Why does one Model know about the other Model?

[DaveToo]

I use EA almost exclusively to reverse-engineer existing Python code, to create diagrams to help my understanding of the product I work with.  I may be missing something fundamental here about how EA works...

What I'm trying to do at the moment is have one open EA project instance model two disparate branches of the same codebase.  One branch is a fork that I'm customizing heavily, that won't end up back in the main product/branch.  I want to compare this against the trunk/main branch. I had expected that two root models within a project would not interact with each other, but this seems not to be the case:

• First, I create a new project and within the default Model space, I reverse/code-engineer import the (python) project source tree.
• Then I create a new Root model in the same open EA project, and check out my branch of the code (into the same Git workspace).  Upon doing so I get a popup about classes missing from the code... classes in the *trunk* model that are not in my branch.
• In both Root Models, the "Class Model" view is set as a root namespace (in fact they are not both named "Class Model" either).

What am I seeing here?

Thanks.

[DaveToo]

I must not be using the right "lingo"?

[Eve]

Effectively, EA is seeing that the classes are the same by its definition (same name and file)

You can stop this behavior by using a different file path for the two versions.

Another way to achieve what you want may be to baseline the trunk and then import the branch directly over it. This allows you to do a comparison using EA's baseline comparison.