a simple, extensible rating system.

dependencies:

• django-generic-aggregation: http://github.com/coleifer/django-generic-aggregation

also, a playground for different stuff i've been reading about in programming collective intelligence, by toby segaran.

this stuff lives in utils.py and is there if you want to experiment (or contribute!).

you'd like to add ratings to some model:

from django.db import models
from ratings.models import Ratings

class Food(models.Model):
    name = models.CharField(max_length=50)

    ratings = Ratings()

now, you can:

# add ratings to things
>>> apple.ratings.rate(user=john, score=1)
<RatedItem: apple rated 1 by john>

>>> apple.ratings.rate(user=jane, score=5)
<RatedItem: apple rated 5 by jane>

# get interesting aggregate data
>>> apple.ratings.all()
[<RatedItem: apple rated 1 by john>, <RatedItem: apple rated 5 by jane>]

>>> apple.ratings.cumulative_score()
6

>>> apple.ratings.average_score()
3.0

# order things by their rating
>>> Food.ratings.order_by_rating()
[<Food: apple>, <Food: orange>]

By default, whenever you add Ratings() to your model it uses the RatedItem model which has a GFK on it. Suppose you are only rating one thing, or would like to have an explicit database constraint -- that's no problem. You can provide a custom RatedItem model with a ForeignKey instead of a GFK. Here's the example from the tests:

class BeverageRating(RatedItemBase):
    content_object = models.ForeignKey('Beverage')


class Beverage(models.Model):
    name = models.CharField(max_length=50)

    ratings = Ratings(BeverageRating)

    def __unicode__(self):
        return self.name

The API is exactly the same.