This is a prototype of aggregate commit authorship signals.

This is how it works:

1. Database is central and contains both the individual signals as well as aggregates.
2. Aggregates are updated with stored procedure as new data appears.
3. The API serves ownership signal based on commit authorship for every directory with a single database query.
4. Code intel tooling schedules scanning repo and adding relevant commits to the history.

TODO

• Database
• GraphQL API wired to the database
• UI
• Pulling most recent commit data
  • and adding them to the database
• Code intel like indexing

Caveats

• The commit adding database code does not consider root directory of repo yet. - simple fix.

Test plan

Spike spike