Closed khoroshevskyi closed 1 year ago
I don't think having duplicate projects in the database is a problem. If two namespaces want to have two identical projects that might diverge eventually, that seems ok. Something like forking a PEP.
still, the idea of project digest is a good one, for other reasons. Like, for reproducibility. a pipeline can say, here's a digest of the input data I ran. Now, it's registered.
Then, you can make pephub be able to look up a project by digest.
I like it.
I have added first version of digest to the pepagent, but I think this functionality has to be in peppy, so everyone can generate it.
Should we add project digest to the database, so we won't have same projects in our database?