This add-on provides models and commands to import country, subregion, region/state, and city data in your database.
The data is pulled from GeoNames and contains cities, subregions, regions/states and countries.
Spatial query support is not required by this application.
This application is very simple and is useful if you want to make a simple address book for example. If you intend to build a fully featured spatial database, you should use django-cities.
Requirements:
- Python >= 3.8
- Django >= 3.2
- MySQL or PostgreSQL or SQLite.
Yes, for some reason, code that used to work on MySQL (not without pain xD) does not work anymore. So we're now using django.db.transaction.atomic which comes from Django 1.6 just to support MySQL quacks.
- GraphQL support
- Built-in admin support
- Rest-Framework support
- Ajax Select Lookup support
See CHANGELOG.
Install django-cities-light:
pip install django-cities-light
Or the development version:
pip install -e [email protected]:yourlabs/django-cities-light.git#egg=cities_light
Add cities_light to your INSTALLED_APPS.
Configure filters to exclude data you don't want, ie.:
CITIES_LIGHT_TRANSLATION_LANGUAGES = ['fr', 'en'] CITIES_LIGHT_INCLUDE_COUNTRIES = ['FR'] CITIES_LIGHT_INCLUDE_CITY_TYPES = ['PPL', 'PPLA', 'PPLA2', 'PPLA3', 'PPLA4', 'PPLC', 'PPLF', 'PPLG', 'PPLL', 'PPLR', 'PPLS', 'STLMT',]
Now, run migrations, it will only create tables for models that are not disabled:
./manage.py migrate
Finally, populate your database with command:
./manage.py cities_light
This command is well documented, consult the help with:
./manage.py help cities_light
By default, update procedure attempts to update all fields, including Country/Region/Subregion/City slugs. But there is an option to keep them intact:
./manage.py cities_light --keep-slugs
The configuration parameter CITIES_LIGHT_CITY_SOURCES, comes with the default value http://download.geonames.org/export/dump/cities15000.zip that has cities with a population over 15000, if you need to load cities with less population please use another source. For the list of available source please check here: http://download.geonames.org/export/dump/readme.txt
Geonames.org is updated on daily basis and its full import is quite slow, so if you want to import the same data multiple times (for example on different servers) it is convenient to use fixtures with the helper management command:
./manage.py cities_light_fixtures dump ./manage.py cities_light_fixtures load
To reduce space, JSON fixtures are compressed with bzip2 and can be fetched from any HTTP server or local filesystem.
Consult the help with:
./manage.py help cities_light_fixtures
Create development virtualenv (you need to have tox installed in your base system):
tox -e dev source .tox/dev/bin/activate
Then run the full import:
test_project/manage.py migrate test_project/manage.py cities_light
There are several environment variables which affect project settings (like DB_ENGINE and CI), you can find them all in test_project/settings.py.
To run the test suite you need to have postgresql or mysql installed with passwordless login, or just use sqlite. Otherwise the tests which try to create/drop database will fail.
Running the full test suite:
tox
To run the tests in specific environment use the following command:
tox -e py37-django31-sqlite
And to run one specific test use this one:
tox -e py37-django31-sqlite -- cities_light/tests/test_form.py::FormTestCase::testCountryFormNameAndContinentAlone
To run it even faster, you can switch to specific tox virtualenv:
source .tox/py37-django18-sqlite/bin/activate CI=true test_project/manage.py test cities_light.tests.test_form.FormTestCase.testCountryFormNameAndContinentAlone
If you want to build the docs, use the following steps:
source .tox/dev/bin/activate cd docs make html
- Add ruff for formatting
- Improve the performance of the import command
- Improve the local development environment with https://tox.wiki/en/legacy/example/devenv.html
You could subscribe to the mailing list ask questions or just be informed of package updates.