# iceberg.catalog

> A module of iceberg's API.

Catalogs: how a table name becomes a `TableMetadata`.

`filesystem` needs nothing but a path. `rest` is a documented stub — see its
module docstring for exactly what is and is not implemented and why. `sql` is
a JDBC-style catalog over sqlite.mojo or postgres.mojo — sqlite for local
development and PyIceberg test parity, PostgreSQL for a catalog several
writers can share — with `_sqldriver` holding the small part that differs
between the two.