Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - Oliver F.

Pages: 1 ... 33 34 [35] 36 37 ... 39
Automation Interface, Add-Ins and Tools / Re: MDA newbie
« on: March 27, 2009, 11:12:26 pm »
The transform engine can't create tagged values on connectors at all.  Follow those instructions if you are wanting to add a tagged value to anything else.  It won't work for connectors.

Simon, I have to add that this is an unfortunate situation especially for those working with XML schemas.
Actually the only way to create a reference link in a XSD model is to add two tagged values to one of the aggregation or association end.
Otherwise the link is treated as a composite.
If the XSD elements are the result of a transformation those tags are getting lost and you have to set them otherwise (mostly manually) on the transformed elements.
The alternative is to use no transformation at all to create XSD files.

I have filed a bug report but the answer was to set the tagged values to accomplish what I want. And here I stay again. :-?


BTW: Is there javadoc from this eaapi.jar. (I find things out, by decompiling stuff....)

Simon said it all but for completeness: People who are used javadoc syntax will find it hard to get used to the API reference Sparx provides as it is generic to fit all supported technologies. It is not that straightforward.

In fact there even is no real JAVA implementation available as the existing API is just a wrapper for the COM objects under Windows.

Though I wonder whether it would be possible to create some useful javadoc from the wrapper API...


To generate a 'ref' attribute, open the tagged-values window for the connector and set the following tagged-values in the 'Connector Source' section :

1. Tag    = anonymousType
    Value = false

2. Tag    = anonymousRole
    Value = true

This does not sound like a serious workaround- imagine hundreds of connections to draw where we would have to set the stereotypes for each of those. OK, UML profiles help here a bit, but as even the online help does not give clues (though greatly showing the ref tag without telling how to accomplish this) it would be a good idea to either adapt the help or introduce a connector property (though it should work with a Reference connector type anyway).


Update: When applying a transformation those tagged values are lost. So this is definitely not an option, I am afraid.

So, in this situation in which i will need to rearrange the same components in several ways (each per document i would to generate), i think i will neccesary have to replicate all my modeldocuments and the easiest way is to copy diagram (Documentation Diagram) in "deep" mode, isn't it?

Apparently yes.
But I have also not tried working with instances of model documents. That could be an option though I doubt that it works. In this case it would be interesting to see whether instances of model documents are processed inside a master document.
If yes then we have the solution, if no, then nothing is lost (and neither won :) )


No that I know of. As the master document is a package its ModelDocument elements are physically attached in the model and the RTF generator traverses each.

What you can do is to create a package which includes artifacts and link that pacakge with several ModelDocuments but I am almost certain that this does not help you.


When you right-click on the profile in the panel "resources" there is a synchronization function "synch tagged values and constrains" that these tagged values add to elements using this profile. I think it's easiest to correct the problem.

And again, SparxSystems does not recommend importing the profile in the resource view but points to using a MDG technology file instead in which case the synchronisation feature is not available any more.


Just is created the Tag Values of the attributes if I drag-and-drop de Attribute (Stereotyped) indo Class.

This is true. You will also not get tagged values for a class if you create and afterwards manually set the stereotype to the one from your profile.

UML profiles seem to be applied only when used via the toolbox.


I save the Profile and after I import into “Resources”

This is the problematic part *sigh*
It does not work if you import the profile via the resources view but wrap the UML profile into a MDG technology file.


If I open a Repository through openFile...I have no Issues (not always at least), but I would like to get the Models for the open instance of EA.

As I said above- afaik you can not access an already open instance from the JAVA API.
I'd be glad if someone can prove me wrong here as I need that functionality, too.


From a quick glance at my code:

Code: [Select]

org.sparx.EARepository eaRepository = new org.sparx.Repository();
if (eaRepository == null)
                //error handling here...
                return false;

        if (!eaRepository.OpenFile(CONNECTION))
             //error handling here...
            return false;

Note that this will not give you access to a running instance but opens a new instance (defined in the connection string). AFAIK there is no way to access the running instance from JAVA via the DCOM bridge.


I am using the rtf-generator.
I am searching a practical way to have my UI included in my report, when I run my documentation of 1 or more Use-cases.
I would prefere to have my UI-diagrams in the User Interface Package.

You can include the user interface diagrams as diagram references below their corresponding use cases but this will not document the UI element properties.

The other option is to directly inlude UI diagrams and check the "Report diagram elements in report" option in the rtf generation dialog.

Second approach: Make yourself familiar with the virtual document concepts which lets can include the use case and UI packages in a single report using different templates. The online help is a good starting point here.


I suggest to read the report generation whitepaper from sparx which is a valuable source of information:

Especially the section about virtual documents might suit your needs.

If specific questions arise feel free to ask again.

Best regards,


7.1 includes various bugfixes and extensions to the automation API so I believe that falls under this category.

If possible upgrade to 7.1, it is worth the improvements anyway.

On the other hand you might try to skip the Iterator class and access the constraints through a for or while loop.

You could also post your code fragment in question here so we can have a look at it.


Automation Interface, Add-Ins and Tools / Re: RunReport on master document
« on: December 16, 2008, 07:45:35 pm »
Now it seems to me that EA is using the master document as normal package and ignores the virtual documents inside. Is there any way to generate a master document with the EA.Interop API?

Well, I find it very likely that this is true. What happens, if you omit the template parameter in the RunReport call?

If this does not change things I suggest to file a bug report. Virtual documents are a rather new feature and it propably it went forgotten in the automation interface.


Hmm...Triggering document generation from the master document level does bring up the dialog, where the template selection field is greyed out. BUT when I continiue, I only get a report describing the empty master document ;(

Any suggestions how to fix this ? (I'm still trying to get one report covering all virtual documents under one master doc.)

Do you have packages added to your document?
Is the "Skip root element" option disabled?

What is your structure below the master document?


Pages: 1 ... 33 34 [35] 36 37 ... 39