django-rank-query 0.1.0

Enable SQL Window functions for use in Django queries. Add rank to your query.

Enable SQL Window functions for use in Django queries. Add rank to your query.

Django in curren’t version (1.10) doesn’t support [window functions](https://www.postgresql.org/docs/9.3/static/functions-window.html) as seen in PostgreSQL. This implemets rank, dense_rank and adds shortcut for upper_rank, which retrieves upper value for rank (rank of last element in the window). The implementation is currently working only on PostgreSQL.

Note: The support for window functions is being already solved in [issue #26608](https://code.djangoproject.com/ticket/26608).

Quickstart

Install rank:

pip install django-rank-query

Usage

U can annotate users by rank (sequenc number in alphabet) of their last name the like this:

from rank import DenseRank, UpperRank, Rank
users = User.objects.all()
users = user.annotate(lower_rank=Rank('last_name'))
users = user.annotate(upper_rank=UpperRank('last_name'))
users = user.annotate(dense_rank=DenseRank('last_name'))

Then you have user query annotated with various ranks. You can retrieve them for example by values function:

user_ranks = users.values('last_name', 'lower_rank', 'upper_rank')

Note: PostgreSQL doesn’t support combining rank functions with GROUP_BY or in WHERE clause. So you cannot use them in aggregation queries.

Rank is counted in the current select query, so filtering will change rank of given elements, so you can’t have sequence number in whole table filtered by ie. his name. This can be solved by using subqueries in SQL, but Django doesn’t support them as far as I know.

Running Tests

Does the code actually work?

source <YOURVIRTUALENV>/bin/activate
(myenv) $ pip install tox
(myenv) $ tox

Credits

Tools used in rendering this package:

• Cookiecutter

• cookiecutter-djangopackage

History

0.1.0 (2017-01-20)

• First release on PyPI.