Module trase.tools.data_release_package.snapshot.destination

Names and prepares the PostgreSQL schema a data release package is snapshotted into.

Every package is snapshotted into a schema of its own, named after the package folder. A snapshot is a replacement, not a merge: a package owns its schema outright, so anything already in there is dropped before the new tables are written. Since the whole snapshot runs in one transaction, a failure part-way through leaves the previous snapshot's schema untouched.

Functions

def check_identifier(identifier: str) ‑> str

Guards the identifiers we interpolate into SQL. Raises if one looks unsafe.

def create_schema(connection, schema: str) ‑> None

Creates schema, owned by the role that also owns the snapshotted tables.

def drop_all_tables(connection, schema: str) ‑> List[str]

Drops every table in schema and returns their names.

def existing_tables(connection, schema: str) ‑> List[str]

The tables currently in schema, including any archived by an earlier run.

def prepare_schema(connection, schema: str) ‑> None

Makes schema exist and be empty, ready for the package to be written into.

def schema_exists(connection, schema: str) ‑> bool

def schema_name(package_dir: pathlib.Path) ‑> str

The PostgreSQL schema a package is snapshotted into, derived from its folder.