OSX users who develop Python applications (any application really) that use PostgreSQL are in luck. If you are looking
for, quite possibly, the easiest way to install, run and use the most advanced open source database on the planet
(provided you are using Mac OSX > 10.7). Postgres.app - will make your life very happy. When you just want to run PostgreSQL and get to coding, Postgres.app from Matt Thompson and the awesome people at Heroku is where its at.

Install PostgreSQL

  1. Use Homebrew and follow this guide or something similar
  2. Use the PostgreSQL binaries and follow this guide
  3. Use postgres.app - it is hard to make an argument against this app. Download it, drag it to applications, DONE.

Use PostgreSQL

If you have ever used postgreSQL from the commandline you have undoubtedly felt this pain:

:::bash pg_ctl -D /usr/local/var/postgres start pg_ctl -D /usr/local/var/postgres status pg_ctl -D /usr/local/var/postgres stop -s -m fast

This is just to start and/or check on the darn thing. Then you can enter the psql command to access the $USER database.
Remember that the $USER database doesn't have a user/password like it would in a production environment.

Use Postgres.app

The postgres app simplifies this process by allowing you to start PostgreSQL like any other mac app, click on the app.
But before you do that; Add one line to your .bashrc or in my case, .zshrc.

:::bash PATH="/Applications/Postgres.app/Contents/MacOS/bin:$PATH"

Once the app initializes, you get the cool little elephant in your toolbar. That's it, PostgreSQL is running and ready to
relation. With this done you can now create a database:

$ psql
psql (9.2.2)
Type "help" for help.

$ michaelmartinez=# CREATE DATABASE my_database;
# If you want to delete the database you just created
$ michaelmartinez=# DROP DATABASE my_database;

note: don't forget the semi-colon!

Use for Development

To use this with Django you need to install the psycopg2 package, then its just a matter of adding the correct parameters
to your DATABASE settings:

"default": {
"ENGINE": "django.db.backends.postgresql_psycopg2",
"NAME": "my_database",
"USER": "",
"HOST": "localhost",
"PORT": "",

To use this with your Flask app, you can use the Flask-SQLAlchemy extension or SQLAlchemy on its own:

from flask import Flask
from flask.ext.sqlalchemy import SQLAlchemy

app = Flask(__name__)
app.config['SQLALCHEMY_DATABASE_URI'] = 'postgresql://localhost/my_database'
db = SQLAlchemy(app)

Vanilla SQLAlchemy

from sqlalchemy import create_engine
engine = create_engine('postgresql://localhost/my_database')

371 words ~ 2 min read

  • Flesch-kincaid Index / Reading Ease: 70.56
  • Flesch-kincaid Grade Level: 5.91
// // // // // // // // // // // // // //