Load Django environment for CLI script

django_cli.py

#!/usr/bin/env python
# Load the Django environment
from django.core.management import setup_environ
import os
import sys

try:
    project_path = os.environ['DJANGO_PROJECT_PATH']
except KeyError:
    raise Exception("Unable to locate Django project.  Set your operating "
                    "system's DJANGO_PROJECT_PATH environment variable to "
                    "point to the root of the Django project.")

if project_path not in sys.path:
    sys.path.append(project_path)

settings_module = os.environ.get('DJANGO_SETTINGS_MODULE')
if settings_module:
    settings = __import__(settings_module)
else:
    import settings
setup_environ(settings)
# End Django environment load.

This script does nothing. It deliberately does nothing so that I can clone it and just add code at the end for Django CLI script. It requires, at minimum, the DJANGO_PROJECT_PATH environment variable be set. I also included an optional DJANGO_SETTINGS_MODULE variable that, if present, will be used as the settings module. E.g., running export DJANGO_SETTINGS_MODULE=settings_dev prior to invoking this script would be similar to ./manage.py --settings=settings_dev. The best way to make sure these environment variables are properly set is to append the export statement to your virtualenv's activate script. You are using virtualenv, right?

Oh and just as an aside, yes, this is really ugly, especially with the way the imports are scattered about, but I recommend putting any additional imports below this script, as some of the core django imports do some metaprogramming at import time that won't work unless it's imported after the execution of setup_environ.