Home » Modeling » EMF » Error found in EMF 2.4, extension of map creates "The interface Entry cannot be implemented mor
| | |
| Re: Error found in EMF 2.4, extension of map creates "The interface Entry cannot be implemented [message #414708 is a reply to message #414707] |
Sun, 18 November 2007 18:16   |
 Philipp Kutter
Messages: 306
Registered: July 2009 |
Senior Member |
|
|
Perfect, this solved the problem. A corresponding validation constraint
might be added... shall I open a feature request?
You ask for the use of extending a map entry class. Here is our approach:
We treat strongly typed maps, as supported in EMF models as first class
citizens of our business models. Especially, since I can use containment
semantics for either of key or value, and since I can use the many-flag
to have values being lists. (I never tried out what a many-flagged
key-feature would give...)
Thanks to the mapping of maps to classes with key/value features, we can
now attach additional attributes or references to the map entries. And
often, there are different variants of these additional attributes or
references, which are then added to different subtypes of the map entry.
Does this make sense to you?
Best, Philipp
Ed Merks wrote:
> Philipp,
>
> I guess I didn't notice this posting for some reason...
>
> If you set instanceClassName to java.util.Map$Entry for
> MapEClass1Extension it works okay. I'm not sure what utility there is
> in extending a map entry EClass though mind you. What are you trying to
> accomplish by doing this?
>
>
> Philipp W. Kutter wrote:
>> Ed, shall I enter a bug report?
>>
>> Best, Philipp
>>
>>
>> Philipp W. Kutter wrote:
>>> Hi.
>>> In the included ZIP, an error is appearing.
>>>
>>> You can as well recreate it by creating a map with ecore in one project,
>>> and then creating an extension of the map in the next one.
>>>
>>> The full error is:
>>> Severity and Description Path Resource Location Creation
>>> Time Id
>>> The interface Entry cannot be implemented more than once with
>>> different arguments: Map.Entry<EClass2,EClass3> and Map.Entry
>>> TestMapsExtension/src/TestMapsExtension/impl
>>> MapEClass1ExtensionImpl.java line 25 1194980174734 13531
>>>
>>> Any idea what this is?
>>>
>>> Best, Philipp
>>>
|
|
|
| Re: Error found in EMF 2.4, extension of map creates "The interface Entry cannot be implemented [message #414710 is a reply to message #414708] |
Sun, 18 November 2007 20:35 |
Ed Merks
Messages: 33216
Registered: July 2009 |
Senior Member |
|
|
This is a multi-part message in MIME format.
--------------070605050407090609060306
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Philipp,
Please add your test case to 207449
<https://bugs.eclipse.org/bugs/show_bug.cgi?id=207449>
It does make sense, though generally if a client simply uses the EMap's
put to add things, just a base entry will be created...
Philipp W. Kutter wrote:
> Perfect, this solved the problem. A corresponding validation
> constraint might be added... shall I open a feature request?
>
>
> You ask for the use of extending a map entry class. Here is our approach:
>
> We treat strongly typed maps, as supported in EMF models as first class
> citizens of our business models. Especially, since I can use containment
> semantics for either of key or value, and since I can use the
> many-flag to have values being lists. (I never tried out what a
> many-flagged key-feature would give...)
>
> Thanks to the mapping of maps to classes with key/value features, we
> can now attach additional attributes or references to the map entries.
> And
> often, there are different variants of these additional attributes or
> references, which are then added to different subtypes of the map entry.
>
> Does this make sense to you?
>
> Best, Philipp
>
>
> Ed Merks wrote:
>> Philipp,
>>
>> I guess I didn't notice this posting for some reason...
>>
>> If you set instanceClassName to java.util.Map$Entry for
>> MapEClass1Extension it works okay. I'm not sure what utility there
>> is in extending a map entry EClass though mind you. What are you
>> trying to accomplish by doing this?
>>
>>
>> Philipp W. Kutter wrote:
>>> Ed, shall I enter a bug report?
>>>
>>> Best, Philipp
>>>
>>>
>>> Philipp W. Kutter wrote:
>>>> Hi.
>>>> In the included ZIP, an error is appearing.
>>>>
>>>> You can as well recreate it by creating a map with ecore in one
>>>> project,
>>>> and then creating an extension of the map in the next one.
>>>>
>>>> The full error is:
>>>> Severity and Description Path Resource Location
>>>> Creation Time Id
>>>> The interface Entry cannot be implemented more than once with
>>>> different arguments: Map.Entry<EClass2,EClass3> and Map.Entry
>>>> TestMapsExtension/src/TestMapsExtension/impl
>>>> MapEClass1ExtensionImpl.java line 25 1194980174734 13531
>>>>
>>>> Any idea what this is?
>>>>
>>>> Best, Philipp
>>>>
--------------070605050407090609060306
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
Philipp,<br>
<br>
Please add your test case to <a name="b207449"
href="https://bugs.eclipse.org/bugs/show_bug.cgi?id=207449">207449</a><br>
<br>
It does make sense, though generally if a client simply uses the EMap's
put to add things, just a base entry will be created...<br>
<br>
<br>
Philipp W. Kutter wrote:
<blockquote cite="mid:fhpvg5$m9t$1@build.eclipse.org" type="cite">Perfect,
this solved the problem. A corresponding validation constraint might be
added... shall I open a feature request?
<br>
<br>
<br>
You ask for the use of extending a map entry class. Here is our
approach:
<br>
<br>
We treat strongly typed maps, as supported in EMF models as first class
<br>
citizens of our business models. Especially, since I can use
containment
<br>
semantics for either of key or value, and since I can use the many-flag
to have values being lists. (I never tried out what a many-flagged
key-feature would give...)
<br>
<br>
Thanks to the mapping of maps to classes with key/value features, we
can now attach additional attributes or references to the map entries.
And
<br>
often, there are different variants of these additional attributes or
references, which are then added to different subtypes of the map
entry.
<br>
<br>
Does this make sense to you?
<br>
<br>
Best, Philipp
<br>
<br>
<br>
Ed Merks wrote:
<br>
<blockquote type="cite">Philipp,
<br>
<br>
I guess I didn't notice this posting for some reason...
<br>
<br>
If you set instanceClassName to java.util.Map$Entry for
MapEClass1Extension it works okay. I'm not sure what utility there is
in extending a map entry EClass though mind you. What are you trying
to accomplish by doing this?
<br>
<br>
<br>
Philipp W. Kutter wrote:
<br>
<blockquote type="cite">Ed, shall I enter a bug report?
<br>
<br>
Best, Philipp
<br>
<br>
<br>
Philipp W. Kutter wrote:
<br>
<blockquote type="cite">Hi.
<br>
In the included ZIP, an error is appearing.
<br>
<br>
You can as well recreate it by creating a map with ecore in one
project,
<br>
and then creating an extension of the map in the next one.
<br>
<br>
The full error is:
<br>
Severity and Description Path Resource Location Creation
Time Id
<br>
The interface Entry cannot be implemented more than once with different
arguments: Map.Entry<EClass2,EClass3> and Map.Entry
TestMapsExtension/src/TestMapsExtension/impl
MapEClass1ExtensionImpl.java line 25 1194980174734 13531
<br>
<br>
Any idea what this is?
<br>
<br>
Best, Philipp
<br>
<br>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
<br>
</body>
</html>
--------------070605050407090609060306--
Ed Merks
Professional Support: https://www.macromodeling.com/
|
|
|
| Re: Error found in EMF 2.4, extension of map creates "The interface Entry cannot be implemented [message #414749 is a reply to message #414710] |
Tue, 20 November 2007 15:35 |
Philipp Kutter
Messages: 306
Registered: July 2009 |
Senior Member |
|
|
Done.
Ed Merks wrote:
> Philipp,
>
> Please add your test case to 207449
> <https://bugs.eclipse.org/bugs/show_bug.cgi?id=207449>
>
> It does make sense, though generally if a client simply uses the EMap's
> put to add things, just a base entry will be created...
>
>
> Philipp W. Kutter wrote:
>> Perfect, this solved the problem. A corresponding validation
>> constraint might be added... shall I open a feature request?
>>
>>
>> You ask for the use of extending a map entry class. Here is our approach:
>>
>> We treat strongly typed maps, as supported in EMF models as first class
>> citizens of our business models. Especially, since I can use containment
>> semantics for either of key or value, and since I can use the
>> many-flag to have values being lists. (I never tried out what a
>> many-flagged key-feature would give...)
>>
>> Thanks to the mapping of maps to classes with key/value features, we
>> can now attach additional attributes or references to the map entries.
>> And
>> often, there are different variants of these additional attributes or
>> references, which are then added to different subtypes of the map entry.
>>
>> Does this make sense to you?
>>
>> Best, Philipp
>>
>>
>> Ed Merks wrote:
>>> Philipp,
>>>
>>> I guess I didn't notice this posting for some reason...
>>>
>>> If you set instanceClassName to java.util.Map$Entry for
>>> MapEClass1Extension it works okay. I'm not sure what utility there
>>> is in extending a map entry EClass though mind you. What are you
>>> trying to accomplish by doing this?
>>>
>>>
>>> Philipp W. Kutter wrote:
>>>> Ed, shall I enter a bug report?
>>>>
>>>> Best, Philipp
>>>>
>>>>
>>>> Philipp W. Kutter wrote:
>>>>> Hi.
>>>>> In the included ZIP, an error is appearing.
>>>>>
>>>>> You can as well recreate it by creating a map with ecore in one
>>>>> project,
>>>>> and then creating an extension of the map in the next one.
>>>>>
>>>>> The full error is:
>>>>> Severity and Description Path Resource Location
>>>>> Creation Time Id
>>>>> The interface Entry cannot be implemented more than once with
>>>>> different arguments: Map.Entry<EClass2,EClass3> and Map.Entry
>>>>> TestMapsExtension/src/TestMapsExtension/impl
>>>>> MapEClass1ExtensionImpl.java line 25 1194980174734 13531
>>>>>
>>>>> Any idea what this is?
>>>>>
>>>>> Best, Philipp
>>>>>
>
|
|
|
Goto Forum:
Current Time: Sun Sep 22 07:31:36 GMT 2024
Powered by FUDForum. Page generated in 0.04050 seconds |
] 
Search
Help
Register
Login
Home
