The postgres datastore implementation.
This should generally be considered by far the slowest implementation, however it provides a few major benefits:
- Transaction changes can be rolled back on error.
IndraDBserver processes can run on the same datastore at the same time.
- You can use all of the postgres tooling to poke around at the results.
- Thanks to foreign keys et al., this is probably less buggy than other implementations.
A datastore that is backed by a postgres database.
A postgres-backed datastore transaction.