postgres.update.pq_update_pg

postgres.update.pq_update_pg(
    table_name,
    schema,
    *,
    data_dir=None,
    user=None,
    host=None,
    dbname=None,
    database=None,
    port=None,
    dst_schema=None,
    alt_table_name=None,
    engine='duckdb',
    force=False,
    create_roles=True,
)

Materialize a parquet file into PostgreSQL when the parquet source is newer.

Parameters

Name	Type	Description	Default
table_name	str	Basename of the source parquet file.	required
schema	str	Name of the source parquet schema directory.	required
data_dir	str	Root directory of the parquet data repository. If omitted, defaults to `DATA_DIR` or the current working directory.	`None`
user	str	Destination PostgreSQL user role.	`None`
host	str	Destination PostgreSQL host name.	`None`
dbname	str	Destination PostgreSQL database name.	`None`
database	str	Alias for `dbname`.	`None`
port	int	Destination PostgreSQL port.	`None`
dst_schema	str	Destination PostgreSQL schema. If omitted, defaults to `schema`.	`None`
alt_table_name	str	Destination PostgreSQL table name. If omitted, defaults to `table_name`.	`None`
engine	(duckdb, adbc)	Engine used to load the parquet file into PostgreSQL.	`"duckdb"`
force	bool	If `True`, import the parquet file even when the source metadata are missing or the destination does not appear stale.	`False`
create_roles	bool	If `True`, ensure destination schema owner and access roles exist and apply ownership and grants to the loaded table.	`True`

Returns

Name	Type	Description
	bool	`True` if the parquet file was imported. `False` if the destination was already current or the source metadata were insufficient and `force` was not requested.

Examples

>>> pq_update_pg("dsi", "crsp", dbname="research")
>>> pq_update_pg("company", "comp", dbname="research", force=True)