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.

Topics - Polymorph

Pages: 1 2 3 [4]
General Board / Relationship Matrix Notes
« on: January 29, 2016, 10:50:06 pm »
Does anyone know if there is a way to construct a relationship matrix where the notes from the model elements are also visible. Say, for example, I had a matrix that maps requirements (shown as rows in the matrix) and use cases (shown as columns) together. It would be very useful if I could not only display the name of the requirements but also have an extra column next to the requirement name that displayed its notes

This would allow you to read the text of the requirement quickly and deduce which use case(s) it should be linked to

Also, how do you insert an image into these posts!!!! I can't work it out

Bugs and Issues / Call Behaviour Action Pins
« on: January 14, 2016, 03:26:42 am »
When I create a call behaviour action (EA 12.1, build 1224) and type it by an activity that
1. already exists in the model
2. has activity parameters

The call behaviour that is created shows the activity pins but does not show their type. If I double click on the activity pin and then click "OK" on the properties dialogue box, EA then updates the diagram and also shows the type of the action pin.

I've checked and the are no (obvious) display options that are hiding the types from the diagram

General Board / Interfaces, blocks and ports
« on: October 05, 2015, 09:17:14 am »
I am new to Enterprise Architect (Rhapsody user). If I create an interface block (in SysML 1.3) and then add a port to a block that implements that interface, I would expect to have access to these operations whenever I place this block on a sequence diagram.

However, when I do this and create a message between this block and an actor, I do not see the operations contained within the interface.

Is this possible?

Uml Process / Project Searches
« on: December 16, 2015, 08:15:56 pm »
I have some custom SQL searches. I wish to make them available to everyone that uses the model.

I am aware that it is possible to export the SQL to an XML file and then have each individual user re-import the search but is there any way to embed these searches within the project database itself?

Need: if I make changes to the search, I want these changes to affect everyone. I don't want each of my users to have to perform an import each and every time I change something.

Uml Process / Showing stereotyped action pins
« on: December 16, 2015, 04:30:01 am »
SysML action pins may be stereotyped with an «optional» keyword.

Does EA allow this stereotyping to be displayed on diagrams? I have tried various diagram and tool settings but so far have had no luck in getting this information to display

Uml Process / Using Connectors on IBDs
« on: November 19, 2015, 07:36:34 am »
I have two questions, one specific to EA and a more general one about modelling best-practice:

1. Specific: The SysML spec says that connectors on IBDs may be typed by associations. Does EA support this?

2. It is not always necessary to link two parts on an IBD via a port - you may legally join the parts directly together without using ports. Can anyone offer some "best practice" tips on when to use each approach?

Here's my thinking: connectors between parts represent purely general, structural relationships (e.g. a wheel is connected to an axle) whereas connectors between ports emphasise the specific behavioural features that are exposed between (e.g. a connection to a car via the diagnositics port reveals "diagnositic data")

Thoughts? Comments? Rotten Tomatoes?

Thanks in advance!

Uml Process / Display Action Pin Direction?
« on: November 10, 2015, 10:31:31 pm »
Is there any way to display the direction of an action pin on an activity diagram?

Uml Process / Displaying Allocation In Block Compartments
« on: November 06, 2015, 12:36:32 am »
I have created an allocation relationship between an «action» and «block»

The SysL spec defines an allocatedTo and allocatedFrom notation to display allocation information on diagrams  (see SysML 1.3 Spec section 15.2.1, page 152)

Does EA support this notation?

Uml Process / Tips for Creating Operations From Actions?
« on: November 05, 2015, 03:08:02 am »
Hi everyone

I have observed that if I drag an operation (owned by a block) onto an activity diagram that EA helpfully creates a callOperation action which is linked to the original operation. This is great!

Does anyone know if there is a way to perform the reverse? That is to say, can I somehow drag pre-existing actions from my model onto a block in such a way that Operations are created or must I manually create them?

Uml Process / Creating Entry/Exit Events on States
« on: October 29, 2015, 01:22:52 am »
I am creating state machines using SysML...

My state machine is nested underneath a block (that has several operations). I had expected to be able to pick and chose the operations from the parent block and reuse it in my state machine

I can add an effect to a transition by selecting the appropriate operation from the owning block using  the Transition Properties dialogue box.

That's great ;D, however, in other situations it is preferable to add entry/do/exit actions to a state itself...

I have worked out that I can add a copy of the operation from the original block to the specific state(s) owned by the state machine...

This is OK... but what if during subsequent work the name of the operation on my parent block changes? Then I'll have an inconsistency in my model. Is there any alternative way of doing this or must I be resigned to having a copy of something that resides elsewhere in the model?

Uml Process / Suppressing CallOperation Detail
« on: October 28, 2015, 09:34:31 pm »
I am using SysML actions on activity diagrams. I have recently discovered the "CallOperation" action - which seems to be an EA specific (rather than SysML) model element. This appears to do precisely what I need (link action and operation together).

However, there is a problem. EA automatically places the name of the operation underneath the action name. This causes readability issues for my diagrams as the action often does not live in the same package as the block that owns the operation; resulting in a very long name


Is there a way that this extra information can be hidden?

Uml Process / Associations between Blocks
« on: October 22, 2015, 01:00:48 am »
Occasionally I create assocations between blocks and rather than the line being a simple line, the line is adorned with two black dots at either end.

Can anyone tell me what this represents? I would post a picture of it but I also can't work out how to insert images into posts!!!

Uml Process / Convert Block to Association Block
« on: October 22, 2015, 07:44:45 pm »
I am using SysML blocks to create an ontology of terms. Sometimes I discover that a block I have created (and linked via associations to other blocks) is also an association block for other types of relationships

Is there a way that I can change this block into an Association Block... I don't really want to have create a new Association Block and then move all the existing reference assocations to this new block.

Uml Process / Event (Signal) Receiptions
« on: October 13, 2015, 09:13:01 am »
I wish to model an asynchornous message sent from an actor into my system on a SysML Sequence Diagram.

I have created a signal element in my model and have created a message from my actor to my system. I have set the "Control Flow Type Sync" to "Asynchoronous" and the Kind to "Signal" on the message properties dialogue.  I have then selected the appropriate «signal» that I want this message to represent by pressing the "signal" button on the same dialogue screen and locating it within the browser.

Now I want to specify a reception on the target block...

I read in the EA help that:
"To define a reception, create an operation in the receiving object and assign the stereotype <<signal>> to it. The reception has the same name as the signal that the object can receive."


When I did this, I noticed that the list of predefined stereotypes available for an operation did not include «signal» (or "reception" or anything else that looked relevant). I therefore concluded that I simply had to type the name of the stereotype into the box rather than browse for it...

I have done this as described and the block representing my system now has a stereotyped operation with the same name as my signal...

Now my question:
What links (if any) exist between this (supposed) reception and the signal itself?

I want there to be traceability between the signal events I create in my model and the receptions themselves. It is of little use to me to have an operation with a name that matches the signal - what happens if in time the signal name were to change? Would I have to go through and manually change all receptions?

I plan to use these sequence diagrams to create «Interface Blocks» to type «Proxy Ports»

Note: when I am creating my sequence diagrams I drag the target block onto the diagram as a "link" not a "lifeline". I don't know what advantages/disadvantages that might cause (other than preventing my model from rapidly filling up with multiple instances of my blocks!)

I want to create diagrams free from all the random colours that EA thinks I want to see! I've managed to remove things like gradient fills from shapes and the page using the Tools>Options>Appearance dialogue box - that was pretty easy.

After some investigation I discovered that you can create a "Template" package and then create diagrams within this with the style settings you require (

This sort of works... if I create a new model element and drop it onto the diagram then it usually assumes the size and colour format options I defined. However, this is pretty tedious... I need to drop all of the elements from all of the possible toolboxes onto each diagram and then remove the colour from each one.

The help mentions Default Appearance Menu (
but this is no help because this seems to only apply to specific model elements (not types of model element such as use cases, blocks etc)

Another option mentioned in the help file is to create a stereotype with the desired properties... This might be OK but don't I then have to first create the basic use case model element and then apply the stereotype to it (or is there a way to force all new instances of use cases to adopt this new stereotype by default)?

Surely there must be an easy wy to  turn off all the (frankly meanlingless) colours from all model elements?

Pages: 1 2 3 [4]