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 - Paolo F Cantoni

Pages: 1 ... 4 5 [6] 7 8 ... 78
Bugs and Issues / Relationship Matrix fails with "Information Items"
« on: August 31, 2017, 05:00:00 pm »
We have our own element metatype called "Information Item" according to the following (partial) MDG definition.
Code: [Select]
<Stereotype name="InfrmtnItm" metatype="Information Item" If we try to select it from the dropdown for the Element Type, EA helpfully (It's an ironic observation, Joyce) replaces  "Information Item" with "InformationItem".  Consequently, we can't use them in a Relationship Matrix.  Pure EAUI!

Needs to be fixed!


Virtualized Connector Ends (VCEs) now honour shapescripts.  BUT it doesn't seem to provide access to all the properties used by the shapescript.  We have a decoration that will provide an indicator that the Notes field for an element is empty.
Code: [Select]
if(hasproperty("notes","")) //Notes empty...
The base object responds correctly, but the VCE does not, it ALWAYS shows the notes as empty regardless of the actuality.


Bugs and Issues / Virtualized Connector End as Diagram Object
« on: August 30, 2017, 05:24:45 pm »
I've just reported another defect in the Virtualized Connector End functionality.  I did a search and found that the only posts that mention that functionality are mine!   ::)

In quickly reviewing them, and having used the functionality from time to time, I've come to the conclusion that a significant part of the problem with the shapes and connectors tot he virtualized connector end is that there is NO corresponding Diagram Object.

With automation, one can place MORE than one instance of an object on a diagram (and AFAIK the Consistency Checker accepts it).  I guess that, at present, EA synthesises a diagram object.  My suggestion why not make that item persistent?  It can be distinguished in some way from the Base diagram object and be referenced in the virtualized connector end.  This would, it seems to me, allow the full functionality of a REAL diagram object, without significantly deviating from the concept of a virtualized connector end.  Because the VCE diagram object can be distinguished from the base object, one can finesse any appropriate functionality through to the base object.


In 1303: Virtualized Connector End needs shapescript I mention that the virtualized connector end is different from the base diagram object.  However, today we discovered that not only does it do that, but it will resize the base diagram object to the default size (at least for Rectangular notation).  Once it's done that, you CAN'T readjust the sizing back to what you started with (not even by "jamming" the DB!!!)


Insert Related elements doesn’t have Select All/None on Connector and Element types.  Anywhere where there can be more than 3 possible values (either permanently or as a result of a query), there should be Select All/None functionality available.  It would also be good to have a "remember settings" in the case of this dialog (even if only on a per session basis).


As far as I'm aware, there's no mechanism in a shapescript to determine whether or not the relationship is Self-Referential.  We'd like to vary the shapescript if the line is self-referential.  Can we please have such a property in the Connector Properties list?


Bugs and Issues / Orthogonal Style Lines don't terminate consistently
« on: August 24, 2017, 04:57:36 pm »
I added 23 objects (all same metatype and size) to a diagram.  I needed to connect them to one object also on the diagram.  So I used the Quicklinker to make the first link and then used [F3] to repeat the relationship for the other objects.  So far, so good!  I noticed after the first half-dozen or so of these operations, that the lines were overlaying each other at the destination end (the one object I'm connecting to).  Cool, I thought!  That's neat!  However, as I kept adding connections and the side to which the destination was connecting changed, the subsequent lines didn't overlay each other on the second side.  That's inconsistent.

Could I suggest that the default terminations of the Orthogonal line styles be the centre of the side EA has decided to attach it to?  In the first case above, the lines were overlayed at one corner of the side.  In the second case, they were distributed across the side.


Now that we can have relationships from relationships to elements, it is defective to make every relationship to a UML Constraint be replaced by an implied (not even a real) Notelink.  We defined a number of relationships between arcs and stereotyped Constraints in our QuickLinker.  Imagine our surprise when, as we released the mouse, we weren't given a set of QuickLinker options, but an implied notelink was automatically created.


Bugs and Issues / Can’t display hidden label for Constraint
« on: August 23, 2017, 10:03:31 am »
We're naming our UML Constraints - so we can tell one from the other.  This will produce a Label.  However, once the label is suppressed (either intentionally or otherwise) there is no mechanism to recover it and make it visible again.

Please Rectify!

Bugs and Issues / Can’t suppress Notes display for Constraint
« on: August 23, 2017, 09:59:57 am »
We're using Stereotyped UML Constraints in our MDG (for technical modelling constraints rather than Business Constraints for which we use the ArchiMate Constraint).  We entered a narrative description of the Constraint in the Notes field and it displays.  BUT on some diagrams, we only want to show a small indication of the constraint (i.e. as if no notes were present).  However, we can't suppress Notes.  A Constraint will not respond to the [Ctrl+SHift+Y] Feature and Compartment Visibility, nor is there a specific context menu item.

Please rectify,

Adding an arc on one diagram with a default line style (say Orthogonal Square), it gets added to other diagrams with Custom Line Style.  Annoying to say the least!

But the real problem is it's WRONG!

Concistency, konsistency, consistensy! TMUffe - after Paolo


This bug is probably related to Proxy Connectors don’t move correctly when moving multiple vertices on diagram.  If you grab a relationship between relationships and try to drag it, it won't move.  This is probably because the ProxyConnectors (actually Connector Proxies) at the ends of the relationship aren't handled correctly.  Interestingly, and annoyingly, if you drag one of the related (end) lines, the proxies move!


If you have aligned vertices (of the same size) and you connect them via an orthogonal (square or rounded) arc, you can only drag it to the edge of the vertex.  You can't drag it past.  If the vertices, are of different sizes or are not aligned, then you can get the additional waypoints to bend the arc as required.  It would be good to be able to do this also when the vertices are the same size and aligned.

I decided it was a feature request, but it could be considered a bug.

If you select multiple vertices on a diagram and move them (as a unit), the arcs between the vertices move appropriately.  However, if there are arcs between arcs (expressing relationships between relationships), the proxy connectors get scrambled in the move.


Uml Process / What is an "Instance"?
« on: August 17, 2017, 10:15:10 am »
As promised in another post, here it is...

When modelling systems and application programs, this is fairly clear.
"Car" is a Classifier, "BMW 318i" is a (specialised) Classifier, "BMW 318i with plate # XXX 888" is an instance of the Classifier BMW 318i".  We implement Classifiers with Classes and Instances with Objects.  Everyone agree?

A classifier can be defined as anything that aggregates a set of instances according to some property set.  Agreed?

Now let's look at modelling some other things.
ArchiMate 3.0 when talking about (Business) Actors observes:
"Business actors may be specific individuals or organizations; e.g., “John Smith” or “ABC Corporation”, or they may be generic;e.g., “Customer” or “Supplier”."

This introduces the notions of "specific" vs "generic".

It seems to me that generic Actors fall under the definition of Classifier whereas specific actors don't (seem to).  Colloquially, we say "John Smith is an instance of an actor".  Can we also say "Customer is an instance of an actor"?

So what's going on?  Should we have Actors (for Generic) and Actor instances (for specific)?


Pages: 1 ... 4 5 [6] 7 8 ... 78