Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

comp_id.module #532

Merged
merged 2 commits into from
Sep 9, 2019
Merged

comp_id.module #532

merged 2 commits into from
Sep 9, 2019

Conversation

corakingdon
Copy link
Collaborator

@corakingdon corakingdon commented Jul 28, 2019

@davidanthoff this is for what we were talking about on Friday, to have the comp_id in a component definition hold the full module instead of just a symbol of the bottom most module. This currently works now (on the MimiIWG example that wasn't working before) and passes all of the Mimi tests (although I had to change three tests that were specifically testing this functionality).

Would it be better though to have the comp_id still store a symbol, but just make sure the symbol is the full module path (i.e. make it hold Symbol("Main.foo.bar") instead of what it would previously store, which was just :bar)? I'm just not sure what the implications are of passing around the actual module object and if there are any disadvantages of this we should be worried about.

@corakingdon corakingdon changed the title WIP Chang comp_id to hold the module instead of symbol WIP comp_id.module Jul 28, 2019
@corakingdon
Copy link
Collaborator Author

I put an example of the problem in an issue here: #533

@davidanthoff davidanthoff requested a review from rjplevin August 2, 2019 00:41
@davidanthoff
Copy link
Collaborator

I think @rjplevin needs to make the call here. To me it seems fine to store the module itself instead of the name. With the caveat that we should think through what that means if someone redefines a module... Probably it would mean that one would then have a reference to an outdated module around...

@corakingdon
Copy link
Collaborator Author

Rich started implementing this differently in his class-based branch as well, to have it store the whole module path as symbols instead. We should remember to talk about this together on Monday!

@corakingdon corakingdon requested review from davidanthoff and removed request for rjplevin September 9, 2019 20:53
@corakingdon
Copy link
Collaborator Author

@davidanthoff I believe we decided to merge this as is for now. Do you want to take a look before I merge it? The travis tests aren't passing because the models are on the old versions

@corakingdon corakingdon changed the title WIP comp_id.module comp_id.module Sep 9, 2019
@davidanthoff
Copy link
Collaborator

Looks good!

We should really try to move to a situation where we don't end up with a broken CI build on master, which seems to be what is going on here, right?

@corakingdon
Copy link
Collaborator Author

Yes, I'm not sure how to prevent that when we have updates with breaking changes. I think Lisa has a better grasp on what the ideal workflow could be. One thing on this PR before I merge it though: should I rename the field in the comp_id to be called module instead? Or is it okay to leave it as module_name?

@davidanthoff
Copy link
Collaborator

davidanthoff commented Sep 9, 2019

should I rename the field in the comp_id to be called module instead?

I'm indifferent.

@corakingdon corakingdon merged commit 75fbdf8 into master Sep 9, 2019
@davidanthoff davidanthoff deleted the comp-module branch September 9, 2019 23:49
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants