ownedrules is not declared in the class
- if I change the Multiplicity to 1, the resulting code =
works
public EList getOwnedrules() {

if (ownedrules =3D=3D null) {

ownedrules =3D new EObjectWithInverseResolvingEList(

MappingRule.class, this,

UMLTPackage.TRANSFORMATION_GROUP__OWNEDRULES,

UMLTPackage.MAPPING_RULE__TGROUP);

}

return ownedrules;

}


In the attachments there are some pictures of the .uml and =
..genmodel to get the context.

Anyone an idea
Regards, Andreas

------=_NextPart_000_07E8_01C73BCA.837935E0
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 6.00.2900.3020" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Whether a class is considered a model =
root is=20
determined in EMF like this:</FONT></DIV>
<DIV><B><FONT color=3D#7f0055 size=3D2></FONT></B>&nbsp;</DIV>
<DIV><B><FONT color=3D#7f0055 size=3D2>public</B></FONT><FONT size=3D2>=20
</FONT><B><FONT color=3D#7f0055 size=3D2>boolean</B></FONT><FONT =
size=3D2>=20
isModelRoot()</FONT></DIV>
<DIV><FONT size=3D2>{</FONT></DIV>
<DIV><B><FONT color=3D#7f0055 size=3D2>&nbsp;&nbsp;&nbsp; =
return</B></FONT><FONT=20
size=3D2> getClassExtendsGenClass() =3D=3D </FONT><B><FONT =
color=3D#7f0055=20
size=3D2>null</B></FONT><FONT size=3D2> || =
getClassExtendsGenClass().getGenModel()=20
!=3D getGenModel();</FONT></DIV>
<DIV><FONT size=3D2>}</DIV></FONT>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Kenn</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt; =
wrote in=20
message <A=20
=
href=3D"news:eoqsqu$i5o$1@utils.eclipse.org">news:eoqsqu$i5o$1@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Kenn,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>I don't know how and why it comes, =
but now, there=20
is no such method anymore when generating the model code -=20
strange!</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>I will send you the details as soon =
as the=20
problem occurs again...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>So, how do you markup a class as a =
"model root"=20
class in the genmodel?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Andreas</FONT></DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Kenn Hussey" &lt;<A=20
href=3D"mailto:khussey@ca.ibm.com">khussey@ca.ibm.com</A>&gt; wrote =
in message=20
<A=20
=
href=3D"news:eopcql$mcq$1@utils.eclipse.org">news:eopcql$mcq$1@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>- Could you provide more details on =
the stack=20
overflow you are seeing? Every "model root" class will have this =
method in=20
it, but it's not clear to me why this would result in a stack=20
overflow...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- Association ends that are owned =
by the=20
assocation are ignored when converting UML models to =
Ecore.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Kenn</FONT></DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"James Bruck" &lt;<A=20
href=3D"mailto:jbruck@ca.ibm.com">jbruck@ca.ibm.com</A>&gt; wrote =
in message=20
<A=20
=
href=3D"news:eopai8$174$1@utils.eclipse.org">news:eopai8$174$1@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>some comments below =
....</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- James.</FONT></DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt;=20
wrote in message <A=20
=
href=3D"news:eolgq1$q5s$1@utils.eclipse.org">news:eolgq1$q5s$1@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi James</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Thanks for your =
answer!</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Fortunately the problem =
disappeared=20
unexpectedly when I changed the Aggregation attribute of the =
memberEnd=20
to "none". After switching back to 'composite' everything =
remained=20
ok...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>But after some experiences with =
the=20
Genmodel there are some further questions:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- For every implementation =
class there is a=20
<FONT size=3D2>getCacheAdapter() method generated (in the=20
uml2.uml-impl-classes there is no such method) - this leads to a =
stack=20
overflow using the generated editor plugin. Deleting the method =
manually=20
resolves the problem. What am I doing wrong with the=20
UML-Model/Genmodel?</FONT></FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ Not sure about this one&nbsp; =
I'll have=20
to investigate and get back to you.&nbsp; </FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- In the UML.merged.uml-model =
used as=20
metamodel, what is the difference beween memberEnd Property with =

&lt;eReference&gt; Stereotype and the ones without=20
Stereotype?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ In general the Ecore =
stereotype provides=20
information when converting from UMLto Ecore.&nbsp; Specific =
code=20
generation details might be necessary for those items with =
stereotypes=20
applied.&nbsp; Sometimes information is provided in the =
stereotype to=20
allow conversion from Ecore to UML and back to Ecore and =
preserve the=20
integrity of the original .ecore model ( ie. round tripping=20
).</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- Using an UML-model as =
metamodel is there=20
a special behaviour in Genmodel when using ownedEnd's in =
associations=20
(as in A_general_generalization)?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ I'm not sure what you mean =
about "special=20
behavior".&nbsp;&nbsp; I'm not aware of unique behavior for =
this.&nbsp;=20
I'll discuss it with Kenn and get back to you if there is =
something=20
unique here.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>And finally a question which is =
maybe not=20
just related with uml2:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Getting all the sources from =
CVS (emf:=20
200701110200 uml2: 200701111331) on 3.3M4 there are several =
errors in=20
the UML2 projects even after updating the classpathes =
(org.eclipse.uml2,=20
.uml2.importer.rose, . Is that just a strange behaviour of my=20
environment?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ I'm not seeing any build =
issues for the=20
uml2 project.&nbsp; I do however have some issues with=20
org.eclipse.emf.importer.rose.&nbsp;&nbsp; .. may not be the =
same issue=20
you are having but these problems are a result of .jars needed =
for=20
ant.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Sometimes there are .readme's =
that explain=20
how to resolve these issues.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>eg.&nbsp;&nbsp; <FONT =
size=3D3><FONT=20
face=3D"Times New Roman">Before running your build scripts or =
the examples=20
provided here, make sure you have the 2 jar files required by =
this=20
plugin ("importer.rose.jar" and "importer.rose.tasks.jar") and =
also the=20
2 jar files required by the =
</FONT><TT>org.eclipse.emf.ant</TT><FONT=20
face=3D"Times New Roman"> plugin ("emf.ant.jar" and =
"emf.ant.tasks.jar").=20
If necessary, Eclipse can generate them for you. All you need to =
do is=20
to run the Ant script created by PDE when you right click on the =

manifest file and select "PDE Tools&gt;Create Ant Build File".=20
</FONT></FONT></FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>but again, I don't see any =
issues with=20
uml.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Greets, Andreas</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: =
5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"James Bruck" &lt;<A=20
href=3D"mailto:jbruck@ca.ibm.com">jbruck@ca.ibm.com</A>&gt; =
wrote in=20
message <A=20
=
href=3D"news:eo93md$g96$2@utils.eclipse.org">news:eo93md$g96$2@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>I just tried this on my =
machine with a=20
contrived example and it seemed to work fine ( If I understood =
this=20
correctly ;)&nbsp; ).</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Would it be possible for you =
to send more=20
information or a snippet of your metamodel so I can inspect=20
it.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>What errors are you seeing in =
the impl=20
class?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Regards,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- James.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: =
5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt;=20
wrote in message <A=20
=
href=3D"news:eo7rou$ob0$1@utils.eclipse.org">news:eo7rou$ob0$1@utils.ecli=
pse.org</A>...</DIV>
<DIV>Hi all</DIV>
<DIV>I'm trying to make a new language metamodel on the =
basis on=20
UML.uml and using UML-Genmodel. In principle, the process is =
working=20
but with the following aspect I get into trouble:</DIV>
<DIV>&nbsp;</DIV>
<DIV>- In my metamodel I have two classes with an 1 -- 0..*=20
association</DIV>
<DIV>- the creation of the genmodel works well</DIV>
<DIV>- after model-code generation, there are errors in the=20
impl-class of the 0..* -side: in the following code-snipped =
the=20
varable ownedrules is not declared in the class</DIV>
<DIV>- if I change the Multiplicity to 1, the resulting code =

works</DIV>
<DIV><FONT face=3DArial size=3D2>
<P align=3Dleft><B><FONT color=3D#7f0055 =
size=3D2>public</B></FONT><FONT=20
size=3D2> EList getOwnedrules() {</P>
<P align=3Dleft></FONT><B><FONT color=3D#7f0055=20
size=3D2>if</B></FONT><FONT size=3D2> (ownedrules =3D=3D =
</FONT><B><FONT=20
color=3D#7f0055 size=3D2>null</B></FONT><FONT size=3D2>) =
{</P>
<P align=3Dleft>ownedrules =3D </FONT><B><FONT =
color=3D#7f0055=20
size=3D2>new</B></FONT><FONT size=3D2>=20
EObjectWithInverseResolvingEList(</P>
<P align=3Dleft>MappingRule.</FONT><B><FONT color=3D#7f0055=20
size=3D2>class</B></FONT><FONT size=3D2>, </FONT><B><FONT =
color=3D#7f0055=20
size=3D2>this</B></FONT><FONT size=3D2>,</P>
<P align=3Dleft>UMLTPackage.</FONT><I><FONT color=3D#0000c0=20
size=3D2>TRANSFORMATION_GROUP__OWNEDRULES</I></FONT><FONT =
size=3D2>,</P>
<P align=3Dleft>UMLTPackage.</FONT><I><FONT color=3D#0000c0=20
size=3D2>MAPPING_RULE__TGROUP</I></FONT><FONT =
size=3D2>);</P>
<P align=3Dleft>}</P>
<P align=3Dleft></FONT><B><FONT color=3D#7f0055=20
size=3D2>return</B></FONT><FONT size=3D2> ownedrules;</P>
<P align=3Dleft>}</FONT></FONT></P></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV>In the attachments there are some pictures of the .uml =
and=20
.genmodel to get the context.</DIV>
<DIV>&nbsp;</DIV>
<DIV>Anyone an idea</DIV>
<DIV>Regards, Andreas</DIV>
<DIV><FONT face=3DArial=20
=
size=3D2></FONT>&nbsp;</DIV></BLOCKQUOTE></BLOCKQUOTE ></BLOCKQUOTE></BLOC=
KQUOTE></BLOCKQUOTE></BLOCKQUOTE></BODY></HTML>

------=_NextPart_000_07E8_01C73BCA.837935E0--
Re: Problem using UML Genmodel [message #468946 is a reply to message #468690] Fri, 26 January 2007 10:15 Go to previous messageGo to next message
Andreas Morf is currently offline Andreas MorfFriend
Messages: 14
Registered: July 2009
Junior Member
After refactoring my metamodel and deleting all of the generated code, the
same problem with the getCacheAdapter appears again:
My metaclass inherits from uml2.uml.Model and therefore the root class
should be ElementImpl but somehow the generator isn't aware of that and
gerates the following method leading to the StackOverflow (pingpong):

protected CacheAdapter getCacheAdapter() {
return CacheAdapter.getCacheAdapter(this);
}

So changing this method to the following, the code works - but in my
opinion, this should not be the preferred way:

protected CacheAdapter getCacheAdapter() {
return super.getCacheAdapter();
}

Why is this generated method different from that in the ElementImpl-Class
Also, there is to say that I Model/genmodel importing/extending Metaclasses
from UML.merged.uml

Any ideas out there?
Greetings, Andreas
Re: Problem using UML Genmodel [message #468955 is a reply to message #468946] Fri, 26 January 2007 19:46 Go to previous message
Kenn Hussey is currently offline Kenn HusseyFriend
Messages: 1620
Registered: July 2009
Senior Member
Andreas,

This is a problem only because you are extending UML2. If you are extending
UML2 and want to use the same cache adapter, overriding the method as you
have is the right choice (the other alternative is to disable cache adapter
support for your generator model). However, we should certainly avoid the
stack overflow for cases where a different cache adapter is used (the
default) - please open a Bugzilla report and we'll have this fixed.

Kenn

"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message
news:epckbf$pal$1@utils.eclipse.org...
> After refactoring my metamodel and deleting all of the generated code, the
> same problem with the getCacheAdapter appears again:
> My metaclass inherits from uml2.uml.Model and therefore the root class
> should be ElementImpl but somehow the generator isn't aware of that and
> gerates the following method leading to the StackOverflow (pingpong):
>
> protected CacheAdapter getCacheAdapter() {
> return CacheAdapter.getCacheAdapter(this);
> }
>
> So changing this method to the following, the code works - but in my
> opinion, this should not be the preferred way:
>
> protected CacheAdapter getCacheAdapter() {
> return super.getCacheAdapter();
> }
>
> Why is this generated method different from that in the ElementImpl-Class
> Also, there is to say that I Model/genmodel importing/extending
> Metaclasses from UML.merged.uml
>
> Any ideas out there?
> Greetings, Andreas
>
Re: Problem using UML Genmodel [message #572451 is a reply to message #468510] Fri, 19 January 2007 02:34 Go to previous message
james bruck is currently offline james bruckFriend
Messages: 1724
Registered: July 2009
Senior Member
This is a multi-part message in MIME format.

------=_NextPart_000_0080_01C73B48.610BDD00
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Hi Andreas,

some comments below ...

- James.
"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eolgq1$q5s$1@utils.eclipse.org...
Hi James
Thanks for your answer!
Fortunately the problem disappeared unexpectedly when I changed the =
Aggregation attribute of the memberEnd to "none". After switching back =
to 'composite' everything remained ok...

But after some experiences with the Genmodel there are some further =
questions:

- For every implementation class there is a getCacheAdapter() method =
generated (in the uml2.uml-impl-classes there is no such method) - this =
leads to a stack overflow using the generated editor plugin. Deleting =
the method manually resolves the problem. What am I doing wrong with the =
UML-Model/Genmodel?

+ Not sure about this one I'll have to investigate and get back to =
you. =20

- In the UML.merged.uml-model used as metamodel, what is the =
difference beween memberEnd Property with <eReference> Stereotype and =
the ones without Stereotype?

+ In general the Ecore stereotype provides information when converting =
from UMLto Ecore. Specific code generation details might be necessary =
for those items with stereotypes applied. Sometimes information is =
provided in the stereotype to allow conversion from Ecore to UML and =
back to Ecore and preserve the integrity of the original .ecore model ( =
ie. round tripping ).

- Using an UML-model as metamodel is there a special behaviour in =
Genmodel when using ownedEnd's in associations (as in =
A_general_generalization)?

+ I'm not sure what you mean about "special behavior". I'm not aware =
of unique behavior for this. I'll discuss it with Kenn and get back to =
you if there is something unique here.

And finally a question which is maybe not just related with uml2:
Getting all the sources from CVS (emf: 200701110200 uml2: =
200701111331) on 3.3M4 there are several errors in the UML2 projects =
even after updating the classpathes (org.eclipse.uml2, =
..uml2.importer.rose, . Is that just a strange behaviour of my =
environment?

+ I'm not seeing any build issues for the uml2 project. I do however =
have some issues with org.eclipse.emf.importer.rose. .. may not be the =
same issue you are having but these problems are a result of .jars =
needed for ant.
Sometimes there are .readme's that explain how to resolve these =
issues.

eg. Before running your build scripts or the examples provided here, =
make sure you have the 2 jar files required by this plugin =
("importer.rose.jar" and "importer.rose.tasks.jar") and also the 2 jar =
files required by the org.eclipse.emf.ant plugin ("emf.ant.jar" and =
"emf.ant.tasks.jar"). If necessary, Eclipse can generate them for you. =
All you need to do is to run the Ant script created by PDE when you =
right click on the manifest file and select "PDE Tools>Create Ant Build =
File".=20

but again, I don't see any issues with uml.


Greets, Andreas


"James Bruck" <jbruck@ca.ibm.com> wrote in message =
news:eo93md$g96$2@utils.eclipse.org...
Hi Andreas,

I just tried this on my machine with a contrived example and it =
seemed to work fine ( If I understood this correctly ;) ).
Would it be possible for you to send more information or a snippet =
of your metamodel so I can inspect it.
What errors are you seeing in the impl class?

Regards,

- James.

"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eo7rou$ob0$1@utils.eclipse.org...
Hi all
I'm trying to make a new language metamodel on the basis on =
UML.uml and using UML-Genmodel. In principle, the process is working but =
with the following aspect I get into trouble:

- In my metamodel I have two classes with an 1 -- 0..* association
- the creation of the genmodel works well
- after model-code generation, there are errors in the impl-class =
of the 0..* -side: in the following code-snipped the varable ownedrules =
is not declared in the class
- if I change the Multiplicity to 1, the resulting code works
public EList getOwnedrules() {

if (ownedrules =3D=3D null) {

ownedrules =3D new EObjectWithInverseResolvingEList(

MappingRule.class, this,

UMLTPackage.TRANSFORMATION_GROUP__OWNEDRULES,

UMLTPackage.MAPPING_RULE__TGROUP);

}

return ownedrules;

}


In the attachments there are some pictures of the .uml and =
..genmodel to get the context.

Anyone an idea
Regards, Andreas

------=_NextPart_000_0080_01C73B48.610BDD00
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 6.00.2900.2995" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>Hi Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>some comments below ...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- James.</FONT></DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt; =
wrote in=20
message <A=20
=
href=3D"news:eolgq1$q5s$1@utils.eclipse.org">news:eolgq1$q5s$1@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi James</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Thanks for your answer!</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Fortunately the problem disappeared =
unexpectedly=20
when I changed the Aggregation attribute of the memberEnd to "none". =
After=20
switching back to 'composite' everything remained ok...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>But after some experiences with the =
Genmodel=20
there are some further questions:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- For every implementation class =
there is a <FONT=20
size=3D2>getCacheAdapter() method generated (in the =
uml2.uml-impl-classes there=20
is no such method) - this leads to a stack overflow using the =
generated editor=20
plugin. Deleting the method manually resolves the problem. What am I =
doing=20
wrong with the UML-Model/Genmodel?</FONT></FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ Not sure about this one&nbsp; I'll =
have to=20
investigate and get back to you.&nbsp; </FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- In the UML.merged.uml-model used as =
metamodel,=20
what is the difference beween memberEnd Property with =
&lt;eReference&gt;=20
Stereotype and the ones without Stereotype?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ In general the Ecore stereotype =
provides=20
information when converting from UMLto Ecore.&nbsp; Specific code =
generation=20
details might be necessary for those items with stereotypes =
applied.&nbsp;=20
Sometimes information is provided in the stereotype to allow =
conversion from=20
Ecore to UML and back to Ecore and preserve the integrity of the =
original=20
.ecore model ( ie. round tripping ).</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- Using an UML-model as metamodel is =
there a=20
special behaviour in Genmodel when using ownedEnd's in associations =
(as in=20
A_general_generalization)?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ I'm not sure what you mean about =
"special=20
behavior".&nbsp;&nbsp; I'm not aware of unique behavior for =
this.&nbsp; I'll=20
discuss it with Kenn and get back to you if there is something unique=20
here.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>And finally a question which is maybe =
not just=20
related with uml2:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Getting all the sources from CVS =
(emf:=20
200701110200 uml2: 200701111331) on 3.3M4 there are several errors in =
the UML2=20
projects even after updating the classpathes (org.eclipse.uml2,=20
.uml2.importer.rose, . Is that just a strange behaviour of my=20
environment?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ I'm not seeing any build issues for =
the uml2=20
project.&nbsp; I do however have some issues with=20
org.eclipse.emf.importer.rose.&nbsp;&nbsp; .. may not be the same =
issue you=20
are having but these problems are a result of .jars needed for=20
ant.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Sometimes there are .readme's that =
explain how to=20
resolve these issues.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>eg.&nbsp;&nbsp; <FONT size=3D3><FONT=20
face=3D"Times New Roman">Before running your build scripts or the =
examples=20
provided here, make sure you have the 2 jar files required by this =
plugin=20
("importer.rose.jar" and "importer.rose.tasks.jar") and also the 2 jar =
files=20
required by the </FONT><TT>org.eclipse.emf.ant</TT><FONT=20
face=3D"Times New Roman"> plugin ("emf.ant.jar" and =
"emf.ant.tasks.jar"). If=20
necessary, Eclipse can generate them for you. All you need to do is to =
run the=20
Ant script created by PDE when you right click on the manifest file =
and select=20
"PDE Tools&gt;Create Ant Build File". </FONT></FONT></FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>but again, I don't see any issues =
with=20
uml.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Greets, Andreas</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"James Bruck" &lt;<A=20
href=3D"mailto:jbruck@ca.ibm.com">jbruck@ca.ibm.com</A>&gt; wrote in =
message=20
<A=20
=
href=3D"news:eo93md$g96$2@utils.eclipse.org">news:eo93md$g96$2@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>I just tried this on my machine =
with a=20
contrived example and it seemed to work fine ( If I understood this=20
correctly ;)&nbsp; ).</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Would it be possible for you to =
send more=20
information or a snippet of your metamodel so I can inspect =
it.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>What errors are you seeing in the =
impl=20
class?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Regards,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- James.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt;=20
wrote in message <A=20
=
href=3D"news:eo7rou$ob0$1@utils.eclipse.org">news:eo7rou$ob0$1@utils.ecli=
pse.org</A>...</DIV>
<DIV>Hi all</DIV>
<DIV>I'm trying to make a new language metamodel on the basis on =
UML.uml=20
and using UML-Genmodel. In principle, the process is working but =
with the=20
following aspect I get into trouble:</DIV>
<DIV>&nbsp;</DIV>
<DIV>- In my metamodel I have two classes with an 1 -- 0..*=20
association</DIV>
<DIV>- the creation of the genmodel works well</DIV>
<DIV>- after model-code generation, there are errors in the =
impl-class of=20
the 0..* -side: in the following code-snipped the varable =
ownedrules is=20
not declared in the class</DIV>
<DIV>- if I change the Multiplicity to 1, the resulting code =
works</DIV>
<DIV><FONT face=3DArial size=3D2>
<P align=3Dleft><B><FONT color=3D#7f0055 =
size=3D2>public</B></FONT><FONT size=3D2>=20
EList getOwnedrules() {</P>
<P align=3Dleft></FONT><B><FONT color=3D#7f0055 =
size=3D2>if</B></FONT><FONT=20
size=3D2> (ownedrules =3D=3D </FONT><B><FONT color=3D#7f0055=20
size=3D2>null</B></FONT><FONT size=3D2>) {</P>
<P align=3Dleft>ownedrules =3D </FONT><B><FONT color=3D#7f0055=20
size=3D2>new</B></FONT><FONT size=3D2> =
EObjectWithInverseResolvingEList(</P>
<P align=3Dleft>MappingRule.</FONT><B><FONT color=3D#7f0055=20
size=3D2>class</B></FONT><FONT size=3D2>, </FONT><B><FONT =
color=3D#7f0055=20
size=3D2>this</B></FONT><FONT size=3D2>,</P>
<P align=3Dleft>UMLTPackage.</FONT><I><FONT color=3D#0000c0=20
size=3D2>TRANSFORMATION_GROUP__OWNEDRULES</I></FONT><FONT =
size=3D2>,</P>
<P align=3Dleft>UMLTPackage.</FONT><I><FONT color=3D#0000c0=20
size=3D2>MAPPING_RULE__TGROUP</I></FONT><FONT size=3D2>);</P>
<P align=3Dleft>}</P>
<P align=3Dleft></FONT><B><FONT color=3D#7f0055 =
size=3D2>return</B></FONT><FONT=20
size=3D2> ownedrules;</P>
<P align=3Dleft>}</FONT></FONT></P></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV>In the attachments there are some pictures of the .uml and =
..genmodel=20
to get the context.</DIV>
<DIV>&nbsp;</DIV>
<DIV>Anyone an idea</DIV>
<DIV>Regards, Andreas</DIV>
<DIV><FONT face=3DArial=20
size=3D2></FONT>&nbsp;</DIV></BLOCKQUOTE></BLOCKQUOTE ></BLOCKQUOTE></BODY=
></HTML>

------=_NextPart_000_0080_01C73B48.610BDD00--
Re: Problem using UML Genmodel [message #572487 is a reply to message #468510] Fri, 19 January 2007 03:01 Go to previous message
Kenn Hussey is currently offline Kenn HusseyFriend
Messages: 1620
Registered: July 2009
Senior Member
This is a multi-part message in MIME format.

------=_NextPart_000_078A_01C73B4C.33147340
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Andreas,

Regarding the errors, please ensure that you only have the following =
projects checked out (from HEAD):

org.eclipse.uml2-feature
org.eclipse.uml2-plugin
org.eclipse.uml2.codegen.ecore
org.eclipse.uml2.codegen.ecore.ui
org.eclipse.uml2.common
org.eclipse.uml2.common.edit
org.eclipse.uml2.doc-feature
org.eclipse.uml2.doc-plugin
org.eclipse.uml2.examples-feature
org.eclipse.uml2.examples-plugin
org.eclipse.uml2.examples.uml.ui
org.eclipse.uml2.sdk-feature
org.eclipse.uml2.tests-feature
org.eclipse.uml2.tests-plugin
org.eclipse.uml2.uml
org.eclipse.uml2.uml.ecore.exporter
org.eclipse.uml2.uml.ecore.importer
org.eclipse.uml2.uml.edit
org.eclipse.uml2.uml.editor
org.eclipse.uml2.uml.resources
org.eclipse.uml2.uml.tests

Kenn

"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eolgq1$q5s$1@utils.eclipse.org...
Hi James
Thanks for your answer!
Fortunately the problem disappeared unexpectedly when I changed the =
Aggregation attribute of the memberEnd to "none". After switching back =
to 'composite' everything remained ok...

But after some experiences with the Genmodel there are some further =
questions:

- For every implementation class there is a getCacheAdapter() method =
generated (in the uml2.uml-impl-classes there is no such method) - this =
leads to a stack overflow using the generated editor plugin. Deleting =
the method manually resolves the problem. What am I doing wrong with the =
UML-Model/Genmodel?
- In the UML.merged.uml-model used as metamodel, what is the =
difference beween memberEnd Property with <eReference> Stereotype and =
the ones without Stereotype?
- Using an UML-model as metamodel is there a special behaviour in =
Genmodel when using ownedEnd's in associations (as in =
A_general_generalization)?

And finally a question which is maybe not just related with uml2:
Getting all the sources from CVS (emf: 200701110200 uml2: =
200701111331) on 3.3M4 there are several errors in the UML2 projects =
even after updating the classpathes (org.eclipse.uml2, =
..uml2.importer.rose, . Is that just a strange behaviour of my =
environment?

Greets, Andreas


"James Bruck" <jbruck@ca.ibm.com> wrote in message =
news:eo93md$g96$2@utils.eclipse.org...
Hi Andreas,

I just tried this on my machine with a contrived example and it =
seemed to work fine ( If I understood this correctly ;) ).
Would it be possible for you to send more information or a snippet =
of your metamodel so I can inspect it.
What errors are you seeing in the impl class?

Regards,

- James.

"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eo7rou$ob0$1@utils.eclipse.org...
Hi all
I'm trying to make a new language metamodel on the basis on =
UML.uml and using UML-Genmodel. In principle, the process is working but =
with the following aspect I get into trouble:

- In my metamodel I have two classes with an 1 -- 0..* association
- the creation of the genmodel works well
- after model-code generation, there are errors in the impl-class =
of the 0..* -side: in the following code-snipped the varable ownedrules =
is not declared in the class
- if I change the Multiplicity to 1, the resulting code works
public EList getOwnedrules() {

if (ownedrules =3D=3D null) {

ownedrules =3D new EObjectWithInverseResolvingEList(

MappingRule.class, this,

UMLTPackage.TRANSFORMATION_GROUP__OWNEDRULES,

UMLTPackage.MAPPING_RULE__TGROUP);

}

return ownedrules;

}


In the attachments there are some pictures of the .uml and =
..genmodel to get the context.

Anyone an idea
Regards, Andreas

------=_NextPart_000_078A_01C73B4C.33147340
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 6.00.2900.3020" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Regarding the errors, please ensure =
that you only=20
have the following projects checked out (from HEAD):</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV>org.eclipse.uml2-feature<BR>org.eclipse.uml2-plugin<BR >org.eclipse.u=
ml2.codegen.ecore<BR>org.eclipse.uml2.codegen.ecore.ui<BR >org.eclipse.uml=
2.common<BR>org.eclipse.uml2.common.edit<BR>org.eclipse.uml2.doc-feature <=
BR>org.eclipse.uml2.doc-plugin<BR>org.eclipse.uml2.examples-feature <BR>or=
g.eclipse.uml2.examples-plugin<BR>org.eclipse.uml2.examples.uml.ui <BR>org=
..eclipse.uml2.sdk-feature<BR>org.eclipse.uml2.tests-feature <BR>org.eclips=
e.uml2.tests-plugin<BR>org.eclipse.uml2.uml<BR>org.eclipse.uml2.uml.ecore=
..exporter<BR>org.eclipse.uml2.uml.ecore.importer<BR>org.eclipse.uml2.uml.=
edit<BR>org.eclipse.uml2.uml.editor<BR>org.eclipse.uml2.uml.resources <BR>=
org.eclipse.uml2.uml.tests<BR></DIV>
<DIV><FONT face=3DArial size=3D2>Kenn</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt; =
wrote in=20
message <A=20
=
href=3D"news:eolgq1$q5s$1@utils.eclipse.org">news:eolgq1$q5s$1@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi James</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Thanks for your answer!</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Fortunately the problem disappeared =
unexpectedly=20
when I changed the Aggregation attribute of the memberEnd to "none". =
After=20
switching back to 'composite' everything remained ok...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>But after some experiences with the =
Genmodel=20
there are some further questions:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- For every implementation class =
there is a <FONT=20
size=3D2>getCacheAdapter() method generated (in the =
uml2.uml-impl-classes there=20
is no such method) - this leads to a stack overflow using the =
generated editor=20
plugin. Deleting the method manually resolves the problem. What am I =
doing=20
wrong with the UML-Model/Genmodel?</FONT></FONT></DIV>
<DIV><FONT face=3DArial size=3D2>- In the UML.merged.uml-model used as =
metamodel,=20
what is the difference beween memberEnd Property with =
&lt;eReference&gt;=20
Stereotype and the ones without Stereotype?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>- Using an UML-model as metamodel is =
there a=20
special behaviour in Genmodel when using ownedEnd's in associations =
(as in=20
A_general_generalization)?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>And finally a question which is maybe =
not just=20
related with uml2:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Getting all the sources from CVS =
(emf:=20
200701110200 uml2: 200701111331) on 3.3M4 there are several errors in =
the UML2=20
projects even after updating the classpathes (org.eclipse.uml2,=20
.uml2.importer.rose, . Is that just a strange behaviour of my=20
environment?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Greets, Andreas</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"James Bruck" &lt;<A=20
href=3D"mailto:jbruck@ca.ibm.com">jbruck@ca.ibm.com</A>&gt; wrote in =
message=20
<A=20
=
href=3D"news:eo93md$g96$2@utils.eclipse.org">news:eo93md$g96$2@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>I just tried this on my machine =
with a=20
contrived example and it seemed to work fine ( If I understood this=20
correctly ;)&nbsp; ).</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Would it be possible for you to =
send more=20
information or a snippet of your metamodel so I can inspect =
it.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>What errors are you seeing in the =
impl=20
class?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Regards,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- James.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt;=20
wrote in message <A=20
=
href=3D"news:eo7rou$ob0$1@utils.eclipse.org">news:eo7rou$ob0$1@utils.ecli=
pse.org</A>...</DIV>
<DIV>Hi all</DIV>
<DIV>I'm trying to make a new language metamodel on the basis on =
UML.uml=20
and using UML-Genmodel. In principle, the process is working but =
with the=20
following aspect I get into trouble:</DIV>
<DIV>&nbsp;</DIV>
<DIV>- In my metamodel I have two classes with an 1 -- 0..*=20
association</DIV>
<DIV>- the creation of the genmodel works well</DIV>
<DIV>- after model-code generation, there are errors in the =
impl-class of=20
the 0..* -side: in the following code-snipped the varable =
ownedrules is=20
not declared in the class</DIV>
<DIV>- if I change the Multiplicity to 1, the resulting code =
works</DIV>
<DIV><FONT face=3DArial size=3D2>
<P align=3Dleft><B><FONT color=3D#7f0055 =
size=3D2>public</B></FONT><FONT size=3D2>=20
EList getOwnedrules() {</P>
<P align=3Dleft></FONT><B><FONT color=3D#7f0055 =
size=3D2>if</B></FONT><FONT=20
size=3D2> (ownedrules =3D=3D </FONT><B><FONT color=3D#7f0055=20
size=3D2>null</B></FONT><FONT size=3D2>) {</P>
<P align=3Dleft>ownedrules =3D </FONT><B><FONT color=3D#7f0055=20
size=3D2>new</B></FONT><FONT size=3D2> =
EObjectWithInverseResolvingEList(</P>
<P align=3Dleft>MappingRule.</FONT><B><FONT color=3D#7f0055=20
size=3D2>class</B></FONT><FONT size=3D2>, </FONT><B><FONT =
color=3D#7f0055=20
size=3D2>this</B></FONT><FONT size=3D2>,</P>
<P align=3Dleft>UMLTPackage.</FONT><I><FONT color=3D#0000c0=20
size=3D2>TRANSFORMATION_GROUP__OWNEDRULES</I></FONT><FONT =
size=3D2>,</P>
<P align=3Dleft>UMLTPackage.</FONT><I><FONT color=3D#0000c0=20
size=3D2>MAPPING_RULE__TGROUP</I></FONT><FONT size=3D2>);</P>
<P align=3Dleft>}</P>
<P align=3Dleft></FONT><B><FONT color=3D#7f0055 =
size=3D2>return</B></FONT><FONT=20
size=3D2> ownedrules;</P>
<P align=3Dleft>}</FONT></FONT></P></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV>In the attachments there are some pictures of the .uml and =
..genmodel=20
to get the context.</DIV>
<DIV>&nbsp;</DIV>
<DIV>Anyone an idea</DIV>
<DIV>Regards, Andreas</DIV>
<DIV><FONT face=3DArial=20
size=3D2></FONT>&nbsp;</DIV></BLOCKQUOTE></BLOCKQUOTE ></BLOCKQUOTE></BODY=
></HTML>

------=_NextPart_000_078A_01C73B4C.33147340--
Re: Problem using UML Genmodel [message #572540 is a reply to message #468524] Fri, 19 January 2007 03:10 Go to previous message
Kenn Hussey is currently offline Kenn HusseyFriend
Messages: 1620
Registered: July 2009
Senior Member
This is a multi-part message in MIME format.

------=_NextPart_000_079D_01C73B4D.6CDA39B0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

- Could you provide more details on the stack overflow you are seeing? =
Every "model root" class will have this method in it, but it's not clear =
to me why this would result in a stack overflow...

- Association ends that are owned by the assocation are ignored when =
converting UML models to Ecore.

Kenn
"James Bruck" <jbruck@ca.ibm.com> wrote in message =
news:eopai8$174$1@utils.eclipse.org...
Hi Andreas,

some comments below ...

- James.
"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eolgq1$q5s$1@utils.eclipse.org...
Hi James
Thanks for your answer!
Fortunately the problem disappeared unexpectedly when I changed the =
Aggregation attribute of the memberEnd to "none". After switching back =
to 'composite' everything remained ok...

But after some experiences with the Genmodel there are some further =
questions:

- For every implementation class there is a getCacheAdapter() method =
generated (in the uml2.uml-impl-classes there is no such method) - this =
leads to a stack overflow using the generated editor plugin. Deleting =
the method manually resolves the problem. What am I doing wrong with the =
UML-Model/Genmodel?

+ Not sure about this one I'll have to investigate and get back to =
you. =20

- In the UML.merged.uml-model used as metamodel, what is the =
difference beween memberEnd Property with <eReference> Stereotype and =
the ones without Stereotype?

+ In general the Ecore stereotype provides information when =
converting from UMLto Ecore. Specific code generation details might be =
necessary for those items with stereotypes applied. Sometimes =
information is provided in the stereotype to allow conversion from Ecore =
to UML and back to Ecore and preserve the integrity of the original =
..ecore model ( ie. round tripping ).

- Using an UML-model as metamodel is there a special behaviour in =
Genmodel when using ownedEnd's in associations (as in =
A_general_generalization)?

+ I'm not sure what you mean about "special behavior". I'm not =
aware of unique behavior for this. I'll discuss it with Kenn and get =
back to you if there is something unique here.

And finally a question which is maybe not just related with uml2:
Getting all the sources from CVS (emf: 200701110200 uml2: =
200701111331) on 3.3M4 there are several errors in the UML2 projects =
even after updating the classpathes (org.eclipse.uml2, =
..uml2.importer.rose, . Is that just a strange behaviour of my =
environment?

+ I'm not seeing any build issues for the uml2 project. I do =
however have some issues with org.eclipse.emf.importer.rose. .. may =
not be the same issue you are having but these problems are a result of =
..jars needed for ant.
Sometimes there are .readme's that explain how to resolve these =
issues.

eg. Before running your build scripts or the examples provided =
here, make sure you have the 2 jar files required by this plugin =
("importer.rose.jar" and "importer.rose.tasks.jar") and also the 2 jar =
files required by the org.eclipse.emf.ant plugin ("emf.ant.jar" and =
"emf.ant.tasks.jar"). If necessary, Eclipse can generate them for you. =
All you need to do is to run the Ant script created by PDE when you =
right click on the manifest file and select "PDE Tools>Create Ant Build =
File".=20

but again, I don't see any issues with uml.


Greets, Andreas


"James Bruck" <jbruck@ca.ibm.com> wrote in message =
news:eo93md$g96$2@utils.eclipse.org...
Hi Andreas,

I just tried this on my machine with a contrived example and it =
seemed to work fine ( If I understood this correctly ;) ).
Would it be possible for you to send more information or a snippet =
of your metamodel so I can inspect it.
What errors are you seeing in the impl class?

Regards,

- James.

"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eo7rou$ob0$1@utils.eclipse.org...
Hi all
I'm trying to make a new language metamodel on the basis on =
UML.uml and using UML-Genmodel. In principle, the process is working but =
with the following aspect I get into trouble:

- In my metamodel I have two classes with an 1 -- 0..* =
association
- the creation of the genmodel works well
- after model-code generation, there are errors in the =
impl-class of the 0..* -side: in the following code-snipped the varable =
ownedrules is not declared in the class
- if I change the Multiplicity to 1, the resulting code works
public EList getOwnedrules() {

if (ownedrules =3D=3D null) {

ownedrules =3D new EObjectWithInverseResolvingEList(

MappingRule.class, this,

UMLTPackage.TRANSFORMATION_GROUP__OWNEDRULES,

UMLTPackage.MAPPING_RULE__TGROUP);

}

return ownedrules;

}


In the attachments there are some pictures of the .uml and =
..genmodel to get the context.

Anyone an idea
Regards, Andreas

------=_NextPart_000_079D_01C73B4D.6CDA39B0
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 6.00.2900.3020" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>- Could you provide more details on the =
stack=20
overflow you are seeing? Every "model root" class will have this method =
in it,=20
but it's not clear to me why this would result in a stack=20
overflow...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- Association ends that are owned by =
the assocation=20
are ignored when converting UML models to Ecore.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Kenn</FONT></DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"James Bruck" &lt;<A=20
href=3D"mailto:jbruck@ca.ibm.com">jbruck@ca.ibm.com</A>&gt; wrote in =
message <A=20
=
href=3D"news:eopai8$174$1@utils.eclipse.org">news:eopai8$174$1@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>some comments below ...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- James.</FONT></DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt; =
wrote=20
in message <A=20
=
href=3D"news:eolgq1$q5s$1@utils.eclipse.org">news:eolgq1$q5s$1@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi James</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Thanks for your =
answer!</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Fortunately the problem disappeared =

unexpectedly when I changed the Aggregation attribute of the =
memberEnd to=20
"none". After switching back to 'composite' everything remained=20
ok...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>But after some experiences with the =
Genmodel=20
there are some further questions:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- For every implementation class =
there is a=20
<FONT size=3D2>getCacheAdapter() method generated (in the=20
uml2.uml-impl-classes there is no such method) - this leads to a =
stack=20
overflow using the generated editor plugin. Deleting the method =
manually=20
resolves the problem. What am I doing wrong with the=20
UML-Model/Genmodel?</FONT></FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ Not sure about this one&nbsp; =
I'll have to=20
investigate and get back to you.&nbsp; </FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- In the UML.merged.uml-model used =
as=20
metamodel, what is the difference beween memberEnd Property with=20
&lt;eReference&gt; Stereotype and the ones without =
Stereotype?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ In general the Ecore stereotype =
provides=20
information when converting from UMLto Ecore.&nbsp; Specific code =
generation=20
details might be necessary for those items with stereotypes =
applied.&nbsp;=20
Sometimes information is provided in the stereotype to allow =
conversion from=20
Ecore to UML and back to Ecore and preserve the integrity of the =
original=20
.ecore model ( ie. round tripping ).</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- Using an UML-model as metamodel =
is there a=20
special behaviour in Genmodel when using ownedEnd's in associations =
(as in=20
A_general_generalization)?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ I'm not sure what you mean about =
"special=20
behavior".&nbsp;&nbsp; I'm not aware of unique behavior for =
this.&nbsp; I'll=20
discuss it with Kenn and get back to you if there is something =
unique=20
here.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>And finally a question which is =
maybe not just=20
related with uml2:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Getting all the sources from CVS =
(emf:=20
200701110200 uml2: 200701111331) on 3.3M4 there are several errors =
in the=20
UML2 projects even after updating the classpathes (org.eclipse.uml2, =

.uml2.importer.rose, . Is that just a strange behaviour of my=20
environment?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ I'm not seeing any build issues =
for the uml2=20
project.&nbsp; I do however have some issues with=20
org.eclipse.emf.importer.rose.&nbsp;&nbsp; .. may not be the same =
issue you=20
are having but these problems are a result of .jars needed for=20
ant.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Sometimes there are .readme's that =
explain how=20
to resolve these issues.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>eg.&nbsp;&nbsp; <FONT =
size=3D3><FONT=20
face=3D"Times New Roman">Before running your build scripts or the =
examples=20
provided here, make sure you have the 2 jar files required by this =
plugin=20
("importer.rose.jar" and "importer.rose.tasks.jar") and also the 2 =
jar files=20
required by the </FONT><TT>org.eclipse.emf.ant</TT><FONT=20
face=3D"Times New Roman"> plugin ("emf.ant.jar" and =
"emf.ant.tasks.jar"). If=20
necessary, Eclipse can generate them for you. All you need to do is =
to run=20
the Ant script created by PDE when you right click on the manifest =
file and=20
select "PDE Tools&gt;Create Ant Build File". =
</FONT></FONT></FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>but again, I don't see any issues =
with=20
uml.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Greets, Andreas</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"James Bruck" &lt;<A=20
href=3D"mailto:jbruck@ca.ibm.com">jbruck@ca.ibm.com</A>&gt; wrote =
in message=20
<A=20
=
href=3D"news:eo93md$g96$2@utils.eclipse.org">news:eo93md$g96$2@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>I just tried this on my machine =
with a=20
contrived example and it seemed to work fine ( If I understood =
this=20
correctly ;)&nbsp; ).</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Would it be possible for you to =
send more=20
information or a snippet of your metamodel so I can inspect=20
it.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>What errors are you seeing in the =
impl=20
class?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Regards,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- James.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt;=20
wrote in message <A=20
=
href=3D"news:eo7rou$ob0$1@utils.eclipse.org">news:eo7rou$ob0$1@utils.ecli=
pse.org</A>...</DIV>
<DIV>Hi all</DIV>
<DIV>I'm trying to make a new language metamodel on the basis on =
UML.uml=20
and using UML-Genmodel. In principle, the process is working but =
with=20
the following aspect I get into trouble:</DIV>
<DIV>&nbsp;</DIV>
<DIV>- In my metamodel I have two classes with an 1 -- 0..*=20
association</DIV>
<DIV>- the creation of the genmodel works well</DIV>
<DIV>- after model-code generation, there are errors in the =
impl-class=20
of the 0..* -side: in the following code-snipped the varable =
ownedrules=20
is not declared in the class</DIV>
<DIV>- if I change the Multiplicity to 1, the resulting code =
works</DIV>
<DIV><FONT face=3DArial size=3D2>
<P align=3Dleft><B><FONT color=3D#7f0055 =
size=3D2>public</B></FONT><FONT=20
size=3D2> EList getOwnedrules() {</P>
<P align=3Dleft></FONT><B><FONT color=3D#7f0055 =
size=3D2>if</B></FONT><FONT=20
size=3D2> (ownedrules =3D=3D </FONT><B><FONT color=3D#7f0055=20
size=3D2>null</B></FONT><FONT size=3D2>) {</P>
<P align=3Dleft>ownedrules =3D </FONT><B><FONT color=3D#7f0055=20
size=3D2>new</B></FONT><FONT size=3D2> =
EObjectWithInverseResolvingEList(</P>
<P align=3Dleft>MappingRule.</FONT><B><FONT color=3D#7f0055=20
size=3D2>class</B></FONT><FONT size=3D2>, </FONT><B><FONT =
color=3D#7f0055=20
size=3D2>this</B></FONT><FONT size=3D2>,</P>
<P align=3Dleft>UMLTPackage.</FONT><I><FONT color=3D#0000c0=20
size=3D2>TRANSFORMATION_GROUP__OWNEDRULES</I></FONT><FONT =
size=3D2>,</P>
<P align=3Dleft>UMLTPackage.</FONT><I><FONT color=3D#0000c0=20
size=3D2>MAPPING_RULE__TGROUP</I></FONT><FONT size=3D2>);</P>
<P align=3Dleft>}</P>
<P align=3Dleft></FONT><B><FONT color=3D#7f0055=20
size=3D2>return</B></FONT><FONT size=3D2> ownedrules;</P>
<P align=3Dleft>}</FONT></FONT></P></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV>In the attachments there are some pictures of the .uml and=20
.genmodel to get the context.</DIV>
<DIV>&nbsp;</DIV>
<DIV>Anyone an idea</DIV>
<DIV>Regards, Andreas</DIV>
<DIV><FONT face=3DArial=20
=
size=3D2></FONT>&nbsp;</DIV></BLOCKQUOTE></BLOCKQUOTE ></BLOCKQUOTE></BLOC=
KQUOTE></BODY></HTML>

------=_NextPart_000_079D_01C73B4D.6CDA39B0--
Re: Problem using UML Genmodel [message #572722 is a reply to message #468526] Fri, 19 January 2007 16:49 Go to previous message
Andreas Morf is currently offline Andreas MorfFriend
Messages: 14
Registered: July 2009
Junior Member
This is a multi-part message in MIME format.

------=_NextPart_000_0036_01C73BF2.2DD55AB0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Kenn,
I don't know how and why it comes, but now, there is no such method =
anymore when generating the model code - strange!
I will send you the details as soon as the problem occurs again...
So, how do you markup a class as a "model root" class in the genmodel?

Andreas
"Kenn Hussey" <khussey@ca.ibm.com> wrote in message =
news:eopcql$mcq$1@utils.eclipse.org...
- Could you provide more details on the stack overflow you are seeing? =
Every "model root" class will have this method in it, but it's not clear =
to me why this would result in a stack overflow...

- Association ends that are owned by the assocation are ignored when =
converting UML models to Ecore.

Kenn
"James Bruck" <jbruck@ca.ibm.com> wrote in message =
news:eopai8$174$1@utils.eclipse.org...
Hi Andreas,

some comments below ...

- James.
"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eolgq1$q5s$1@utils.eclipse.org...
Hi James
Thanks for your answer!
Fortunately the problem disappeared unexpectedly when I changed =
the Aggregation attribute of the memberEnd to "none". After switching =
back to 'composite' everything remained ok...

But after some experiences with the Genmodel there are some =
further questions:

- For every implementation class there is a getCacheAdapter() =
method generated (in the uml2.uml-impl-classes there is no such method) =
- this leads to a stack overflow using the generated editor plugin. =
Deleting the method manually resolves the problem. What am I doing wrong =
with the UML-Model/Genmodel?

+ Not sure about this one I'll have to investigate and get back =
to you. =20

- In the UML.merged.uml-model used as metamodel, what is the =
difference beween memberEnd Property with <eReference> Stereotype and =
the ones without Stereotype?

+ In general the Ecore stereotype provides information when =
converting from UMLto Ecore. Specific code generation details might be =
necessary for those items with stereotypes applied. Sometimes =
information is provided in the stereotype to allow conversion from Ecore =
to UML and back to Ecore and preserve the integrity of the original =
..ecore model ( ie. round tripping ).

- Using an UML-model as metamodel is there a special behaviour in =
Genmodel when using ownedEnd's in associations (as in =
A_general_generalization)?

+ I'm not sure what you mean about "special behavior". I'm not =
aware of unique behavior for this. I'll discuss it with Kenn and get =
back to you if there is something unique here.

And finally a question which is maybe not just related with uml2:
Getting all the sources from CVS (emf: 200701110200 uml2: =
200701111331) on 3.3M4 there are several errors in the UML2 projects =
even after updating the classpathes (org.eclipse.uml2, =
..uml2.importer.rose, . Is that just a strange behaviour of my =
environment?

+ I'm not seeing any build issues for the uml2 project. I do =
however have some issues with org.eclipse.emf.importer.rose. .. may =
not be the same issue you are having but these problems are a result of =
..jars needed for ant.
Sometimes there are .readme's that explain how to resolve these =
issues.

eg. Before running your build scripts or the examples provided =
here, make sure you have the 2 jar files required by this plugin =
("importer.rose.jar" and "importer.rose.tasks.jar") and also the 2 jar =
files required by the org.eclipse.emf.ant plugin ("emf.ant.jar" and =
"emf.ant.tasks.jar"). If necessary, Eclipse can generate them for you. =
All you need to do is to run the Ant script created by PDE when you =
right click on the manifest file and select "PDE Tools>Create Ant Build =
File".=20

but again, I don't see any issues with uml.


Greets, Andreas


"James Bruck" <jbruck@ca.ibm.com> wrote in message =
news:eo93md$g96$2@utils.eclipse.org...
Hi Andreas,

I just tried this on my machine with a contrived example and it =
seemed to work fine ( If I understood this correctly ;) ).
Would it be possible for you to send more information or a =
snippet of your metamodel so I can inspect it.
What errors are you seeing in the impl class?

Regards,

- James.

"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eo7rou$ob0$1@utils.eclipse.org...
Hi all
I'm trying to make a new language metamodel on the basis on =
UML.uml and using UML-Genmodel. In principle, the process is working but =
with the following aspect I get into trouble:

- In my metamodel I have two classes with an 1 -- 0..* =
association
- the creation of the genmodel works well
- after model-code generation, there are errors in the =
impl-class of the 0..* -side: in the following code-snipped the varable =
ownedrules is not declared in the class
- if I change the Multiplicity to 1, the resulting code works
public EList getOwnedrules() {

if (ownedrules =3D=3D null) {

ownedrules =3D new EObjectWithInverseResolvingEList(

MappingRule.class, this,

UMLTPackage.TRANSFORMATION_GROUP__OWNEDRULES,

UMLTPackage.MAPPING_RULE__TGROUP);

}

return ownedrules;

}


In the attachments there are some pictures of the .uml and =
..genmodel to get the context.

Anyone an idea
Regards, Andreas

------=_NextPart_000_0036_01C73BF2.2DD55AB0
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 6.00.5730.11" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>Kenn,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>I don't know how and why it comes, but =
now, there=20
is no such method anymore when generating the model code - =
strange!</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>I will send you the details as soon as =
the problem=20
occurs again...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>So, how do you markup a class as a =
"model root"=20
class in the genmodel?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Andreas</FONT></DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Kenn Hussey" &lt;<A=20
href=3D"mailto:khussey@ca.ibm.com">khussey@ca.ibm.com</A>&gt; wrote in =
message=20
<A=20
=
href=3D"news:eopcql$mcq$1@utils.eclipse.org">news:eopcql$mcq$1@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>- Could you provide more details on =
the stack=20
overflow you are seeing? Every "model root" class will have this =
method in it,=20
but it's not clear to me why this would result in a stack=20
overflow...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- Association ends that are owned by =
the=20
assocation are ignored when converting UML models to =
Ecore.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Kenn</FONT></DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"James Bruck" &lt;<A=20
href=3D"mailto:jbruck@ca.ibm.com">jbruck@ca.ibm.com</A>&gt; wrote in =
message=20
<A=20
=
href=3D"news:eopai8$174$1@utils.eclipse.org">news:eopai8$174$1@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>some comments below =
....</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- James.</FONT></DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt;=20
wrote in message <A=20
=
href=3D"news:eolgq1$q5s$1@utils.eclipse.org">news:eolgq1$q5s$1@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi James</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Thanks for your =
answer!</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Fortunately the problem =
disappeared=20
unexpectedly when I changed the Aggregation attribute of the =
memberEnd to=20
"none". After switching back to 'composite' everything remained=20
ok...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>But after some experiences with =
the Genmodel=20
there are some further questions:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- For every implementation class =
there is a=20
<FONT size=3D2>getCacheAdapter() method generated (in the=20
uml2.uml-impl-classes there is no such method) - this leads to a =
stack=20
overflow using the generated editor plugin. Deleting the method =
manually=20
resolves the problem. What am I doing wrong with the=20
UML-Model/Genmodel?</FONT></FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ Not sure about this one&nbsp; =
I'll have to=20
investigate and get back to you.&nbsp; </FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- In the UML.merged.uml-model =
used as=20
metamodel, what is the difference beween memberEnd Property with=20
&lt;eReference&gt; Stereotype and the ones without=20
Stereotype?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ In general the Ecore stereotype =
provides=20
information when converting from UMLto Ecore.&nbsp; Specific code=20
generation details might be necessary for those items with =
stereotypes=20
applied.&nbsp; Sometimes information is provided in the stereotype =
to=20
allow conversion from Ecore to UML and back to Ecore and preserve =
the=20
integrity of the original .ecore model ( ie. round tripping=20
).</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- Using an UML-model as metamodel =
is there a=20
special behaviour in Genmodel when using ownedEnd's in =
associations (as in=20
A_general_generalization)?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ I'm not sure what you mean =
about "special=20
behavior".&nbsp;&nbsp; I'm not aware of unique behavior for =
this.&nbsp;=20
I'll discuss it with Kenn and get back to you if there is =
something unique=20
here.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>And finally a question which is =
maybe not=20
just related with uml2:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Getting all the sources from CVS =
(emf:=20
200701110200 uml2: 200701111331) on 3.3M4 there are several errors =
in the=20
UML2 projects even after updating the classpathes =
(org.eclipse.uml2,=20
.uml2.importer.rose, . Is that just a strange behaviour of my=20
environment?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ I'm not seeing any build issues =
for the=20
uml2 project.&nbsp; I do however have some issues with=20
org.eclipse.emf.importer.rose.&nbsp;&nbsp; .. may not be the same =
issue=20
you are having but these problems are a result of .jars needed for =

ant.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Sometimes there are .readme's =
that explain=20
how to resolve these issues.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>eg.&nbsp;&nbsp; <FONT =
size=3D3><FONT=20
face=3D"Times New Roman">Before running your build scripts or the =
examples=20
provided here, make sure you have the 2 jar files required by this =
plugin=20
("importer.rose.jar" and "importer.rose.tasks.jar") and also the 2 =
jar=20
files required by the </FONT><TT>org.eclipse.emf.ant</TT><FONT=20
face=3D"Times New Roman"> plugin ("emf.ant.jar" and =
"emf.ant.tasks.jar"). If=20
necessary, Eclipse can generate them for you. All you need to do =
is to run=20
the Ant script created by PDE when you right click on the manifest =
file=20
and select "PDE Tools&gt;Create Ant Build File".=20
</FONT></FONT></FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>but again, I don't see any issues =
with=20
uml.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Greets, Andreas</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"James Bruck" &lt;<A=20
href=3D"mailto:jbruck@ca.ibm.com">jbruck@ca.ibm.com</A>&gt; =
wrote in=20
message <A=20
=
href=3D"news:eo93md$g96$2@utils.eclipse.org">news:eo93md$g96$2@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>I just tried this on my machine =
with a=20
contrived example and it seemed to work fine ( If I understood =
this=20
correctly ;)&nbsp; ).</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Would it be possible for you to =
send more=20
information or a snippet of your metamodel so I can inspect=20
it.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>What errors are you seeing in =
the impl=20
class?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Regards,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- James.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: =
5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt;=20
wrote in message <A=20
=
href=3D"news:eo7rou$ob0$1@utils.eclipse.org">news:eo7rou$ob0$1@utils.ecli=
pse.org</A>...</DIV>
<DIV>Hi all</DIV>
<DIV>I'm trying to make a new language metamodel on the basis =
on=20
UML.uml and using UML-Genmodel. In principle, the process is =
working=20
but with the following aspect I get into trouble:</DIV>
<DIV>&nbsp;</DIV>
<DIV>- In my metamodel I have two classes with an 1 -- 0..*=20
association</DIV>
<DIV>- the creation of the genmodel works well</DIV>
<DIV>- after model-code generation, there are errors in the =
impl-class=20
of the 0..* -side: in the following code-snipped the varable=20
ownedrules is not declared in the class</DIV>
<DIV>- if I change the Multiplicity to 1, the resulting code=20
works</DIV>
<DIV><FONT face=3DArial size=3D2>
<P align=3Dleft><B><FONT color=3D#7f0055 =
size=3D2>public</B></FONT><FONT=20
size=3D2> EList getOwnedrules() {</P>
<P align=3Dleft></FONT><B><FONT color=3D#7f0055 =
size=3D2>if</B></FONT><FONT=20
size=3D2> (ownedrules =3D=3D </FONT><B><FONT color=3D#7f0055=20
size=3D2>null</B></FONT><FONT size=3D2>) {</P>
<P align=3Dleft>ownedrules =3D </FONT><B><FONT color=3D#7f0055 =

size=3D2>new</B></FONT><FONT size=3D2>=20
EObjectWithInverseResolvingEList(</P>
<P align=3Dleft>MappingRule.</FONT><B><FONT color=3D#7f0055=20
size=3D2>class</B></FONT><FONT size=3D2>, </FONT><B><FONT =
color=3D#7f0055=20
size=3D2>this</B></FONT><FONT size=3D2>,</P>
<P align=3Dleft>UMLTPackage.</FONT><I><FONT color=3D#0000c0=20
size=3D2>TRANSFORMATION_GROUP__OWNEDRULES</I></FONT><FONT =
size=3D2>,</P>
<P align=3Dleft>UMLTPackage.</FONT><I><FONT color=3D#0000c0=20
size=3D2>MAPPING_RULE__TGROUP</I></FONT><FONT size=3D2>);</P>
<P align=3Dleft>}</P>
<P align=3Dleft></FONT><B><FONT color=3D#7f0055=20
size=3D2>return</B></FONT><FONT size=3D2> ownedrules;</P>
<P align=3Dleft>}</FONT></FONT></P></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV>In the attachments there are some pictures of the .uml =
and=20
.genmodel to get the context.</DIV>
<DIV>&nbsp;</DIV>
<DIV>Anyone an idea</DIV>
<DIV>Regards, Andreas</DIV>
<DIV><FONT face=3DArial=20
=
size=3D2></FONT>&nbsp;</DIV></BLOCKQUOTE></BLOCKQUOTE ></BLOCKQUOTE></BLOC=
KQUOTE></BLOCKQUOTE></BODY></HTML>

------=_NextPart_000_0036_01C73BF2.2DD55AB0--
Re: Problem using UML Genmodel [message #572776 is a reply to message #468690] Fri, 19 January 2007 18:05 Go to previous message
Kenn Hussey is currently offline Kenn HusseyFriend
Messages: 1620
Registered: July 2009
Senior Member
This is a multi-part message in MIME format.

------=_NextPart_000_07E8_01C73BCA.837935E0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Andreas,

Whether a class is considered a model root is determined in EMF like =
this:

public boolean isModelRoot()
{
return getClassExtendsGenClass() =3D=3D null || =
getClassExtendsGenClass().getGenModel() !=3D getGenModel();
}

Kenn

"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eoqsqu$i5o$1@utils.eclipse.org...
Kenn,
I don't know how and why it comes, but now, there is no such method =
anymore when generating the model code - strange!
I will send you the details as soon as the problem occurs again...
So, how do you markup a class as a "model root" class in the genmodel?

Andreas
"Kenn Hussey" <khussey@ca.ibm.com> wrote in message =
news:eopcql$mcq$1@utils.eclipse.org...
- Could you provide more details on the stack overflow you are =
seeing? Every "model root" class will have this method in it, but it's =
not clear to me why this would result in a stack overflow...

- Association ends that are owned by the assocation are ignored when =
converting UML models to Ecore.

Kenn
"James Bruck" <jbruck@ca.ibm.com> wrote in message =
news:eopai8$174$1@utils.eclipse.org...
Hi Andreas,

some comments below ...

- James.
"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eolgq1$q5s$1@utils.eclipse.org...
Hi James
Thanks for your answer!
Fortunately the problem disappeared unexpectedly when I changed =
the Aggregation attribute of the memberEnd to "none". After switching =
back to 'composite' everything remained ok...

But after some experiences with the Genmodel there are some =
further questions:

- For every implementation class there is a getCacheAdapter() =
method generated (in the uml2.uml-impl-classes there is no such method) =
- this leads to a stack overflow using the generated editor plugin. =
Deleting the method manually resolves the problem. What am I doing wrong =
with the UML-Model/Genmodel?

+ Not sure about this one I'll have to investigate and get back =
to you. =20

- In the UML.merged.uml-model used as metamodel, what is the =
difference beween memberEnd Property with <eReference> Stereotype and =
the ones without Stereotype?

+ In general the Ecore stereotype provides information when =
converting from UMLto Ecore. Specific code generation details might be =
necessary for those items with stereotypes applied. Sometimes =
information is provided in the stereotype to allow conversion from Ecore =
to UML and back to Ecore and preserve the integrity of the original =
..ecore model ( ie. round tripping ).

- Using an UML-model as metamodel is there a special behaviour =
in Genmodel when using ownedEnd's in associations (as in =
A_general_generalization)?

+ I'm not sure what you mean about "special behavior". I'm not =
aware of unique behavior for this. I'll discuss it with Kenn and get =
back to you if there is something unique here.

And finally a question which is maybe not just related with =
uml2:
Getting all the sources from CVS (emf: 200701110200 uml2: =
200701111331) on 3.3M4 there are several errors in the UML2 projects =
even after updating the classpathes (org.eclipse.uml2, =
..uml2.importer.rose, . Is that just a strange behaviour of my =
environment?

+ I'm not seeing any build issues for the uml2 project. I do =
however have some issues with org.eclipse.emf.importer.rose. .. may =
not be the same issue you are having but these problems are a result of =
..jars needed for ant.
Sometimes there are .readme's that explain how to resolve these =
issues.

eg. Before running your build scripts or the examples provided =
here, make sure you have the 2 jar files required by this plugin =
("importer.rose.jar" and "importer.rose.tasks.jar") and also the 2 jar =
files required by the org.eclipse.emf.ant plugin ("emf.ant.jar" and =
"emf.ant.tasks.jar"). If necessary, Eclipse can generate them for you. =
All you need to do is to run the Ant script created by PDE when you =
right click on the manifest file and select "PDE Tools>Create Ant Build =
File".=20

but again, I don't see any issues with uml.


Greets, Andreas


"James Bruck" <jbruck@ca.ibm.com> wrote in message =
news:eo93md$g96$2@utils.eclipse.org...
Hi Andreas,

I just tried this on my machine with a contrived example and =
it seemed to work fine ( If I understood this correctly ;) ).
Would it be possible for you to send more information or a =
snippet of your metamodel so I can inspect it.
What errors are you seeing in the impl class?

Regards,

- James.

"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message =
news:eo7rou$ob0$1@utils.eclipse.org...
Hi all
I'm trying to make a new language metamodel on the basis on =
UML.uml and using UML-Genmodel. In principle, the process is working but =
with the following aspect I get into trouble:

- In my metamodel I have two classes with an 1 -- 0..* =
association
- the creation of the genmodel works well
- after model-code generation, there are errors in the =
impl-class of the 0..* -side: in the following code-snipped the varable =
ownedrules is not declared in the class
- if I change the Multiplicity to 1, the resulting code =
works
public EList getOwnedrules() {

if (ownedrules =3D=3D null) {

ownedrules =3D new EObjectWithInverseResolvingEList(

MappingRule.class, this,

UMLTPackage.TRANSFORMATION_GROUP__OWNEDRULES,

UMLTPackage.MAPPING_RULE__TGROUP);

}

return ownedrules;

}


In the attachments there are some pictures of the .uml and =
..genmodel to get the context.

Anyone an idea
Regards, Andreas

------=_NextPart_000_07E8_01C73BCA.837935E0
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 6.00.2900.3020" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Whether a class is considered a model =
root is=20
determined in EMF like this:</FONT></DIV>
<DIV><B><FONT color=3D#7f0055 size=3D2></FONT></B>&nbsp;</DIV>
<DIV><B><FONT color=3D#7f0055 size=3D2>public</B></FONT><FONT size=3D2>=20
</FONT><B><FONT color=3D#7f0055 size=3D2>boolean</B></FONT><FONT =
size=3D2>=20
isModelRoot()</FONT></DIV>
<DIV><FONT size=3D2>{</FONT></DIV>
<DIV><B><FONT color=3D#7f0055 size=3D2>&nbsp;&nbsp;&nbsp; =
return</B></FONT><FONT=20
size=3D2> getClassExtendsGenClass() =3D=3D </FONT><B><FONT =
color=3D#7f0055=20
size=3D2>null</B></FONT><FONT size=3D2> || =
getClassExtendsGenClass().getGenModel()=20
!=3D getGenModel();</FONT></DIV>
<DIV><FONT size=3D2>}</DIV></FONT>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Kenn</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt; =
wrote in=20
message <A=20
=
href=3D"news:eoqsqu$i5o$1@utils.eclipse.org">news:eoqsqu$i5o$1@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Kenn,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>I don't know how and why it comes, =
but now, there=20
is no such method anymore when generating the model code -=20
strange!</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>I will send you the details as soon =
as the=20
problem occurs again...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>So, how do you markup a class as a =
"model root"=20
class in the genmodel?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Andreas</FONT></DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Kenn Hussey" &lt;<A=20
href=3D"mailto:khussey@ca.ibm.com">khussey@ca.ibm.com</A>&gt; wrote =
in message=20
<A=20
=
href=3D"news:eopcql$mcq$1@utils.eclipse.org">news:eopcql$mcq$1@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>- Could you provide more details on =
the stack=20
overflow you are seeing? Every "model root" class will have this =
method in=20
it, but it's not clear to me why this would result in a stack=20
overflow...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- Association ends that are owned =
by the=20
assocation are ignored when converting UML models to =
Ecore.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Kenn</FONT></DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"James Bruck" &lt;<A=20
href=3D"mailto:jbruck@ca.ibm.com">jbruck@ca.ibm.com</A>&gt; wrote =
in message=20
<A=20
=
href=3D"news:eopai8$174$1@utils.eclipse.org">news:eopai8$174$1@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>some comments below =
....</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- James.</FONT></DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt;=20
wrote in message <A=20
=
href=3D"news:eolgq1$q5s$1@utils.eclipse.org">news:eolgq1$q5s$1@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi James</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Thanks for your =
answer!</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Fortunately the problem =
disappeared=20
unexpectedly when I changed the Aggregation attribute of the =
memberEnd=20
to "none". After switching back to 'composite' everything =
remained=20
ok...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>But after some experiences with =
the=20
Genmodel there are some further questions:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- For every implementation =
class there is a=20
<FONT size=3D2>getCacheAdapter() method generated (in the=20
uml2.uml-impl-classes there is no such method) - this leads to a =
stack=20
overflow using the generated editor plugin. Deleting the method =
manually=20
resolves the problem. What am I doing wrong with the=20
UML-Model/Genmodel?</FONT></FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ Not sure about this one&nbsp; =
I'll have=20
to investigate and get back to you.&nbsp; </FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- In the UML.merged.uml-model =
used as=20
metamodel, what is the difference beween memberEnd Property with =

&lt;eReference&gt; Stereotype and the ones without=20
Stereotype?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ In general the Ecore =
stereotype provides=20
information when converting from UMLto Ecore.&nbsp; Specific =
code=20
generation details might be necessary for those items with =
stereotypes=20
applied.&nbsp; Sometimes information is provided in the =
stereotype to=20
allow conversion from Ecore to UML and back to Ecore and =
preserve the=20
integrity of the original .ecore model ( ie. round tripping=20
).</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- Using an UML-model as =
metamodel is there=20
a special behaviour in Genmodel when using ownedEnd's in =
associations=20
(as in A_general_generalization)?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ I'm not sure what you mean =
about "special=20
behavior".&nbsp;&nbsp; I'm not aware of unique behavior for =
this.&nbsp;=20
I'll discuss it with Kenn and get back to you if there is =
something=20
unique here.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>And finally a question which is =
maybe not=20
just related with uml2:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Getting all the sources from =
CVS (emf:=20
200701110200 uml2: 200701111331) on 3.3M4 there are several =
errors in=20
the UML2 projects even after updating the classpathes =
(org.eclipse.uml2,=20
.uml2.importer.rose, . Is that just a strange behaviour of my=20
environment?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>+ I'm not seeing any build =
issues for the=20
uml2 project.&nbsp; I do however have some issues with=20
org.eclipse.emf.importer.rose.&nbsp;&nbsp; .. may not be the =
same issue=20
you are having but these problems are a result of .jars needed =
for=20
ant.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Sometimes there are .readme's =
that explain=20
how to resolve these issues.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>eg.&nbsp;&nbsp; <FONT =
size=3D3><FONT=20
face=3D"Times New Roman">Before running your build scripts or =
the examples=20
provided here, make sure you have the 2 jar files required by =
this=20
plugin ("importer.rose.jar" and "importer.rose.tasks.jar") and =
also the=20
2 jar files required by the =
</FONT><TT>org.eclipse.emf.ant</TT><FONT=20
face=3D"Times New Roman"> plugin ("emf.ant.jar" and =
"emf.ant.tasks.jar").=20
If necessary, Eclipse can generate them for you. All you need to =
do is=20
to run the Ant script created by PDE when you right click on the =

manifest file and select "PDE Tools&gt;Create Ant Build File".=20
</FONT></FONT></FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>but again, I don't see any =
issues with=20
uml.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Greets, Andreas</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: =
5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"James Bruck" &lt;<A=20
href=3D"mailto:jbruck@ca.ibm.com">jbruck@ca.ibm.com</A>&gt; =
wrote in=20
message <A=20
=
href=3D"news:eo93md$g96$2@utils.eclipse.org">news:eo93md$g96$2@utils.ecli=
pse.org</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>Hi Andreas,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>I just tried this on my =
machine with a=20
contrived example and it seemed to work fine ( If I understood =
this=20
correctly ;)&nbsp; ).</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Would it be possible for you =
to send more=20
information or a snippet of your metamodel so I can inspect=20
it.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>What errors are you seeing in =
the impl=20
class?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Regards,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>- James.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: =
5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Andreas Morf" &lt;<A=20
=
href=3D"mailto:andreas_morf@bluewin.ch">andreas_morf@bluewin.ch</A>&gt;=20
wrote in message <A=20
=
href=3D"news:eo7rou$ob0$1@utils.eclipse.org">news:eo7rou$ob0$1@utils.ecli=
pse.org</A>...</DIV>
<DIV>Hi all</DIV>
<DIV>I'm trying to make a new language metamodel on the =
basis on=20
UML.uml and using UML-Genmodel. In principle, the process is =
working=20
but with the following aspect I get into trouble:</DIV>
<DIV>&nbsp;</DIV>
<DIV>- In my metamodel I have two classes with an 1 -- 0..*=20
association</DIV>
<DIV>- the creation of the genmodel works well</DIV>
<DIV>- after model-code generation, there are errors in the=20
impl-class of the 0..* -side: in the following code-snipped =
the=20
varable ownedrules is not declared in the class</DIV>
<DIV>- if I change the Multiplicity to 1, the resulting code =

works</DIV>
<DIV><FONT face=3DArial size=3D2>
<P align=3Dleft><B><FONT color=3D#7f0055 =
size=3D2>public</B></FONT><FONT=20
size=3D2> EList getOwnedrules() {</P>
<P align=3Dleft></FONT><B><FONT color=3D#7f0055=20
size=3D2>if</B></FONT><FONT size=3D2> (ownedrules =3D=3D =
</FONT><B><FONT=20
color=3D#7f0055 size=3D2>null</B></FONT><FONT size=3D2>) =
{</P>
<P align=3Dleft>ownedrules =3D </FONT><B><FONT =
color=3D#7f0055=20
size=3D2>new</B></FONT><FONT size=3D2>=20
EObjectWithInverseResolvingEList(</P>
<P align=3Dleft>MappingRule.</FONT><B><FONT color=3D#7f0055=20
size=3D2>class</B></FONT><FONT size=3D2>, </FONT><B><FONT =
color=3D#7f0055=20
size=3D2>this</B></FONT><FONT size=3D2>,</P>
<P align=3Dleft>UMLTPackage.</FONT><I><FONT color=3D#0000c0=20
size=3D2>TRANSFORMATION_GROUP__OWNEDRULES</I></FONT><FONT =
size=3D2>,</P>
<P align=3Dleft>UMLTPackage.</FONT><I><FONT color=3D#0000c0=20
size=3D2>MAPPING_RULE__TGROUP</I></FONT><FONT =
size=3D2>);</P>
<P align=3Dleft>}</P>
<P align=3Dleft></FONT><B><FONT color=3D#7f0055=20
size=3D2>return</B></FONT><FONT size=3D2> ownedrules;</P>
<P align=3Dleft>}</FONT></FONT></P></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV>In the attachments there are some pictures of the .uml =
and=20
.genmodel to get the context.</DIV>
<DIV>&nbsp;</DIV>
<DIV>Anyone an idea</DIV>
<DIV>Regards, Andreas</DIV>
<DIV><FONT face=3DArial=20
=
size=3D2></FONT>&nbsp;</DIV></BLOCKQUOTE></BLOCKQUOTE ></BLOCKQUOTE></BLOC=
KQUOTE></BLOCKQUOTE></BLOCKQUOTE></BODY></HTML>

------=_NextPart_000_07E8_01C73BCA.837935E0--
Re: Problem using UML Genmodel [message #574069 is a reply to message #468690] Fri, 26 January 2007 10:15 Go to previous message
Andreas Morf is currently offline Andreas MorfFriend
Messages: 14
Registered: July 2009
Junior Member
After refactoring my metamodel and deleting all of the generated code, the
same problem with the getCacheAdapter appears again:
My metaclass inherits from uml2.uml.Model and therefore the root class
should be ElementImpl but somehow the generator isn't aware of that and
gerates the following method leading to the StackOverflow (pingpong):

protected CacheAdapter getCacheAdapter() {
return CacheAdapter.getCacheAdapter(this);
}

So changing this method to the following, the code works - but in my
opinion, this should not be the preferred way:

protected CacheAdapter getCacheAdapter() {
return super.getCacheAdapter();
}

Why is this generated method different from that in the ElementImpl-Class
Also, there is to say that I Model/genmodel importing/extending Metaclasses
from UML.merged.uml

Any ideas out there?
Greetings, Andreas
Re: Problem using UML Genmodel [message #574201 is a reply to message #468946] Fri, 26 January 2007 19:46 Go to previous message
Kenn Hussey is currently offline Kenn HusseyFriend
Messages: 1620
Registered: July 2009
Senior Member
Andreas,

This is a problem only because you are extending UML2. If you are extending
UML2 and want to use the same cache adapter, overriding the method as you
have is the right choice (the other alternative is to disable cache adapter
support for your generator model). However, we should certainly avoid the
stack overflow for cases where a different cache adapter is used (the
default) - please open a Bugzilla report and we'll have this fixed.

Kenn

"Andreas Morf" <andreas_morf@bluewin.ch> wrote in message
news:epckbf$pal$1@utils.eclipse.org...
> After refactoring my metamodel and deleting all of the generated code, the
> same problem with the getCacheAdapter appears again:
> My metaclass inherits from uml2.uml.Model and therefore the root class
> should be ElementImpl but somehow the generator isn't aware of that and
> gerates the following method leading to the StackOverflow (pingpong):
>
> protected CacheAdapter getCacheAdapter() {
> return CacheAdapter.getCacheAdapter(this);
> }
>
> So changing this method to the following, the code works - but in my
> opinion, this should not be the preferred way:
>
> protected CacheAdapter getCacheAdapter() {
> return super.getCacheAdapter();
> }
>
> Why is this generated method different from that in the ElementImpl-Class
> Also, there is to say that I Model/genmodel importing/extending
> Metaclasses from UML.merged.uml
>
> Any ideas out there?
> Greetings, Andreas
>
Previous Topic:Eclipse UML2 and other tools compatibility
Next Topic:Platform resource set/get
Goto Forum:
  


Current Time: Fri Apr 19 23:21:14 GMT 2024

Powered by FUDForum. Page generated in 0.04936 seconds
.:: Contact :: Home ::.

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

Back to the top