Merge pull request #4 from M1hacka/cardinality-sum-function

Added CardinalitySum function

README.md

@@ -106,18 +106,21 @@ MyModel.objects.filter(hll__cardinality=3).count()
 ```
 
 ### Aggregate functions
-In order to count aggregations and annotations, library provides 3 aggregate functions:
+In order to count aggregations and annotations, library provides 4 aggregate functions:
 * `django_pg_hll.aggregate.Cardinality`
   Counts cardinality of hll field
 * `django_pg_hll.aggregate.UnionAgg`
   Aggregates multiple hll fields to one hll.
 * `django_pg_hll.aggregate.UnionAggCardinality`
   Counts cardinality of hll, combined by UnionAgg function. In fact, it does `Cardinality(UnionAgg(hll))`.  
   P. s. django doesn't give ability to use function inside function.
+* `django_pg_hll.aggregate.CardinalitySum`
+  Counts sum of multiple rows hll cardinalities. In fact, it does `Sum(Cardinality(hll))`.  
+  P. s. django doesn't give ability to use function inside function.
 ```python
 from django.db import models
 from django_pg_hll import HllField, HllInteger
-from django_pg_hll.aggregate import Cardinality, UnionAggCardinality
+from django_pg_hll.aggregate import Cardinality, UnionAggCardinality, CardinalitySum
 
 
 class ForeignModel(models.Model):
@@ -130,16 +133,21 @@ class MyModel(models.Model):
 
 MyModel.objects.bulk_create([
    MyModel(fk=1, hll=HllInteger(1)),
-   MyModel(fk=2, hll=HllInteger(2) | HllInteger(3)),
+   MyModel(fk=2, hll=HllInteger(2) | HllInteger(3) | HllInteger(4)),
    MyModel(fk=3, hll=HllInteger(4))
 ])
 
 MyModel.objects.annotate(card=Cardinality('hll_field')).values_list('id', 'card')
-# outputs (1, 1), (2, 2), (3, 1)
+# outputs (1, 1), (2, 3), (3, 1)
 
-# Count cardinality for hll, built from 
+# Count cardinality for hll, build by union of all rows
+# 4 element exists in rows with fk=2 and fk=3. After union it gives single result 
 ForeignModel.objects.annotate(card=UnionAggCardinality('testmodel__hll_field')).values_list('card', flat=True)
 # outputs [4]
+
+# Count sum of cardinalities for each row
+ForeignModel.objects.annotate(card=CardinalitySum('testmodel__hll_field')).values_list('card', flat=True)
+# outputs [5]
 ```
 
 ### [django-pg-bulk-update](https://github.com/M1hacka/django-pg-bulk-update) integration

setup.py

@@ -13,7 +13,7 @@
 
 setup(
     name='django-pg-hll',
-    version='1.1.0',
+    version='1.2.0',
     packages=['django_pg_hll'],
     package_dir={'': 'src'},
     url='https://github.com/M1hacka/django-pg-hll',

src/django_pg_hll/aggregate.py

@@ -13,11 +13,21 @@ class UnionAgg(Aggregate):
     output_field = HllField()
 
 
-class UnionAggCardinality(UnionAgg):
+class UnionAggCardinality(Aggregate):
     """
     I haven't found a way to combine function inside function in django.
     So, I've written function to get aggregate cardinality with one call
     """
     function = 'hll_union_agg'
     template = 'hll_cardinality(%(function)s(%(expressions)s))'
     output_field = IntegerField()
+
+
+class CardinalitySum(Aggregate):
+    """
+    I haven't found a way to combine function inside function in django.
+    So, I've written function to get sum cardinality with one call
+    """
+    function = 'hll_cardinality'
+    template = 'SUM(%(function)s(%(expressions)s))'
+    output_field = IntegerField()

tests/test_hll_field.py

@@ -6,7 +6,7 @@
 from django.test import TestCase
 
 from django_pg_hll import HllEmpty, HllInteger
-from django_pg_hll.aggregate import Cardinality, UnionAgg, UnionAggCardinality
+from django_pg_hll.aggregate import Cardinality, UnionAgg, UnionAggCardinality, CardinalitySum
 
 # !!! Don't remove this import, or bulk_update will not see function name
 from django_pg_hll.bulk_update import HllConcatFunction
@@ -92,17 +92,18 @@ class TestAggregation(TestCase):
     def setUp(self):
         TestModel.objects.bulk_create([
             TestModel(id=100501, hll_field=HllEmpty()),
-            TestModel(id=100502, hll_field=HllInteger(1)),
+            TestModel(id=100502, hll_field=HllInteger(1) | HllInteger(2)),
             TestModel(id=100503, hll_field=HllInteger(2))
         ])
 
     def test_cardinality_transform_filter(self):
         self.assertEqual(1, TestModel.objects.filter(hll_field__cardinality=0).count())
-        self.assertEqual(2, TestModel.objects.filter(hll_field__cardinality=1).count())
-        self.assertEqual(0, TestModel.objects.filter(hll_field__cardinality=2).count())
+        self.assertEqual(1, TestModel.objects.filter(hll_field__cardinality=1).count())
+        self.assertEqual(1, TestModel.objects.filter(hll_field__cardinality=2).count())
+        self.assertEqual(0, TestModel.objects.filter(hll_field__cardinality=3).count())
 
     def test_cardinality_aggregate_function(self):
-        self.assertEqual({0, 1}, set(TestModel.objects.annotate(card=Cardinality('hll_field')).
+        self.assertEqual({0, 1, 2}, set(TestModel.objects.annotate(card=Cardinality('hll_field')).
                          values_list('card', flat=True)))
 
     def test_union_aggregate_function(self):
@@ -126,6 +127,15 @@ def test_union_aggregate_cardinality_function(self):
 
         self.assertEqual(2, card)
 
+    def test_cardinality_sum_function(self):
+        fk_instance = FKModel.objects.create()
+        TestModel.objects.all().update(fk=fk_instance)
+
+        card = FKModel.objects.annotate(card=CardinalitySum('testmodel__hll_field')).\
+            values_list('card', flat=True)[0]
+
+        self.assertEqual(3, card)
+
 
 @skipIf(not django_pg_bulk_update_available(), 'django-pg-bulk-update library is not installed')
 class TestBulkUpdate(TestCase):