Cloud Spanner support for Django

ORM plugin for using Cloud Spanner as a database backend for Django.

How it works

Overall design

Internals

Installation

Using this library requires a Google Cloud Platform project with the Cloud Spanner API enabled. See the Spanner Python client documentation for details.

The version of django-google-spanner must correspond to your version of Django. For example, django-google-spanner 2.2.x works with Django 2.2.y (Note: this is the only supported version at this time).

The minor release numbers of Django may not correspond to the minor release numbers of django-google-spanner. Use the latest minor release of each.

To install from PyPI:

pip3 install django-google-spanner

To install from source:

git clone [email protected]:googleapis/python-spanner-django.git
cd python-spanner-django
pip3 install -e .

Configuring settings.py

After installation, you'll have to update your Django settings.py file as follows.

• Add django_spanner as the very first entry in the INSTALLED_APPS settings:

  INSTALLED_APPS = [
      'django_spanner',
      ...
  ]

• Edit the DATABASES settings to point to an EXISTING database, as shown in the following example:

  DATABASES = {
      'default': {
          'ENGINE': 'django_spanner',
          'PROJECT': '<GCP_project_id>',
          'INSTANCE': '<instance_id>',
          'NAME': '<database_name>',
      }
  }

• In order to retrieve the Cloud Spanner credentials from a JSON file, the credentials_uri parameter can also be supplied in the OPTIONS field:

  DATABASES = {
      'default': {
          'ENGINE': 'django_spanner',
          'PROJECT': '<GCP_project_id>',
          'INSTANCE': '<instance_id>',
          'NAME': '<database_name>',
          'OPTIONS': {
              'credentials_uri': '<credentials_uri>',
          },
      },
  }

Executing a query

from google.cloud.spanner_dbapi import connect

connection = connect('<instance_id>', '<database_id>')
cursor = connection.cursor()

cursor.execute(
    "SELECT *"
    "FROM Singers"
    "WHERE SingerId = 15"
)

results = cursor.fetchall()

Current limitations

AutoField generates random IDs

Spanner doesn't have support for auto-generating primary key values. Therefore, django-google-spanner monkey-patches AutoField to generate a random UUID4. It generates a default using Field's default option which means AutoFields will have a value when a model instance is created. For example:

>>> ExampleModel()
>>> ExampleModel.pk
4229421414948291880

To avoid hotspotting, these IDs are not monotonically increasing. This means that sorting models by ID isn't guaranteed to return them in the order in which they were created.

ForeignKey constraints aren't created (#313)

Spanner does not support ON DELETE CASCADE when creating foreign-key constraints, so this is not supported in django-google-spanner.

Check constraints aren't supported

Spanner does not support CHECK constraints so one isn't created for PositiveIntegerField and CheckConstraint can't be used.

No native support for DecimalField

Spanner's support for Decimal types is limited to NUMERIC precision. Higher-precision values can be stored as strings instead.

Variance and StdDev database functions aren't supported

Spanner does not support these functions.

Meta.order_with_respect_to model option isn't supported

This feature uses a column name that starts with an underscore (_order) which Spanner doesn't allow.

Random QuerySet ordering isn't supported

Spanner does not support it and will throw an exception. For example:

>>> ExampleModel.objects.order_by('?')
...
django.db.utils.ProgrammingError: 400 Function not found: RANDOM ... FROM
example_model ORDER BY RANDOM() ASC

Schema migrations

There are some limitations on schema changes to consider:

• No support for renaming tables and columns;
• A column's type can't be changed;
• A table's primary key can't be altered.

DurationField arithmetic doesn't work with DateField values (#253)

Spanner requires using different functions for arithmetic depending on the column type:

• TIMESTAMP columns (DateTimeField) require TIMESTAMP_ADD or TIMESTAMP_SUB
• DATE columns (DateField) require DATE_ADD or DATE_SUB

Django does not provide ways to determine which database function to use. DatabaseOperations.combine_duration_expression() arbitrarily uses TIMESTAMP_ADD and TIMESTAMP_SUB. Therefore, if you use a DateField in a DurationField expression, you'll likely see an error such as:

"No matching signature for function TIMESTAMP\_ADD for argument types:
DATE, INTERVAL INT64 DATE\_TIME\_PART."

Computations that yield FLOAT64 values cannot be assigned to INT64 columns

Spanner does not support this (#331) and will throw an error:

>>> ExampleModel.objects.update(integer=F('integer') / 2)
...
django.db.utils.ProgrammingError: 400 Value of type FLOAT64 cannot be
assigned to integer, which has type INT64 [at 1:46]\nUPDATE
example_model SET integer = (example_model.integer /...

Addition with null values crash

Additions cannot include None values. For example:

>>> Book.objects.annotate(adjusted_rating=F('rating') + None)
...
google.api_core.exceptions.InvalidArgument: 400 Operands of + cannot be literal
NULL ...