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 - Guillaume

Pages: [1] 2 3 ... 44
Bugs and Issues / Call Operation action and Ctrl+L shortcut
« on: November 19, 2018, 07:33:47 pm »

I created a call operation action that I need to associate with a SysML block operation. I expected the Ctrl+L keyboard shortcut to let me achieve that however it looks for behaviour instead of operations.

Suggestions and Requests / Text Elt & Note: Link to Package or Diagram Name
« on: November 16, 2018, 08:06:39 pm »

UML/SysML annotations are great to display dynamic info on a diagram by linking it to an element or a diagram.
For an element such as a class, one can associate the element notes, an attribute's description.

For diagrams, the annotation content can be linked with the diagram notes.

It would be nice to extend this with the following:
- For an element, have the Element Name so my annotation would display the package/model name. The same would need to be done for a diagram
- Include this feature for text elements (A in the common toolbox).

The aim is to generate dynamic title elements to help users in browsing through the model via diagram, or via the web if using Prolaborate web solution.

Hi Takeshi,

Thanks for the link. This rule applies to all SysML diagrams i.e. not limited to activity diagrams.
I managed to hide the frame, however it's going to be annoying having to do that on each diagram. It would be nice to have an option that disables this for the SysML project.


General Board / Re: Logical Data Modelling in SPARX
« on: October 30, 2018, 08:51:06 pm »
Hi David,

I have a client that uses Sparx EA for Data Modelling where we manage Conceptual, Logical Physical Data Models.
Usually the Physical Data Models (PDM) are generated via a reverse on a existing DB to understand an existing schema prior to apply changes in the analysis model (LDM).
Logical Data Models are UML class diagrams defined during the analysis process, either "from scratch" or based on an existing PDM, yet without the technology tied with.
Conceptual Data Models are linked either with standards or pivot models ; this can be useful to establish mapping from your LDM.

Sparx EA also provides model transformation features (MDA) to generate a data model for a given DBMS from a UML class diagram but I've seen this less used.
The DB Builder is quite nice if you work on Data Models.


I opened a SysML project with EA14 (latest build). I noticed that I get frames displayed in the back on SysML diagrams.
Opening this project again in EA13 shows that a package has automatically been added.

What is the reason for doing this? (is it related with the SysML OMG spec?)
Is there a way to disable it?


Should I report a bug, or can you please look into it?


Hi Simon,

I carried some tests as you suggested and I still don't get the trigger (Break) whilst it's counting down.

See the example below:


I defined a statemachine diagram to simulate in the interpreted mode.
When I entered a state, a variable value is set e.g. this.waitingTime = 10 and decreased until reaching 0, triggering another state and associated waiting time.

I initially wanted to run a "sleep" function but didn't find anything.
Currently this is a self transition that exits and enters the state with an effect: this.waitingTime--;
There's another outgoing transition when this.waitingTime reaches 0 to move to the next state.
Note: I tried to use operations (Do) with a Behavior code that also decreases the value but it's only executed once.

On one state I want to have a trigger that lets the user leave this state. Unfortunately this trigger is not available when executing the simulation. It seems that the simulation engine does not have time to provide this waiting trigger as it keeps changing the state.

Is there any workaround?

Bugs and Issues / State machine manual simulation and substates
« on: October 08, 2018, 11:56:45 pm »

I'm reviewing EA state machine simulation functions, starting with the manual simulation.
I noticed that my system state machine didn't let me choose  the next trigger when more than one outgoing transition is defined on a state. I eventually figured out that using a transition out from a state that contains a state machine breaks the manual simulation.
Consider the following:
Initial -> OFF
OFF -> ON via the "on" trigger
ON -> OFF via the "off" trigger
ON has a state machine with STOP, PLAY, PAUSE. I have a 2 transitions from PLAY: Stop and Pause.

The manual simulation only prompts to choose between Stop & Pause to leave the "ON.PLAY" state if I delete the trigger to leave state ON to OFF.

Is there a way round to make this work, or is it a limitation in using the manual simulation?

Note: I found a solution in linking all ON substates to the OFF main state with the same trigger name (end states don't work). It doesn't look as clear as the original diagram but all scenarios work. It's similar to Sparx EA Example called "Nested Traces Example"

Thanks for the clarifications.
Consider a Hardware block, I wanted to have my SysML Block with a "Hardware" stereotype, and when I get the part's equivalent in the IBD, I would have liked to see this part with the "Hardware" stereotype as well.
As it currently stands, I have to differentiate the parts stereotype e.g. "Hardware part".

It's not a big issue and I'm keeping this approach based on your comments.

Hi Geert,

Sounds like a good idea except that I don't want the parts to have the tagged values set on the blocks.


I want to use let's say "MyType" stereotype on a number of elements e.g. a MyType SysML block and a MyType Part.
When I export the UML Profile, the resulting XML file is missing the parts' stereotypes definitions. It seems like EA doesn't like stereotype duplicates.

As a workaround I can create separate UML Profiles to embed in my MDG, but I was wondering if this is a known issue.

Note that I found that running a ReloadDiagram after the OnPostNewConnector in some instances makes EA crash.
I also tried running a SaveDiagram before and noticed it happens. I'm not sure there is a way to sort this so I unfortunately need to disable this automated diagram reload unless there's a way round.


With a script, I update an element on a diagram when a connector is deleted. I'd like to reload the diagram to see the changes.
However since the trigger event is caught before deleting the connector, doing this crashes EA.

When creating a new connector it works because the event is OnPostNewConnector.

It would have been nice to have a PostDeleteConnector with the diagram ID and some of the deleted connector's details like its type, stereotype, + source & target object ids so I can refresh the diagram.
The only way round I can think of is to create a fake element to trigger a reload but I don't really like that.

Any comments on addressing this scenario so the user doesn't have to save & reload manually?

Automation Interface, Add-Ins and Tools / Re: Help with deploying add-ins
« on: September 26, 2018, 12:28:21 am »
Hi Uffe,

Have you got an update on getting an EA Addin template available from VS?


Pages: [1] 2 3 ... 44