These functions are available from the django.contrib.postgres.aggregates\nmodule. They are described in more detail in the PostgreSQL docs.
ArrayAggLink to this heading#Returns a list of values, including nulls, concatenated into an array, or\ndefault if there are no values.
An optional boolean argument that determines if array values\nwill be distinct. Defaults to False.
An optional string of a field name (with an optional "-" prefix\nwhich indicates descending order) or an expression (or a tuple or list\nof strings and/or expressions) that specifies the ordering of the\nelements in the result list.
Examples:
\nfrom django.db.models import F\n\nArrayAgg("a_field", order_by="-some_field")\nArrayAgg("a_field", order_by=F("some_field").desc())\nBitAndLink to this heading#Returns an int of the bitwise AND of all non-null input values, or\ndefault if all values are null.
BitOrLink to this heading#Returns an int of the bitwise OR of all non-null input values, or\ndefault if all values are null.
BitXorLink to this heading#Returns an int of the bitwise XOR of all non-null input values, or\ndefault if all values are null.
BoolAndLink to this heading#Returns True, if all input values are true, default if all values\nare null or if there are no values, otherwise False.
Usage example:
\nclass Comment(models.Model):\n body = models.TextField()\n published = models.BooleanField()\n rank = models.IntegerField()\n>>> from django.db.models import Q\n>>> from django.contrib.postgres.aggregates import BoolAnd\n>>> Comment.objects.aggregate(booland=BoolAnd("published"))\n{'booland': False}\n>>> Comment.objects.aggregate(booland=BoolAnd(Q(rank__lt=100)))\n{'booland': True}\nBoolOrLink to this heading#Returns True if at least one input value is true, default if all\nvalues are null or if there are no values, otherwise False.
Usage example:
\nclass Comment(models.Model):\n body = models.TextField()\n published = models.BooleanField()\n rank = models.IntegerField()\n>>> from django.db.models import Q\n>>> from django.contrib.postgres.aggregates import BoolOr\n>>> Comment.objects.aggregate(boolor=BoolOr("published"))\n{'boolor': True}\n>>> Comment.objects.aggregate(boolor=BoolOr(Q(rank__gt=2)))\n{'boolor': False}\nJSONBAggLink to this heading#Returns the input values as a JSON array, or default if there are\nno values. You can query the result using key and index lookups.
An optional boolean argument that determines if array values will be\ndistinct. Defaults to False.
An optional string of a field name (with an optional "-" prefix\nwhich indicates descending order) or an expression (or a tuple or list\nof strings and/or expressions) that specifies the ordering of the\nelements in the result list.
Examples are the same as for ArrayAgg.order_by.
Usage example:
\nclass Room(models.Model):\n number = models.IntegerField(unique=True)\n\n\nclass HotelReservation(models.Model):\n room = models.ForeignKey("Room", on_delete=models.CASCADE)\n start = models.DateTimeField()\n end = models.DateTimeField()\n requirements = models.JSONField(blank=True, null=True)\n>>> from django.contrib.postgres.aggregates import JSONBAgg\n>>> Room.objects.annotate(\n... requirements=JSONBAgg(\n... "hotelreservation__requirements",\n... order_by="-hotelreservation__start",\n... )\n... ).filter(requirements__0__sea_view=True).values("number", "requirements")\n<QuerySet [{'number': 102, 'requirements': [\n {'parking': False, 'sea_view': True, 'double_bed': False},\n {'parking': True, 'double_bed': True}\n]}]>\nStringAggLink to this heading#Returns the input values concatenated into a string, separated by\nthe delimiter string, or default if there are no values.
Required argument. A string, Value, or\nexpression representing the string for separating values. For example,\nValue(",").
An optional boolean argument that determines if concatenated values\nwill be distinct. Defaults to False.
An optional string of a field name (with an optional "-" prefix\nwhich indicates descending order) or an expression (or a tuple or list\nof strings and/or expressions) that specifies the ordering of the\nelements in the result string.
Examples are the same as for ArrayAgg.order_by.
Usage example:
\nclass Publication(models.Model):\n title = models.CharField(max_length=30)\n\n\nclass Article(models.Model):\n headline = models.CharField(max_length=100)\n publications = models.ManyToManyField(Publication)\n>>> article = Article.objects.create(headline="NASA uses Python")\n>>> article.publications.create(title="The Python Journal")\n<Publication: Publication object (1)>\n>>> article.publications.create(title="Science News")\n<Publication: Publication object (2)>\n>>> from django.contrib.postgres.aggregates import StringAgg\n>>> Article.objects.annotate(\n... publication_names=StringAgg(\n... "publications__title",\n... delimiter=", ",\n... order_by="publications__title",\n... )\n... ).values("headline", "publication_names")\n<QuerySet [{\n 'headline': 'NASA uses Python', 'publication_names': 'Science News, The Python Journal'\n}]>\ny and xLink to this heading#The arguments y and x for all these functions can be the name of a\nfield or an expression returning a numeric data. Both are required.
CorrLink to this heading#Returns the correlation coefficient as a float, or default if there\naren’t any matching rows.
CovarPopLink to this heading#Returns the population covariance as a float, or default if there\naren’t any matching rows.
Optional. By default CovarPop returns the general population\ncovariance. However, if sample=True, the return value will be the\nsample population covariance.
RegrAvgXLink to this heading#Returns the average of the independent variable (sum(x)/N) as a\nfloat, or default if there aren’t any matching rows.
RegrAvgYLink to this heading#Returns the average of the dependent variable (sum(y)/N) as a\nfloat, or default if there aren’t any matching rows.
RegrCountLink to this heading#Returns an int of the number of input rows in which both expressions\nare not null.
RegrInterceptLink to this heading#Returns the y-intercept of the least-squares-fit linear equation determined\nby the (x, y) pairs as a float, or default if there aren’t any\nmatching rows.
RegrR2Link to this heading#Returns the square of the correlation coefficient as a float, or\ndefault if there aren’t any matching rows.
RegrSlopeLink to this heading#Returns the slope of the least-squares-fit linear equation determined\nby the (x, y) pairs as a float, or default if there aren’t any\nmatching rows.
RegrSXXLink to this heading#Returns sum(x^2) - sum(x)^2/N («sum of squares» of the independent\nvariable) as a float, or default if there aren’t any matching rows.
RegrSXYLink to this heading#Returns sum(x*y) - sum(x) * sum(y)/N («sum of products» of independent\ntimes dependent variable) as a float, or default if there aren’t\nany matching rows.
RegrSYYLink to this heading#Returns sum(y^2) - sum(y)^2/N («sum of squares» of the dependent\nvariable) as a float, or default if there aren’t any matching rows.
We will use this example table:
\n| FIELD1 | FIELD2 | FIELD3 |\n|--------|--------|--------|\n| foo | 1 | 13 |\n| bar | 2 | (null) |\n| test | 3 | 13 |\nHere’s some examples of some of the general-purpose aggregation functions:
\n>>> TestModel.objects.aggregate(result=StringAgg("field1", delimiter=";"))\n{'result': 'foo;bar;test'}\n>>> TestModel.objects.aggregate(result=ArrayAgg("field2"))\n{'result': [1, 2, 3]}\n>>> TestModel.objects.aggregate(result=ArrayAgg("field1"))\n{'result': ['foo', 'bar', 'test']}\nThe next example shows the usage of statistical aggregate functions. The\nunderlying math will be not described (you can read about this, for example, at\nwikipedia):
\n>>> TestModel.objects.aggregate(count=RegrCount(y="field3", x="field2"))\n{'count': 2}\n>>> TestModel.objects.aggregate(\n... avgx=RegrAvgX(y="field3", x="field2"), avgy=RegrAvgY(y="field3", x="field2")\n... )\n{'avgx': 2, 'avgy': 13}\n