#postgresql

bin+lib postgresfixture

Easily create and manage PostgreSQL clusters on demand for testing and development

8 unstable releases (3 breaking)

0.5.0 Oct 8, 2023
0.4.0 Oct 8, 2023
0.3.3 Feb 26, 2023
0.3.1 Apr 9, 2022
0.2.3 Mar 13, 2020

#1065 in Database interfaces

Apache-2.0

105KB
2K SLoC

rust-postgresfixture

A Rust library and command-line tool for creating standalone PostgreSQL clusters and databases, useful for experimentation, development, and testing.

It's based on the Python postgresfixture library which saw heavy use in MAAS. That was (and is) a useful tool when experimenting with PostgreSQL. For example we could use it to bring up a cluster to run a development server. However, it came into its own in MAAS's test suites, and was key to making MAAS's test suites faster.

This Rust version started out as a straightforward port but it has deviated significantly from the design of its Python counterpart.

This code works and seems to be reliable, but the command-line and API may change before 1.0, potentially causing breakage. If this is a problem I suggest pinning on a specific version and checking back once in a while to see if it can be upgraded, or use something automated like Dependabot.

Command-line utility

After installing Cargo, cargo install postgresfixture will install a postgresfixture binary in ~/.cargo/bin, which the Cargo installation process will probably have added to your PATH.

Note that this tool does not come with any PostgreSQL runtimes. You must install these yourself and add their bin directories to PATH. To select a specific runtime you must set PATH such that the runtime you want to use is before any others. The runtimes subcommand can show you what is available and what runtime will actually be used.

$ postgresfixture --help
Easily create and manage PostgreSQL clusters on demand for testing and development.

Usage: postgresfixture <COMMAND>

Commands:
  shell     Start a psql shell, creating and starting the cluster as necessary
  exec      Execute an arbitrary command, creating and starting the cluster as necessary
  runtimes  List discovered PostgreSQL runtimes
  help      Print this message or the help of the given subcommand(s)

Options:
  -h, --help     Print help
  -V, --version  Print version

$ postgresfixture runtimes
   10.22      /opt/homebrew/Cellar/postgresql@10/10.22_6/bin
   11.21      /opt/homebrew/Cellar/postgresql@11/11.21/bin
   12.16      /opt/homebrew/Cellar/postgresql@12/12.16/bin
   13.12      /opt/homebrew/Cellar/postgresql@13/13.12/bin
   14.9       /opt/homebrew/Cellar/postgresql@14/14.9/bin
   15.4       /opt/homebrew/Cellar/postgresql@15/15.4/bin
=> 16.0       /opt/homebrew/bin

$ postgresfixture shell
postgres=# select …

$ postgresfixture exec pg_dump
--
-- PostgreSQL database dump
--
…

Use as a library

The essential functionality in this crate is in the Cluster struct and its implementation. This covers the logic you need to create, run, and destroy PostgreSQL clusters of any officially supported version (and a few older versions that are not supported upstream).

use postgresfixture::prelude::*;
for runtime in strategy::default().runtimes() {
  let data_dir = tempdir::TempDir::new("data")?;
  let cluster = Cluster::new(&data_dir, runtime)?;
  cluster.start()?;
  assert_eq!(cluster.databases()?, vec!["postgres", "template0", "template1"]);
  let mut conn = cluster.connect("template1")?;
  let rows = conn.query("SELECT 1234 -- …", &[])?;
  let collations: Vec<i32> = rows.iter().map(|row| row.get(0)).collect();
  assert_eq!(collations, vec![1234]);
  cluster.stop()?;
}
# Ok::<(), ClusterError>(())

You may want to use this with the functions in the coordinate module like run_and_stop and run_and_destroy. These add locking to the setup and teardown steps of using a cluster so that multiple processes can safely share a single on-demand cluster.

Contributing

If you feel the urge to hack on this code, here's how to get started:

Running the tests

After installing the source (see above) run tests with: cargo test.

However, it's important to test against multiple versions of PostgreSQL. The tests will look for all PostgreSQL runtimes on PATH and run tests for all of them.

First you must install multiple versions of PostgreSQL on your machine. Read on for platform-specific notes. Once you've installed the versions you want, [postgresfixture::runtime::strategy::default()] may be able to automatically find them – and, since this function is used by tests, those runtimes will automatically be tested.

Debian & Ubuntu

From https://wiki.postgresql.org/wiki/Apt:

$ sudo apt-get install -y postgresql-common
$ sudo sh /usr/share/postgresql-common/pgdg/apt.postgresql.org.sh -y
$ sudo apt-get install -y postgresql-{9.{4,5,6},10,11,12,13}  # Adjust as necessary.

macOS

Using Homebrew:

$ brew install postgresql  # Latest version.
$ brew install postgresql@{9.{4,5,6},10,11,12,13}  # Adjust as necessary.

Making a release

  1. Bump version in Cargo.toml.
  2. Paste updated --help output into README.md (this file; see near the top). On macOS the command cargo run -- --help | pbcopy is helpful.
  3. Build and test: cargo build && cargo test. The latter on its own does do a build, but a test build can hide warnings about dead code, so do both.
  4. Commit with message "Bump version to $VERSION."
  5. Tag with "v$VERSION", e.g. git tag v1.0.10.
  6. Push: git push && git push --tags.
  7. Publish: cargo publish.

License

This project is licensed under the Apache 2.0 License. See the LICENSE file for details.

Dependencies

~21–31MB
~386K SLoC