v16 - t_xref MOFProps weird entries for non-Generalization Set Generalizations

[Paolo F Cantoni]

I'll report some issues with Generalization Sets over the next little while.  This one started as a potential defect in t_xref, but before reporting the defect, I thought I try to get some clarification.

As per Q's excellent "inside EA" book and searches in the forum, we went looking for Generalization Set names in the t_xref MOFProps entries fr the specified Generlaization Links.  We found them there, BUT we also found MOFProps entries that held the values for the StereotypeEx that would normally be found in the t_xref Stereotypes entry.  The MOFProp entries were duplicates of the Stereotypes entries except they had behaviour="generalizationSet"  whereas the Stereotype entry didn't.

Code: [Select]

X.Name   Type   Behavior   Description
Stereotypes connector property             @STEREO;Name=Rstrctn;FQName=MDG::Rstrctn;@ENDSTEREO;
MOFProps   connector property   generalizationSet   @STEREO;Name=Rstrctn;FQName=MDG::Rstrctn;@ENDSTEREO;It's unclear why these entries are there, but I suspect they (the generalizationSet) shouldn't be there.  We intend to clear them out.  Does anyone know anything about them?

TIA,
Paolo

[qwerty]

Honestly no. And I fear you are in an area which not so many people are actually endeavoring (except for UML spec reviewers like Axel Scheithauer on SO, Neil Capey from Sparx - unfortunately not answering on SO; maybe Eve here - also not on SO).

q.

[Eve]

The generalizationSet row does appear to have had its description copied from the Stereotypes row. It doesn't hold any of the information of the generalization set at all. I can't reproduce that happening with just the tool, so I can't comment further.

[Paolo F Cantoni]

The generalizationSet row does appear to have had its description copied from the Stereotypes row. It doesn't hold any of the information of the generalization set at all. I can't reproduce that happening with just the tool, so I can't comment further.

Thanks, Eve,
For clarification, are you describing my example or what you may have found in your repository?


I agree; the reason for the creation of the anomalous rows is probably shrouded in the mists of time.  Some of these Generalization Sets were originally created a few years and many versions ago.  We never had occasion before to look in this part of t_xref, so they may have been "lying dormant" since then.


We've now purged the anomalous rows.


Paolo

[Eve]

The generalizationSet row does appear to have had its description copied from the Stereotypes row. It doesn't hold any of the information of the generalization set at all. I can't reproduce that happening with just the tool, so I can't comment further.

Thanks, Eve,
For clarification, are you describing my example or what you may have found in your repository?

I'm getting a correct xref in the row for generalizationSet. It tried setting the stereotype in the GUI both before and after creating the generalization set. Neither impacted the result.