v15.2 - Accessing Namespace information via API

[Paolo F Cantoni]

We're playing around with the Namespace options.  Is it possible to get the "effective" namespace of an object via the API?  We can see that setting/unsetting the "Suppress Namespace" option in the ribbon will cause the associated change in the repository.  But we can't see the changes in the "Show Namespace" option for the diagram.

The "Show Namespace" option for the diagram doesn't seem to actually toggle the showing of the namespace of an item, it merely toggles the display of the lowest package name (even if the FQN option is set), regardless of whether it is supposed to suppress the namespace.

Is it possible to get a REAL FQN namespace path from EA or must we calculate it ourselves via traversal?

TIA,
Paolo

[Geert Bellekens]

I think you'll have to determine the namespace package for yourself.

Once you have the namespace package, you can get the fully qualified name from the API though.

Geert

[Paolo F Cantoni]

I think you'll have to determine the namespace package for yourself.

Once you have the namespace package, you can get the fully qualified name from the API though.

Geert

Sorry, Geert, I'm not quite following what you're saying.

<Clarification>This is NOT about Reverse Engineering/Generating Code, it's about creating arbitrary items in REAL namespaces (which I think I first touched on a decade ago)</Clarification>

Which namespace package are you referring to?  The Namespace root?

Which is the call to return the FQN?  Sorry to be "dense", but I've had a hard day...   

Paolo

[Geert Bellekens]

With "Namespace" package I mean the package that gets the red earmark after you flag it as a namespace.
It can be used to limit your searches in the construct picker.

The fully qualified name of an element can be found in the API through EA.Element.FQName

Geert

[Paolo F Cantoni]

With "Namespace" package I mean the package that gets the red earmark after you flag it as a namespace.
It can be used to limit your searches in the construct picker.

The fully qualified name of an element can be found in the API through EA.Element.FQName

Geert

Thanks Geert,

A quick test using VBScript reveals that the output of EA.Element.FQName is the same as is visible on the diagram, in other words, it is NOT a fully qualified name including a namespace structure.  It merely reports the first level package (nest), regardless of any setting.

If I have a structure:
A package (namespace root)
+B package
++C package (suppress namespace)
+++D package
++++E item

I expect the FQName of E to be A.B.D.E
If I remove the property of Suppress Namespace on C then I expect the FQNanme of E to be: A.B.C.D.E

Have I miss understood the concept of Namespaces (and/or Fully qualified names)?

Paolo

[Geert Bellekens]

I think the namespaces are mostly used for code generation. They do not play a part in the FQName

Geert

[Paolo F Cantoni]

I think the namespaces are mostly used for code generation. They do not play a part in the FQName

Geert

Yes, that's why it's controlled under the namespace section of the diagram properties...

(that's irony, )

Thanks for your help, Geert.  Saved me a lot of potential frustrations.

Paolo

[Eve]

The namespace root is the global namespace. It's name will not appear in any paths.

If I have a structure:
A package (namespace root)
+B package
++C package (suppress namespace)
+++D package
++++E item

So E should have a fully qualified name of B.D.E.
[/quote]

[Paolo F Cantoni]

The namespace root is the global namespace. It's name will not appear in any paths.

If I have a structure:
A package (namespace root)
+B package
++C package (suppress namespace)
+++D package
++++E item

So E should have a fully qualified name of B.D.E.

Sorry, Eve, my bad. Thanks for reminding me.  It's been so long since I gave the "namespace" mechanism away as a "dud" that I'd forgotten.

Oh well, it's back to "using EA in spite of EA, not because of it".

Paolo