postgresql-wheel
A Python wheel for Linux containing a complete, self-contained, locally installable PostgreSQL database server.
All servers run as the Python process user in a local path, so this wheel does not require root or sudo privledges. Databases can be initialized in any directory.
Servers can be setup and torn down in test fixtures with no additional outside dependencies.
Currently this wheel only works for most flavors of Linux.
Postgres is compiled in the same "manylinux" environments provided by the cibuildwheel tool using Github Actions and directly archived into the wheel's "package_data".
The wheel can be installed with pip:
$ pip install postgresql-wheel
Postgres binaries in the package can be found in the directory pointed to by the postgresql.pg_bin global variable. Function wrappers around all of the postgres binary programs, like initdb and pg_ctl functions are provided for convenience:
>>> from postgresql import initdb, pg_ctl
>>> initdb('-D testdatabase')
>>> pg_ctl('-D testdatabase -o "-p 5678" -l testdatabase.log start')
>>> import psycopg2
>>> c = psycopg2.connect("postgres://localhost:5678/postgres") # connect with local client
>>> with c.cursor() as q:
>>> q.execute("select version()")
>>> print(q.fetchall())
...
[('PostgreSQL 13.4 on x86_64-pc-linux-gnu, compiled by gcc (GCC) 8.3.1 20190311 (Red Hat 8.3.1-3), 64-bit',)]
>>> pg_ctl('-D testdatabase stop)
For the purposes of testing, convenience functions are provided for setting up and tearing down databases:
>>> pgdata, con_str = postgresql.setup()
>>> postgresql.psql(f'-h {con_str} -c "select version()"')
>>> postgresql.teardown(pgdata)
There is also a pytest fixture called tmp_postgres that returns a new connection string to a temporary databse local domain socket, and can be used in a pytest:
>>> from postgresql import tmp_postgres
>>> def test_foo(tmp_postgres):
... postgresql.psql(f'-h {tmp_postgres} -c "select version()")
841 Dec 21, 2022
33 Dec 8, 2022
3 Nov 24, 2021
7.2k Jan 9, 2023
703 Dec 30, 2022
957 Dec 30, 2022
1.3k Dec 25, 2022
5.1k Jan 8, 2023
3.7k Jan 8, 2023
2.5k Dec 27, 2022
66 Dec 02, 2022
1 Jan 11, 2022
6.5k Jan 01, 2023
6 May 30, 2022
6.9k Jan 03, 2023
7 Aug 21, 2022
2.3k Jan 06, 2023
1.9k Jan 04, 2023
1 Feb 26, 2022
1 Feb 08, 2022
3 Nov 19, 2021
5.5k Jan 08, 2023
2.6k Dec 27, 2022
2 Oct 05, 2022
348 Jan 03, 2023
1.2k Jan 03, 2023