Oh, I didn't answer you first question.

I'll be developing this initially just for the UML2 activity diagram.

Bob

On 12 April 2011 11:13, Bob Tarling <[email protected]> wrote:
> Thomas said:
>> One question on NotatedItem: what if the fig is not associated with a 
>> metatype, but with just an
>> attribute of a metatype?
>
> That is where the NotationType enumeration comes in. Both the metatype
> and notation type are supplied.
>
> Mostly the notation type will be NAME but as an example for an
> AssociationEnd there will be a NAME and a MULTIPLICITY
>
> Bob
>
> On 12 April 2011 10:59, Thomas Neustupny <[email protected]> wrote:
>> Hi Bob,
>>
>> moving notation code to modules with the architecture you proposed and 
>> thereby minimize the notation related events is great. Let's see how things 
>> work when you try it for a new diagram type. Which one are you thinking of?
>>
>> One question on NotatedItem: what if the fig is not associated with a 
>> metatype, but with just an attribute of a metatype?
>>
>> Thomas
>>
>> -------- Original-Nachricht --------
>>> Datum: Thu, 7 Apr 2011 20:30:18 +0100
>>> Von: Bob Tarling <[email protected]>
>>> An: [email protected]
>>> Betreff: Re: [argouml-dev] Moving UML2 activity diagrams forward
>>
>>> Hi Mark
>>>
>>> Notations are nothing new, it is just this implementation that is new.
>>>
>>> Try selecting File->Notations->Java and dropping some classes on a
>>> diagram.
>>>
>>> However what we don't currently have are notations implemented for the
>>> new UML2 diagrams.
>>>
>>> There has been talk of refactoring the notations for some time but I'd
>>> prefer to implement a brand new diagram based on fresh principles and
>>> not waiting for the refactoring to catch up.
>>>
>>> The problem many Figs have at the comment is that they redraw too
>>> frequently partly because they listen to far too many things or redraw
>>> completely instead of only partly redrawing..
>>>
>>> With these new diagrams I'm trying to get a better architecture so
>>> that Fig listen to the minimum they need to. For that the notations
>>> also need to fire less.
>>>
>>> If I can prove this method for a new diagram type then there is no
>>> chance of me breaking the notation in the process for our existing
>>> diagrams.
>>>
>>> As other diagrams begin to become modules of their own they can
>>> migrate at that stage to the new notation system.
>>>
>>> Regards
>>>
>>> Bob
>>>
>>>
>>>
>>> On 7 April 2011 17:37, Mark Fortner <[email protected]> wrote:
>>> > Hi Bob,
>>> > Just to be clear, this framework is meant to store the information about
>>> the
>>> > items in the diagram
>>> > (typically the data that we currently find in the XMI file) is that
>>> right?
>>> > And would there still be the same degree of correspondence between the
>>> > diagrams and the
>>> > XMI file that exists today?  By that I mean that if we see a Fig whose
>>> ID is
>>> > 123 then the
>>> > XMI file will also contain an item whose ID is 123?
>>> > You describe the NotationLanguage interface as "The interface that any
>>> > notation languages should
>>> > implement (e.g. UML, Java etc)."  I assume that the NotationLanguage
>>> for
>>> > Java would mean
>>> > some Java-specific notation like perhaps "annotations" like
>>> > "@PostInitialization" or "@Test".
>>> > Is that correct?
>>> > How would NotationLanguages be registered with the NotationUtility?
>>>  Would
>>> > this be through the
>>> > Services API (or some other discovery mechanism), or hard-coded?
>>> > Regards,
>>> > Mark
>>> >
>>> > On Thu, Apr 7, 2011 at 7:41 AM, Bob Tarling <[email protected]>
>>> wrote:
>>> >>
>>> >> For UML2 activity diagrams I halted for a while thinking about how to
>>> >> apply notations. I think the notations are an incredibly important
>>> >> subsystem as they remove knowledge of both GEF and so will help with
>>> >> any move to a different diagram implementation should it be required
>>> >> in future.
>>> >>
>>> >> I have prepared something on the wiki regarding how I would like to
>>> >> see notations split out into its own module and improved -
>>> >> http://argouml.tigris.org/wiki/Notation_Architecture
>>> >>
>>> >> I'd appreciate any feedback.
>>> >>
>>> >> I propose to leave the old notation architecture in place while this
>>> >> is developed for the UML2 activity diagram only.
>>> >>
>>> >> Once this is working I'll move the state diagram forward to a similar
>>> >> level of functionality and move any common code for state and activity
>>> >> into another shared module.
>>> >>
>>> >> Regards
>>> >>
>>> >> Bob
>>> >>
>>> >> ------------------------------------------------------
>>> >>
>>> >>
>>> http://argouml.tigris.org/ds/viewMessage.do?dsForumId=450&dsMessageId=2717692
>>> >>
>>> >> To unsubscribe from this discussion, e-mail:
>>> >> [[email protected]].
>>> >> To be allowed to post to the list contact the mailing list moderator,
>>> >> email: [[email protected]]
>>> >
>>> >
>>>
>>> ------------------------------------------------------
>>> http://argouml.tigris.org/ds/viewMessage.do?dsForumId=450&dsMessageId=2717786
>>>
>>> To unsubscribe from this discussion, e-mail:
>>> [[email protected]].
>>> To be allowed to post to the list contact the mailing list moderator,
>>> email: [[email protected]]
>>
>> --
>> Empfehlen Sie GMX DSL Ihren Freunden und Bekannten und wir
>> belohnen Sie mit bis zu 50,- Euro! https://freundschaftswerbung.gmx.de
>>
>> ------------------------------------------------------
>> http://argouml.tigris.org/ds/viewMessage.do?dsForumId=450&dsMessageId=2719047
>>
>> To unsubscribe from this discussion, e-mail: 
>> [[email protected]].
>> To be allowed to post to the list contact the mailing list moderator, email: 
>> [[email protected]]
>>
>

------------------------------------------------------
http://argouml.tigris.org/ds/viewMessage.do?dsForumId=450&dsMessageId=2719051

To unsubscribe from this discussion, e-mail: 
[[email protected]].
To be allowed to post to the list contact the mailing list moderator, email: 
[[email protected]]

Reply via email to