Skip to main content

django-pg-returning

A small library implementing PostgreSQL ability to return rows in DML statements for Django.
Link to PostgreSQL docs

Requirements

  • Python Python 3.5+ Previous versions may also work, but are not tested with CI
  • django >= 1.8
    Previous versions may also work, but are not tested with CI.
    bulk_create_returning method doesn't support .only() and .defer() filters for django before 1.10.
  • psycopg2-binary
  • typing for python < 3.5
  • PostgreSQL 9.4+
    Previous versions may also work, but are not tested with CI.

Installation

Install via pip:
pip install django-pg-returning
or via setup.py:
python setup.py install

Usage

Integration

The easiest way to integrate, is to inherit your model from UpdateReturningModel instead of django.db.models.Model. It already has redeclared Manager, supporting returning operations.

from django.db import models
from django_pg_returning import UpdateReturningModel

class MyModel(UpdateReturningModel):   
    field = models.IntegerField()

If you already have custom manager, you can implement get_queryset() method in it:

from django.db import models
from django_pg_returning import UpdateReturningQuerySet, UpdateReturningModel

class MyManager(models.Manager):
    def get_queryset(self):
        return UpdateReturningQuerySet(using=self.db, model=self.model)

class MyModel(UpdateReturningModel):
    objects = MyManager()

    field = models.IntegerField()

And if you have custom manager you can use a mixin:

from django.db import models
from django_pg_returning import UpdateReturningMixin, UpdateReturningModel

class MyQuerySet(models.QuerySet, UpdateReturningMixin):
    pass

class MyManager(models.Manager):
    def get_queryset(self):
        return MyQuerySet(using=self.db, model=self.model)

class MyModel(UpdateReturningModel):
    objects = MyManager()

    field = models.IntegerField()

Methods

QuerySet methods

After QuerySet mixin is integrated with your model, your QuerySet-s will have 3 additional methods:

from django.db.models import Value

# Any django queryset you like
qs = MyModel.objects.all()

# Update and return a ReturningQuerySet, described below
result = qs.update_returning(field=1)

# Delete data and return a ReturningQuerySet, described below
result = qs.delete_returning()

# Acts like django's QuerySet.create() method, but updates all model fields to values stored in database
# Can be used to retrieve values, saved by database default/triggers etc.
result = MyModel.objects.create_returning(field=Value(1) + Value(2))
print(result.field)  # prints: "3" instead of "Value(1) + Value(2)"

# Acts like django's QuerySet.bulk_create() method, but updates all model fields to values stored in database
# Can be used to retrieve values, saved by database default/triggers etc.
result = MyModel.objects.bulk_create_returning([MyModel(field=Value(1) + Value(2))])
print(result[0].field)  # prints: "3" instead of "Value(1) + Value(2)"

By default methods get all fields, fetched by the model. To limit fields returned, you can use standard QuerySet.only() and QuerySet.defer() methods.
create_returning doesn't support these methods.
bulk_create_returning doesn't support these methods for django before 1.10.

Model methods

If model instance is created, basic save() method is called.
If model is updated, database record is updated, and saved fields are refreshed with database values. This may be useful, if you update fields with F() expressions. By default all fields are saved and refreshed. Use update_fields to specify concrete fields to save and refresh.

from django.db.models import Value, F

instance = MyModel(pk=1, field=Value(1))
instance.save_returning()
print(instance.field)
# Output: 2 
# if basic save() called: F('field') + Value(1)

instance.field = F('field') + 1

# Basic save method will not change field and you don't know, what value is in database
instance.save()
print(instance.field)
# Output: F('field') + Value(1)

# Library method gives ability to fetch updated result 
instance.save_returning()
print(instance.field)
# Output: 2

Important notes:

  1. If you don't fetch field, and then try to get it, library acts as django does - makes extra database query to fetch attribute deferred.
  2. These queries are not lazy, as well as basic QuerySet.update() and QuerySet.delete() methods.
  3. Primary key field is fetched not looking at limiting methods, as django needs it to form a QuerySet

ReturningQuerySet

The result of returning functions is django_pg_returning.ReturningQuerySet. It is based on django's RawQuerySet, but adds some extra methods to be used easier. The main difference is that ReturningQuerySet caches query results, while RawQuerySet executes query each time it is iterated. All ReturningQuerySet methods are not executed on database side, they are executed in python on cached result. The only way, ReturningQuerySet makes extra database query - is deferred field loading, described above. Implemented methods:

# UPDATE ... RETURNING query is executed here once. The result is cached.
result = MyModel.objects.all().update_returning(field=1)

# Get number of values fetched
print(result.count(), len(result))
# Output: 1, 1

# Index and slicing. Note that the order of result is not guaranteed by the database.
print(result[1], result[0:2])
# Output: MyModel(...), [MyModel(...), MyModel(...), MyModel(...)]

# Sintax sugar for indexing
print(result.first(), result.last())
# Output: MyModel(...), MyModel(...)

# Fetching values and values_list. Both methods use cache and return lists, not ValuesQuerySet like django does.
# values() method cakked without fields will return all fields, fetched in returning method.
# values_list() method called without fields will raise exception, as order or fields in result tuple is not obvious.

print(result.values())
# Output: [{'id': 1, 'field': 1}, {'id': 2, 'field': 2}]

print(result.values('field'))
# Output: [{'field': 1}, {'field': 2}]

print(result.values_list('field', flat=True))
# Output: [1, 2]

print(result.values_list('field', 'id', named=True))
# Output: [Row(field=1, id=1), Row(field=2, id=2)]

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

django-pg-returning-1.3.1.tar.gz (12.6 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

django_pg_returning-1.3.1-py2.py3-none-any.whl (12.6 kB view details)

Uploaded Python 2Python 3

File details

Details for the file django-pg-returning-1.3.1.tar.gz.

File metadata

  • Download URL: django-pg-returning-1.3.1.tar.gz
  • Upload date:
  • Size: 12.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.3.0 pkginfo/1.6.1 requests/2.25.1 setuptools/49.2.1 requests-toolbelt/0.9.1 tqdm/4.55.1 CPython/3.9.1

File hashes

Hashes for django-pg-returning-1.3.1.tar.gz
Algorithm Hash digest
SHA256 d06a3b2ba66284a1b9f84eb6140af5f852c2203faf58d3083fd1776f2ae15fe8
MD5 c78e04cfa40343773df36865052c47c2
BLAKE2b-256 15817f18bff43bd65b081b83cd626b3350140ccc670c9f0823a5c2a5c7c70922

See more details on using hashes here.

File details

Details for the file django_pg_returning-1.3.1-py2.py3-none-any.whl.

File metadata

  • Download URL: django_pg_returning-1.3.1-py2.py3-none-any.whl
  • Upload date:
  • Size: 12.6 kB
  • Tags: Python 2, Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.3.0 pkginfo/1.6.1 requests/2.25.1 setuptools/49.2.1 requests-toolbelt/0.9.1 tqdm/4.55.1 CPython/3.9.1

File hashes

Hashes for django_pg_returning-1.3.1-py2.py3-none-any.whl
Algorithm Hash digest
SHA256 6b17221bcd1eebd0fa54da3f382b00db07a10be4ce43c735f3490ef5b6098512
MD5 12953f7b059a1225e3ed21193865d2ba
BLAKE2b-256 fa37b93c9f982020aa6ad8b0276d9e0da9d6856644cce89644e90e0959f68389

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page