Eclipse Community Forums
Forum Search:

Search      Help    Register    Login    Home
Home » Modeling » UML2 Tools » diamod direction in case of association
diamod direction in case of association [message #470227] Fri, 27 April 2007 19:01 Go to next message
Pradeep is currently offline Pradeep
Messages: 61
Registered: July 2009
Member
In the following case, a catalog has one or more views. I am just wondering
whether the diamond symbol is correctly placed. I think it should be with
the Catalog.

Thanks
Pradeep


  • Attachment: arrow.PNG
    (Size: 11.02KB, Downloaded 100 times)
Re: diamod direction in case of association [message #470229 is a reply to message #470227] Mon, 30 April 2007 06:54 Go to previous messageGo to next message
Michael Golubev is currently offline Michael Golubev
Messages: 373
Registered: July 2009
Senior Member
Hello Pradeep,

In general, the diamond symbol presents and location depends on the
corresponding propertie's aggregation value (Property#getAggregation()).
In prinicple it is possible to see the link with diamonds for _both_ ends,
in case if the aggregation value for both of property ends is COMPOSITE.

In particular, I believe that in your model, there are 3 properties that
make sense

1. Catalog::views that is NOT actually part of the association (in properties
view, it does NOT have any association set)
2. Association::views : View, aggregation = COMPOSITE, that is show as link
source with diamond decoration.
3. Association::<unnamed> : Catalog, aggregation = NONE, that is shown as
link target

I am almost sure that (1) is not related to the association, because we EXPLICITLY
FILTERING
properties that are parts of binary associations from being shown in attributes
compartment.

Please check your model resource, and provide me with actual model data in
case I am wrong here.

Also please note that we have recently implemented the set of actions (#182749)
that helps to tranfsorm
single or paired set of class properties typed by different classifiers into
the association between them.
I believe that it will simplify the creation of the sample picture attached
avoiding possible confusion with Catalog::views and Association::views.

Does it make sense?

Regards,
Michael


> In the following case, a catalog has one or more views. I am just
> wondering whether the diamond symbol is correctly placed. I think it
> should be with the Catalog.
>
> Thanks
> Pradeep
Re: diamod direction in case of association [message #470239 is a reply to message #470229] Tue, 01 May 2007 02:07 Go to previous messageGo to next message
Pradeep is currently offline Pradeep
Messages: 61
Registered: July 2009
Member
Hi Michael,

Thanks for the explanation. I think I should study more about associations.
I am still quite confused.
I attached the purchase order diagram that is generated using the editor.
Please find it attached.
The original one is available at the following URL.

http://www.eclipse.org/modeling/mdt/uml2/docs/articles/Getti ng_Started_with_UML2/article.html

If you compare these two diagrams, the diamond shapes are positioned exactly
in the reverse order. I am not sure which one is correct now.

Pradeep


"Michael Golubev" <michael.golubev@borland.com> wrote in message
news:80ef4340a5c18c9593dde7dba0c@news.eclipse.org...
> Hello Pradeep,
>
> In general, the diamond symbol presents and location depends on the
> corresponding propertie's aggregation value (Property#getAggregation()).
> In prinicple it is possible to see the link with diamonds for _both_ ends,
> in case if the aggregation value for both of property ends is COMPOSITE.
>
> In particular, I believe that in your model, there are 3 properties that
> make sense
>
> 1. Catalog::views that is NOT actually part of the association (in
> properties
> view, it does NOT have any association set)
> 2. Association::views : View, aggregation = COMPOSITE, that is show as
> link
> source with diamond decoration.
> 3. Association::<unnamed> : Catalog, aggregation = NONE, that is shown as
> link target
>
> I am almost sure that (1) is not related to the association, because we
> EXPLICITLY
> FILTERING
> properties that are parts of binary associations from being shown in
> attributes
> compartment.
>
> Please check your model resource, and provide me with actual model data in
> case I am wrong here.
>
> Also please note that we have recently implemented the set of actions
> (#182749)
> that helps to tranfsorm
> single or paired set of class properties typed by different classifiers
> into
> the association between them.
> I believe that it will simplify the creation of the sample picture
> attached
> avoiding possible confusion with Catalog::views and Association::views.
>
> Does it make sense?
>
> Regards,
> Michael
>
>
>> In the following case, a catalog has one or more views. I am just
>> wondering whether the diamond symbol is correctly placed. I think it
>> should be with the Catalog.
>>
>> Thanks
>> Pradeep
>
>


Re: diamod direction in case of association [message #470245 is a reply to message #470239] Tue, 01 May 2007 03:01 Go to previous messageGo to next message
Pradeep is currently offline Pradeep
Messages: 61
Registered: July 2009
Member
One more observation,

In the generated diagram, attributes are placed on both association links
and inside the respective classes.
But in the diagram on the webpage, we see attributes only on the links which
I think is less redundant.

Pradeep


"Pradeep" <pradeep@live.com> wrote in message
news:f16lg1$hq7$1@build.eclipse.org...
> Hi Michael,
>
> Thanks for the explanation. I think I should study more about
> associations. I am still quite confused.
> I attached the purchase order diagram that is generated using the editor.
> Please find it attached.
> The original one is available at the following URL.
>
> http://www.eclipse.org/modeling/mdt/uml2/docs/articles/Getti ng_Started_with_UML2/article.html
>
> If you compare these two diagrams, the diamond shapes are positioned
> exactly in the reverse order. I am not sure which one is correct now.
>
> Pradeep
>
>
> "Michael Golubev" <michael.golubev@borland.com> wrote in message
> news:80ef4340a5c18c9593dde7dba0c@news.eclipse.org...
>> Hello Pradeep,
>>
>> In general, the diamond symbol presents and location depends on the
>> corresponding propertie's aggregation value (Property#getAggregation()).
>> In prinicple it is possible to see the link with diamonds for _both_
>> ends,
>> in case if the aggregation value for both of property ends is COMPOSITE.
>>
>> In particular, I believe that in your model, there are 3 properties that
>> make sense
>>
>> 1. Catalog::views that is NOT actually part of the association (in
>> properties
>> view, it does NOT have any association set)
>> 2. Association::views : View, aggregation = COMPOSITE, that is show as
>> link
>> source with diamond decoration.
>> 3. Association::<unnamed> : Catalog, aggregation = NONE, that is shown as
>> link target
>>
>> I am almost sure that (1) is not related to the association, because we
>> EXPLICITLY
>> FILTERING
>> properties that are parts of binary associations from being shown in
>> attributes
>> compartment.
>>
>> Please check your model resource, and provide me with actual model data
>> in
>> case I am wrong here.
>>
>> Also please note that we have recently implemented the set of actions
>> (#182749)
>> that helps to tranfsorm
>> single or paired set of class properties typed by different classifiers
>> into
>> the association between them.
>> I believe that it will simplify the creation of the sample picture
>> attached
>> avoiding possible confusion with Catalog::views and Association::views.
>>
>> Does it make sense?
>>
>> Regards,
>> Michael
>>
>>
>>> In the following case, a catalog has one or more views. I am just
>>> wondering whether the diamond symbol is correctly placed. I think it
>>> should be with the Catalog.
>>>
>>> Thanks
>>> Pradeep
>>
>>
>
>
>
Re: diamod direction in case of association [message #470249 is a reply to message #470245] Wed, 02 May 2007 05:17 Go to previous messageGo to next message
Michael Golubev is currently offline Michael Golubev
Messages: 373
Registered: July 2009
Senior Member
Hello Pradeep,

Thank you for persistence in convinicing me, I see that I was wrong claiming
this as implemented correctly.
Just reporoduced the issue, submitted bugzillas #185045 and #185046.

Regards,
Michael


> One more observation,
>
> In the generated diagram, attributes are placed on both association
> links
> and inside the respective classes.
> But in the diagram on the webpage, we see attributes only on the links
> which
> I think is less redundant.
> Pradeep
>
> "Pradeep" <pradeep@live.com> wrote in message
> news:f16lg1$hq7$1@build.eclipse.org...
>
>> Hi Michael,
>>
>> Thanks for the explanation. I think I should study more about
>> associations. I am still quite confused.
>> I attached the purchase order diagram that is generated using the
>> editor.
>> Please find it attached.
>> The original one is available at the following URL.
>> http://www.eclipse.org/modeling/mdt/uml2/docs/articles/Getti ng_Starte
>> d_with_UML2/article.html
>>
>> If you compare these two diagrams, the diamond shapes are positioned
>> exactly in the reverse order. I am not sure which one is correct now.
>>
>> Pradeep
>>
>> "Michael Golubev" <michael.golubev@borland.com> wrote in message
>> news:80ef4340a5c18c9593dde7dba0c@news.eclipse.org...
>>
>>> Hello Pradeep,
>>>
>>> In general, the diamond symbol presents and location depends on the
>>> corresponding propertie's aggregation value
>>> (Property#getAggregation()).
>>> In prinicple it is possible to see the link with diamonds for _both_
>>> ends,
>>> in case if the aggregation value for both of property ends is
>>> COMPOSITE.
>>> In particular, I believe that in your model, there are 3 properties
>>> that make sense
>>>
>>> 1. Catalog::views that is NOT actually part of the association (in
>>> properties
>>> view, it does NOT have any association set)
>>> 2. Association::views : View, aggregation = COMPOSITE, that is show
>>> as
>>> link
>>> source with diamond decoration.
>>> 3. Association::<unnamed> : Catalog, aggregation = NONE, that is
>>> shown as
>>> link target
>>> I am almost sure that (1) is not related to the association, because
>>> we
>>> EXPLICITLY
>>> FILTERING
>>> properties that are parts of binary associations from being shown in
>>> attributes
>>> compartment.
>>> Please check your model resource, and provide me with actual model
>>> data
>>> in
>>> case I am wrong here.
>>> Also please note that we have recently implemented the set of
>>> actions
>>> (#182749)
>>> that helps to tranfsorm
>>> single or paired set of class properties typed by different
>>> classifiers
>>> into
>>> the association between them.
>>> I believe that it will simplify the creation of the sample picture
>>> attached
>>> avoiding possible confusion with Catalog::views and
>>> Association::views.
>>> Does it make sense?
>>>
>>> Regards,
>>> Michael
>>>> In the following case, a catalog has one or more views. I am just
>>>> wondering whether the diamond symbol is correctly placed. I think
>>>> it should be with the Catalog.
>>>>
>>>> Thanks
>>>> Pradeep
Re: diamod direction in case of association [message #470267 is a reply to message #470249] Wed, 02 May 2007 18:13 Go to previous message
Pradeep is currently offline Pradeep
Messages: 61
Registered: July 2009
Member
:)

"Michael Golubev" <michael.golubev@borland.com> wrote in message
news:80ef4340aa3f8c95ac2a94927c8@news.eclipse.org...
> Hello Pradeep,
>
> Thank you for persistence in convinicing me, I see that I was wrong
> claiming this as implemented correctly.
> Just reporoduced the issue, submitted bugzillas #185045 and #185046.
> Regards, Michael
>
>
>> One more observation,
>>
>> In the generated diagram, attributes are placed on both association
>> links
>> and inside the respective classes.
>> But in the diagram on the webpage, we see attributes only on the links
>> which
>> I think is less redundant.
>> Pradeep
>>
>> "Pradeep" <pradeep@live.com> wrote in message
>> news:f16lg1$hq7$1@build.eclipse.org...
>>
>>> Hi Michael,
>>>
>>> Thanks for the explanation. I think I should study more about
>>> associations. I am still quite confused.
>>> I attached the purchase order diagram that is generated using the
>>> editor.
>>> Please find it attached.
>>> The original one is available at the following URL.
>>> http://www.eclipse.org/modeling/mdt/uml2/docs/articles/Getti ng_Starte
>>> d_with_UML2/article.html
>>>
>>> If you compare these two diagrams, the diamond shapes are positioned
>>> exactly in the reverse order. I am not sure which one is correct now.
>>>
>>> Pradeep
>>>
>>> "Michael Golubev" <michael.golubev@borland.com> wrote in message
>>> news:80ef4340a5c18c9593dde7dba0c@news.eclipse.org...
>>>
>>>> Hello Pradeep,
>>>>
>>>> In general, the diamond symbol presents and location depends on the
>>>> corresponding propertie's aggregation value
>>>> (Property#getAggregation()).
>>>> In prinicple it is possible to see the link with diamonds for _both_
>>>> ends,
>>>> in case if the aggregation value for both of property ends is
>>>> COMPOSITE.
>>>> In particular, I believe that in your model, there are 3 properties
>>>> that make sense
>>>>
>>>> 1. Catalog::views that is NOT actually part of the association (in
>>>> properties
>>>> view, it does NOT have any association set)
>>>> 2. Association::views : View, aggregation = COMPOSITE, that is show
>>>> as
>>>> link
>>>> source with diamond decoration.
>>>> 3. Association::<unnamed> : Catalog, aggregation = NONE, that is
>>>> shown as
>>>> link target
>>>> I am almost sure that (1) is not related to the association, because
>>>> we
>>>> EXPLICITLY
>>>> FILTERING
>>>> properties that are parts of binary associations from being shown in
>>>> attributes
>>>> compartment.
>>>> Please check your model resource, and provide me with actual model
>>>> data
>>>> in
>>>> case I am wrong here.
>>>> Also please note that we have recently implemented the set of
>>>> actions
>>>> (#182749)
>>>> that helps to tranfsorm
>>>> single or paired set of class properties typed by different
>>>> classifiers
>>>> into
>>>> the association between them.
>>>> I believe that it will simplify the creation of the sample picture
>>>> attached
>>>> avoiding possible confusion with Catalog::views and
>>>> Association::views.
>>>> Does it make sense?
>>>>
>>>> Regards,
>>>> Michael
>>>>> In the following case, a catalog has one or more views. I am just
>>>>> wondering whether the diamond symbol is correctly placed. I think
>>>>> it should be with the Catalog.
>>>>>
>>>>> Thanks
>>>>> Pradeep
>
>
Re: diamod direction in case of association [message #588811 is a reply to message #470227] Mon, 30 April 2007 06:54 Go to previous message
Michael Golubev is currently offline Michael Golubev
Messages: 373
Registered: July 2009
Senior Member
Hello Pradeep,

In general, the diamond symbol presents and location depends on the
corresponding propertie's aggregation value (Property#getAggregation()).
In prinicple it is possible to see the link with diamonds for _both_ ends,
in case if the aggregation value for both of property ends is COMPOSITE.

In particular, I believe that in your model, there are 3 properties that
make sense

1. Catalog::views that is NOT actually part of the association (in properties
view, it does NOT have any association set)
2. Association::views : View, aggregation = COMPOSITE, that is show as link
source with diamond decoration.
3. Association::<unnamed> : Catalog, aggregation = NONE, that is shown as
link target

I am almost sure that (1) is not related to the association, because we EXPLICITLY
FILTERING
properties that are parts of binary associations from being shown in attributes
compartment.

Please check your model resource, and provide me with actual model data in
case I am wrong here.

Also please note that we have recently implemented the set of actions (#182749)
that helps to tranfsorm
single or paired set of class properties typed by different classifiers into
the association between them.
I believe that it will simplify the creation of the sample picture attached
avoiding possible confusion with Catalog::views and Association::views.

Does it make sense?

Regards,
Michael


> In the following case, a catalog has one or more views. I am just
> wondering whether the diamond symbol is correctly placed. I think it
> should be with the Catalog.
>
> Thanks
> Pradeep
Re: diamod direction in case of association [message #588868 is a reply to message #470229] Tue, 01 May 2007 02:07 Go to previous message
Pradeep is currently offline Pradeep
Messages: 61
Registered: July 2009
Member
Hi Michael,

Thanks for the explanation. I think I should study more about associations.
I am still quite confused.
I attached the purchase order diagram that is generated using the editor.
Please find it attached.
The original one is available at the following URL.

http://www.eclipse.org/modeling/mdt/uml2/docs/articles/Getti ng_Started_with_UML2/article.html

If you compare these two diagrams, the diamond shapes are positioned exactly
in the reverse order. I am not sure which one is correct now.

Pradeep


"Michael Golubev" <michael.golubev@borland.com> wrote in message
news:80ef4340a5c18c9593dde7dba0c@news.eclipse.org...
> Hello Pradeep,
>
> In general, the diamond symbol presents and location depends on the
> corresponding propertie's aggregation value (Property#getAggregation()).
> In prinicple it is possible to see the link with diamonds for _both_ ends,
> in case if the aggregation value for both of property ends is COMPOSITE.
>
> In particular, I believe that in your model, there are 3 properties that
> make sense
>
> 1. Catalog::views that is NOT actually part of the association (in
> properties
> view, it does NOT have any association set)
> 2. Association::views : View, aggregation = COMPOSITE, that is show as
> link
> source with diamond decoration.
> 3. Association::<unnamed> : Catalog, aggregation = NONE, that is shown as
> link target
>
> I am almost sure that (1) is not related to the association, because we
> EXPLICITLY
> FILTERING
> properties that are parts of binary associations from being shown in
> attributes
> compartment.
>
> Please check your model resource, and provide me with actual model data in
> case I am wrong here.
>
> Also please note that we have recently implemented the set of actions
> (#182749)
> that helps to tranfsorm
> single or paired set of class properties typed by different classifiers
> into
> the association between them.
> I believe that it will simplify the creation of the sample picture
> attached
> avoiding possible confusion with Catalog::views and Association::views.
>
> Does it make sense?
>
> Regards,
> Michael
>
>
>> In the following case, a catalog has one or more views. I am just
>> wondering whether the diamond symbol is correctly placed. I think it
>> should be with the Catalog.
>>
>> Thanks
>> Pradeep
>
>


Re: diamod direction in case of association [message #588914 is a reply to message #470239] Tue, 01 May 2007 03:01 Go to previous message
Pradeep is currently offline Pradeep
Messages: 61
Registered: July 2009
Member
One more observation,

In the generated diagram, attributes are placed on both association links
and inside the respective classes.
But in the diagram on the webpage, we see attributes only on the links which
I think is less redundant.

Pradeep


"Pradeep" <pradeep@live.com> wrote in message
news:f16lg1$hq7$1@build.eclipse.org...
> Hi Michael,
>
> Thanks for the explanation. I think I should study more about
> associations. I am still quite confused.
> I attached the purchase order diagram that is generated using the editor.
> Please find it attached.
> The original one is available at the following URL.
>
> http://www.eclipse.org/modeling/mdt/uml2/docs/articles/Getti ng_Started_with_UML2/article.html
>
> If you compare these two diagrams, the diamond shapes are positioned
> exactly in the reverse order. I am not sure which one is correct now.
>
> Pradeep
>
>
> "Michael Golubev" <michael.golubev@borland.com> wrote in message
> news:80ef4340a5c18c9593dde7dba0c@news.eclipse.org...
>> Hello Pradeep,
>>
>> In general, the diamond symbol presents and location depends on the
>> corresponding propertie's aggregation value (Property#getAggregation()).
>> In prinicple it is possible to see the link with diamonds for _both_
>> ends,
>> in case if the aggregation value for both of property ends is COMPOSITE.
>>
>> In particular, I believe that in your model, there are 3 properties that
>> make sense
>>
>> 1. Catalog::views that is NOT actually part of the association (in
>> properties
>> view, it does NOT have any association set)
>> 2. Association::views : View, aggregation = COMPOSITE, that is show as
>> link
>> source with diamond decoration.
>> 3. Association::<unnamed> : Catalog, aggregation = NONE, that is shown as
>> link target
>>
>> I am almost sure that (1) is not related to the association, because we
>> EXPLICITLY
>> FILTERING
>> properties that are parts of binary associations from being shown in
>> attributes
>> compartment.
>>
>> Please check your model resource, and provide me with actual model data
>> in
>> case I am wrong here.
>>
>> Also please note that we have recently implemented the set of actions
>> (#182749)
>> that helps to tranfsorm
>> single or paired set of class properties typed by different classifiers
>> into
>> the association between them.
>> I believe that it will simplify the creation of the sample picture
>> attached
>> avoiding possible confusion with Catalog::views and Association::views.
>>
>> Does it make sense?
>>
>> Regards,
>> Michael
>>
>>
>>> In the following case, a catalog has one or more views. I am just
>>> wondering whether the diamond symbol is correctly placed. I think it
>>> should be with the Catalog.
>>>
>>> Thanks
>>> Pradeep
>>
>>
>
>
>
Re: diamod direction in case of association [message #588935 is a reply to message #470245] Wed, 02 May 2007 05:17 Go to previous message
Michael Golubev is currently offline Michael Golubev
Messages: 373
Registered: July 2009
Senior Member
Hello Pradeep,

Thank you for persistence in convinicing me, I see that I was wrong claiming
this as implemented correctly.
Just reporoduced the issue, submitted bugzillas #185045 and #185046.

Regards,
Michael


> One more observation,
>
> In the generated diagram, attributes are placed on both association
> links
> and inside the respective classes.
> But in the diagram on the webpage, we see attributes only on the links
> which
> I think is less redundant.
> Pradeep
>
> "Pradeep" <pradeep@live.com> wrote in message
> news:f16lg1$hq7$1@build.eclipse.org...
>
>> Hi Michael,
>>
>> Thanks for the explanation. I think I should study more about
>> associations. I am still quite confused.
>> I attached the purchase order diagram that is generated using the
>> editor.
>> Please find it attached.
>> The original one is available at the following URL.
>> http://www.eclipse.org/modeling/mdt/uml2/docs/articles/Getti ng_Starte
>> d_with_UML2/article.html
>>
>> If you compare these two diagrams, the diamond shapes are positioned
>> exactly in the reverse order. I am not sure which one is correct now.
>>
>> Pradeep
>>
>> "Michael Golubev" <michael.golubev@borland.com> wrote in message
>> news:80ef4340a5c18c9593dde7dba0c@news.eclipse.org...
>>
>>> Hello Pradeep,
>>>
>>> In general, the diamond symbol presents and location depends on the
>>> corresponding propertie's aggregation value
>>> (Property#getAggregation()).
>>> In prinicple it is possible to see the link with diamonds for _both_
>>> ends,
>>> in case if the aggregation value for both of property ends is
>>> COMPOSITE.
>>> In particular, I believe that in your model, there are 3 properties
>>> that make sense
>>>
>>> 1. Catalog::views that is NOT actually part of the association (in
>>> properties
>>> view, it does NOT have any association set)
>>> 2. Association::views : View, aggregation = COMPOSITE, that is show
>>> as
>>> link
>>> source with diamond decoration.
>>> 3. Association::<unnamed> : Catalog, aggregation = NONE, that is
>>> shown as
>>> link target
>>> I am almost sure that (1) is not related to the association, because
>>> we
>>> EXPLICITLY
>>> FILTERING
>>> properties that are parts of binary associations from being shown in
>>> attributes
>>> compartment.
>>> Please check your model resource, and provide me with actual model
>>> data
>>> in
>>> case I am wrong here.
>>> Also please note that we have recently implemented the set of
>>> actions
>>> (#182749)
>>> that helps to tranfsorm
>>> single or paired set of class properties typed by different
>>> classifiers
>>> into
>>> the association between them.
>>> I believe that it will simplify the creation of the sample picture
>>> attached
>>> avoiding possible confusion with Catalog::views and
>>> Association::views.
>>> Does it make sense?
>>>
>>> Regards,
>>> Michael
>>>> In the following case, a catalog has one or more views. I am just
>>>> wondering whether the diamond symbol is correctly placed. I think
>>>> it should be with the Catalog.
>>>>
>>>> Thanks
>>>> Pradeep
Re: diamod direction in case of association [message #588972 is a reply to message #470249] Wed, 02 May 2007 18:13 Go to previous message
Pradeep is currently offline Pradeep
Messages: 61
Registered: July 2009
Member
:)

"Michael Golubev" <michael.golubev@borland.com> wrote in message
news:80ef4340aa3f8c95ac2a94927c8@news.eclipse.org...
> Hello Pradeep,
>
> Thank you for persistence in convinicing me, I see that I was wrong
> claiming this as implemented correctly.
> Just reporoduced the issue, submitted bugzillas #185045 and #185046.
> Regards, Michael
>
>
>> One more observation,
>>
>> In the generated diagram, attributes are placed on both association
>> links
>> and inside the respective classes.
>> But in the diagram on the webpage, we see attributes only on the links
>> which
>> I think is less redundant.
>> Pradeep
>>
>> "Pradeep" <pradeep@live.com> wrote in message
>> news:f16lg1$hq7$1@build.eclipse.org...
>>
>>> Hi Michael,
>>>
>>> Thanks for the explanation. I think I should study more about
>>> associations. I am still quite confused.
>>> I attached the purchase order diagram that is generated using the
>>> editor.
>>> Please find it attached.
>>> The original one is available at the following URL.
>>> http://www.eclipse.org/modeling/mdt/uml2/docs/articles/Getti ng_Starte
>>> d_with_UML2/article.html
>>>
>>> If you compare these two diagrams, the diamond shapes are positioned
>>> exactly in the reverse order. I am not sure which one is correct now.
>>>
>>> Pradeep
>>>
>>> "Michael Golubev" <michael.golubev@borland.com> wrote in message
>>> news:80ef4340a5c18c9593dde7dba0c@news.eclipse.org...
>>>
>>>> Hello Pradeep,
>>>>
>>>> In general, the diamond symbol presents and location depends on the
>>>> corresponding propertie's aggregation value
>>>> (Property#getAggregation()).
>>>> In prinicple it is possible to see the link with diamonds for _both_
>>>> ends,
>>>> in case if the aggregation value for both of property ends is
>>>> COMPOSITE.
>>>> In particular, I believe that in your model, there are 3 properties
>>>> that make sense
>>>>
>>>> 1. Catalog::views that is NOT actually part of the association (in
>>>> properties
>>>> view, it does NOT have any association set)
>>>> 2. Association::views : View, aggregation = COMPOSITE, that is show
>>>> as
>>>> link
>>>> source with diamond decoration.
>>>> 3. Association::<unnamed> : Catalog, aggregation = NONE, that is
>>>> shown as
>>>> link target
>>>> I am almost sure that (1) is not related to the association, because
>>>> we
>>>> EXPLICITLY
>>>> FILTERING
>>>> properties that are parts of binary associations from being shown in
>>>> attributes
>>>> compartment.
>>>> Please check your model resource, and provide me with actual model
>>>> data
>>>> in
>>>> case I am wrong here.
>>>> Also please note that we have recently implemented the set of
>>>> actions
>>>> (#182749)
>>>> that helps to tranfsorm
>>>> single or paired set of class properties typed by different
>>>> classifiers
>>>> into
>>>> the association between them.
>>>> I believe that it will simplify the creation of the sample picture
>>>> attached
>>>> avoiding possible confusion with Catalog::views and
>>>> Association::views.
>>>> Does it make sense?
>>>>
>>>> Regards,
>>>> Michael
>>>>> In the following case, a catalog has one or more views. I am just
>>>>> wondering whether the diamond symbol is correctly placed. I think
>>>>> it should be with the Catalog.
>>>>>
>>>>> Thanks
>>>>> Pradeep
>
>
Previous Topic:classes in packages.
Next Topic:classes in packages.
Goto Forum:
  


Current Time: Sat Aug 23 03:48:06 EDT 2014

Powered by FUDForum. Page generated in 0.02015 seconds