Skip to main content


Eclipse Community Forums
Forum Search:

Search      Help    Register    Login    Home
Home » Modeling » UML2 » Should ValueSpecification be really a TypedElement?
Should ValueSpecification be really a TypedElement? [message #477567] Thu, 03 July 2008 12:42 Go to next message
Timothy Marc is currently offline Timothy MarcFriend
Messages: 547
Registered: July 2009
Senior Member
Hi all,

after studying the values and instances concepts of the UML2 superstructure,
i was wondering that a ValueSpecification is also inhereting TypedElement.
What's the main reason for this? In the case of being a
LiteralSpecification, its type is directly combined with the value argument.
In the case of beeing a InstanceValue, the instance references a
InstanceSpecification, that points to a concrete classifier.
IMHO, there is no sense in defining the ValueSpecification as a
TypedElement, or is it for the enhancing mechanism of the metamodel?

Hope, this question isn't too silly.

-- Timothy
Re: Should ValueSpecification be really a TypedElement? [message #477568 is a reply to message #477567] Thu, 03 July 2008 13:33 Go to previous message
Eclipse UserFriend
Originally posted by: cdamus.zeligsoft.com

--=-9KySbOAP3Q+PoKvj1WU7
Content-Type: text/plain
Content-Transfer-Encoding: 7bit

Hi, Timothy,

For me, the TypedElement-ness of ValueSpecification is most valuable
(tee hee) in the OpaqueExpression. Here, you can constrain the type for
expressions that otherwise are not interpreted by your tool.

Otherwise, for InstanceValues, there is the problem of the fact that
InstanceSpecifications can have multiple classifiers, which doesn't
match very well with the single type of a TypedElement. However, the
type of a TypedElement is optional, so I expect that often the right
thing to do is just to leave it blank.

HTH,

Christian


On Thu, 2008-07-03 at 14:42 +0200, Timothy Marc wrote:

> Hi all,
>
> after studying the values and instances concepts of the UML2 superstructure,
> i was wondering that a ValueSpecification is also inhereting TypedElement.
> What's the main reason for this? In the case of being a
> LiteralSpecification, its type is directly combined with the value argument.
> In the case of beeing a InstanceValue, the instance references a
> InstanceSpecification, that points to a concrete classifier.
> IMHO, there is no sense in defining the ValueSpecification as a
> TypedElement, or is it for the enhancing mechanism of the metamodel?
>
> Hope, this question isn't too silly.
>
> -- Timothy
>
>

--=-9KySbOAP3Q+PoKvj1WU7
Content-Type: text/html; charset=utf-8

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 TRANSITIONAL//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; CHARSET=UTF-8">
<META NAME="GENERATOR" CONTENT="GtkHTML/3.16.0">
</HEAD>
<BODY>
Hi, Timothy,<BR>
<BR>
For me, the TypedElement-ness of ValueSpecification is most valuable (tee hee) in the OpaqueExpression.&nbsp; Here, you can constrain the type for expressions that otherwise are not interpreted by your tool.<BR>
<BR>
Otherwise, for InstanceValues, there is the problem of the fact that InstanceSpecifications can have multiple classifiers, which doesn't match very well with the single type of a TypedElement.&nbsp; However, the type of a TypedElement is optional, so I expect that often the right thing to do is just to leave it blank.<BR>
<BR>
HTH,<BR>
<BR>
Christian<BR>
<BR>
<BR>
On Thu, 2008-07-03 at 14:42 +0200, Timothy Marc wrote:
<BLOCKQUOTE TYPE=CITE>
<PRE>
<FONT COLOR="#000000">Hi all,</FONT>

<FONT COLOR="#000000">after studying the values and instances concepts of the UML2 superstructure, </FONT>
<FONT COLOR="#000000">i was wondering that a ValueSpecification is also inhereting TypedElement. </FONT>
<FONT COLOR="#000000">What's the main reason for this? In the case of being a </FONT>
<FONT COLOR="#000000">LiteralSpecification, its type is directly combined with the value argument. </FONT>
<FONT COLOR="#000000">In the case of beeing a InstanceValue, the instance references a </FONT>
<FONT COLOR="#000000">InstanceSpecification, that points to a concrete classifier.</FONT>
<FONT COLOR="#000000">IMHO, there is no sense in defining the ValueSpecification as a </FONT>
<FONT COLOR="#000000">TypedElement, or is it for the enhancing mechanism of the metamodel?</FONT>

<FONT COLOR="#000000">Hope, this question isn't too silly.</FONT>

<FONT COLOR="#000000">-- Timothy </FONT>


</PRE>
</BLOCKQUOTE>
</BODY>
</HTML>

--=-9KySbOAP3Q+PoKvj1WU7--
Re: Should ValueSpecification be really a TypedElement? [message #626759 is a reply to message #477567] Thu, 03 July 2008 13:33 Go to previous message
Eclipse UserFriend
Originally posted by: cdamus.zeligsoft.com

--=-9KySbOAP3Q+PoKvj1WU7
Content-Type: text/plain
Content-Transfer-Encoding: 7bit

Hi, Timothy,

For me, the TypedElement-ness of ValueSpecification is most valuable
(tee hee) in the OpaqueExpression. Here, you can constrain the type for
expressions that otherwise are not interpreted by your tool.

Otherwise, for InstanceValues, there is the problem of the fact that
InstanceSpecifications can have multiple classifiers, which doesn't
match very well with the single type of a TypedElement. However, the
type of a TypedElement is optional, so I expect that often the right
thing to do is just to leave it blank.

HTH,

Christian


On Thu, 2008-07-03 at 14:42 +0200, Timothy Marc wrote:

> Hi all,
>
> after studying the values and instances concepts of the UML2 superstructure,
> i was wondering that a ValueSpecification is also inhereting TypedElement.
> What's the main reason for this? In the case of being a
> LiteralSpecification, its type is directly combined with the value argument.
> In the case of beeing a InstanceValue, the instance references a
> InstanceSpecification, that points to a concrete classifier.
> IMHO, there is no sense in defining the ValueSpecification as a
> TypedElement, or is it for the enhancing mechanism of the metamodel?
>
> Hope, this question isn't too silly.
>
> -- Timothy
>
>

--=-9KySbOAP3Q+PoKvj1WU7
Content-Type: text/html; charset=utf-8

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 TRANSITIONAL//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; CHARSET=UTF-8">
<META NAME="GENERATOR" CONTENT="GtkHTML/3.16.0">
</HEAD>
<BODY>
Hi, Timothy,<BR>
<BR>
For me, the TypedElement-ness of ValueSpecification is most valuable (tee hee) in the OpaqueExpression.&nbsp; Here, you can constrain the type for expressions that otherwise are not interpreted by your tool.<BR>
<BR>
Otherwise, for InstanceValues, there is the problem of the fact that InstanceSpecifications can have multiple classifiers, which doesn't match very well with the single type of a TypedElement.&nbsp; However, the type of a TypedElement is optional, so I expect that often the right thing to do is just to leave it blank.<BR>
<BR>
HTH,<BR>
<BR>
Christian<BR>
<BR>
<BR>
On Thu, 2008-07-03 at 14:42 +0200, Timothy Marc wrote:
<BLOCKQUOTE TYPE=CITE>
<PRE>
<FONT COLOR="#000000">Hi all,</FONT>

<FONT COLOR="#000000">after studying the values and instances concepts of the UML2 superstructure, </FONT>
<FONT COLOR="#000000">i was wondering that a ValueSpecification is also inhereting TypedElement. </FONT>
<FONT COLOR="#000000">What's the main reason for this? In the case of being a </FONT>
<FONT COLOR="#000000">LiteralSpecification, its type is directly combined with the value argument. </FONT>
<FONT COLOR="#000000">In the case of beeing a InstanceValue, the instance references a </FONT>
<FONT COLOR="#000000">InstanceSpecification, that points to a concrete classifier.</FONT>
<FONT COLOR="#000000">IMHO, there is no sense in defining the ValueSpecification as a </FONT>
<FONT COLOR="#000000">TypedElement, or is it for the enhancing mechanism of the metamodel?</FONT>

<FONT COLOR="#000000">Hope, this question isn't too silly.</FONT>

<FONT COLOR="#000000">-- Timothy </FONT>


</PRE>
</BLOCKQUOTE>
</BODY>
</HTML>

--=-9KySbOAP3Q+PoKvj1WU7--
Previous Topic:Should ValueSpecification be really a TypedElement?
Next Topic:Unable to wrap long names... and other n00b questions :)
Goto Forum:
  


Current Time: Sat Aug 20 03:19:09 GMT 2022

Powered by FUDForum. Page generated in 0.02988 seconds
.:: Contact :: Home ::.

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

Back to the top