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 - P Robert

Pages: [1]
1
Thanks for your answers.
1) Qwerty suggested that I should open a bug, am I not in the bugs and issues forum? Is there another way to "submit a bug"?
Yes indeed it is not a high priority bug (I mean the soft still works) it just makes my diagram look less professional am I'm the picky kind ;)

2) Eve: you said if you "autosize" it regulates the size (to the bigger one which is fine for me as long as they all have the same size): can you explain how you do that? Pretty sure it is less of a pain than manually sizing each :)

Thanks

2
Who said it was the same element twice? Look at the screenshot I provided: it is two different elements, created in two different ways (one from the toolbox and another previously existing one dragged from the project browser). Fact is I'm not duplicating items if they already exist in a previous diagram, I reuse them (which is the right way to use building blocks). And yes in some diagram I may have more than one "input" therefore more than one start event, may it be a good or bad BPMN practice is off topic (and so what if it helps me illustrate my message well...). But regardless of the type of objects, it does the same for gateways, end event and most of the BPMN shapes and it is annoying.

3
Hi,

I'm writing BPMN process daigrams and wondering if we can determine where the name of an arrow is displayed on the diagram? Currently it is next to the destination end of the arrow (on sequence flow arrows) and was serching for a way to show them next to the source end of the arrow instead? This would make it more readable when using gateway elements since the "yes" and "no" text would be next to the gateway element rather than at the end of potentially long sequence flow arrows...

suggestion: it should be that way by default when writing BPMN diagrams... ;)

Thanks

4
Hi,

I'm using Sparx 15 (latest version) and creating a new BPMN 2.0 business process diagram.
When I create a new item (dragged from the toolbox directly into a diagram) the item shape has a default size but if I drag an existing item from the item browser into the same diagram it shows bigger than originally set when created from the toolbox. That is super annoying as I'm constantly resizing stuff to make it constant... What's the issue?
Example:
https://drive.google.com/file/d/1ZnfrBjjx-0CGfgRH1lLjAM9oSUWjDZHA/view?usp=sharing



5
Just to close the loop, I've completed parsing the XML, retrieving the metaclass and creating the connector programatically.
That almost did the trick: stereotype was correctly applied but some other properties of the connector were not correctly set. Found out that in the "apply" XML tags there were some default property values configured.

<AppliesTo>
<Apply type="Association">
<Property name="direction" value="Unspecified"/>
<Property name="compositionKind" value="Aggregate at Source"/>
<Property name="_MeaningForwards" value="Aggregation of"/>
<Property name="_MeaningBackwards" value="Aggregates"/>
</Apply>
</AppliesTo>

I found out through the MDG Technology SDK about those Special Attributes (https://sparxsystems.com/enterprise_architect_user_guide/14.0/modeling_tools/supportedattributes_2.html) that could be defined and had to figure out for those that I were interested in how to apply them to the connector.

So more investigation than initially planned but it does work now  :)

Thanks for your help!

6
No, there is no better way.
Either you parse the XML, or you hardcode all the info you need.

Or, and this is a long shot, you might be able to use the fully qualified stereotype in the AddNew() operations as the type parameter.
So something like
Code: [Select]
myConnector = myElement.connectors.AddNew("myName","Archimate2::ArchiMate_UsedBy");(double check to make sure the syntax and names are correct)
Then you can check the type of the just created connector.

Not sure if this works, but it's worth a try.

Geert

Just as an FYI, this does not work. Would have been great though!

7
The user specified locations are indeed stored in the registry.
The others are "fixed"

The problem might reading the technologies imported into the model (t_document) and the ones in the internalTechnologies.
You'll have to decode the format first.
The others are plain xml so you could read those, but I'm not convinced parsing the mdg xml's is the best approach.

Geert

Would you happen to know which registry key I'm supposed to look at? :)

Thanks

Phil

8
Well there's...
  • MDGTechnologies folder
  • InternalTechnologies folder (encrypted)
  • The user's APPDATA folder
  • Stored internally in the model
  • On any one of the alternate paths that the user has defined
  • At a specified URL
  • Provided by an add-in at start-up
  • Anything I've missed...?

Yeah there are many possibilities, my wish is to understand how EA does figure out where to access the file so that I can do the same.
Any idea on how this is done? Since EA does load those files on startup, there must be a reference to the file location in a registry or in the database or in a one of EA's config file or the like...?

Thanks

Philippe

9
Hi,

I'm building a framework which would work for every type of the technology.
Parsing the XML file is no issue I do have handy APIs for that ;)
However I would like to be able to determine the location of the XML file programatically so that I can avoid hardcoding or putting in a config file the XML file's path.
If you know of a way to determine this it would be great.

Thanks

Phil

10
Thanks. That's what I figured out but wanted to be sure before I start parsing the XML. Definitely be something that should be added as part of the automation interface.
Do you know of the best way to determine the XML file location then?
I wouldn't want to "hardcode" the path even though it is probably a pretty standard one unless you changed the install folder...

11
Hi,

I'm trying to create a connector using an MDG Technology stereotype (Archimate in my case).
The stereotype defined in the MDG technology "maps" to a base connector type in the MDG tech XML file.

Example:
<Stereotype notes="" name="ArchiMate_UsedBy" hideicon="0" borderwidth="-1" bordercolor="-1" fontcolor="-1" bgcolor="-1" metatype="UsedBy">
  [...]
  -<AppliesTo>
    -<Apply type="Dependency">
      [...]
    </Apply>
  </AppliesTo>
</Stereotype>

So (if I'm getting this right) when creating a connector with the "Archimate_UsedBy" stereotype EA will create a "Dependency" type connector.
I want to be able to set the right "base type" (Dependency) when creating a new connector based on the "stereotype" (Archimate_UsedBy).

Problem is, this mapping does seem to exist only in the MDG XML file, so it cannot be accessed (or at least I haven't figured out how) from the automation interface... Unless I parse the XML myself, which seems to be troublesome (would need first to figure out where the XML is, then parse it manually :-\)

Before I jump into this I was wondering if there's a cleaner way to achieve getting access to the MDG technology info from the automation interface?

Thanks

Phil

12
General Board / Archimate elements used multiple times
« on: June 12, 2013, 05:29:17 am »
Hi,

I'm trying to put together a Archimate 2.0 diagram in which I would like to have multiple instances of the same element.

A good example is a data element: I have multiple systems in my whole IT landscape which use the part data object. So for each system, I would like to display that the software either realizes (create) or uses (consume a local copy of the data) the part data object.
However, when I try to drag the Parts data object more than once on my diagram, I'm getting an error message saying:

 "The diagram already contains an instance of the element you are trying to paste. Currently, only one instance is supported, so you cannot paste the element here."

I don't want to have the element only once and linking all systems to it for 2 reasons:
1) it will quickly become unreadable
2) I want to be able to show the difference between a system which gets a local copy of the data and system which accesses the part data directly into another system

I've seen that I can create local copies of an linked element which allows me to have it multiple times, but they really are distinct element afterward... This does pollute the model a lot and prevents traceability afterward (I wouldn't be able to do an analysis of what systems are using part data because I would then need to check each copy of the part data object to see what it is relating to, so this becomes a nightmare to manage).

Any idea how we can (or if we can) change that setting to allow multiple instances of the element in the same diagram?

Pages: [1]