Skip to main content

An identity mapper for the Django ORM

Project description

© 2014-2015 Thomas Khyn © 2009 David Cramer

An identity mapper for the Django ORM. This is a fork of django-idmapper, which is no longer maintained.

What is it?

django-idmap is a Django application which:

  • loads only once the instances in memory the first time they are needed

  • share them througout your interpreter until the request is finished

Indeed, the default django behavior is to expose different instances for the same database entry between the start and the end of the request. It has one main consequence: the temporary attributes you may set are lost if you want to access the same database object in another place in your code.

Tested with django 1.8+ and latest minor versions of python 2 and 3.

Installation

As straightforward as it can be, using pip:

pip install django-idmap

You then need to add 'idmap' to your INSTALLED_APPS.

Usage

To use the exposed shared memory model you simply need to inherit from it (instead of models.Model). This enable all queries (and relational queries) to this model to use the shared memory instance cache, effectively creating a single instance for each unique row (based on primary key) in the queryset.

You can chose between 2 caching modes:

  • Weak references mode: the instance will be removed from the cache once there are no more references to it. This is the default behavior

  • Strong references mode: the instance will only be removed from the cache when it is flushed

Note that django-idmap clears the cache when the request_finished or post_syncdb signal is sent. This default behavior can be modified by disconnecting the exposed idmap.flush_cache function from these signals.

Examples

You can import idmap.models as you would import django.db.models. idmap.models exposes all what is exposed by django.db.models plus the SharedMemoryModel model class.

You may mix and match SharedMemoryModels with regular Models:

from idmap import models

class MyModel(models.SharedMemoryModel):
    name = models.CharField(...)
    fkey = models.ForeignKey('Other')

class Other(models.Model):
    name = models.CharField(...)

If you want to use strong references for a particular model, simply set use_strong_refs to True in the derived model class:

from idmap import models

class MyModel(models.SharedMemoryModel):
   use_strong_refs = True
   [...]

With strong references, the model will be loaded only once from the database, until it is explicitly erased from the cache.

You may want to use the functions or class methods:

  • idmap.flush_cache() to erase the whole cache

  • SharedMemoryModel.flush_instance_cache() to erase the cache for one model

  • SharedMemoryModel.flush_cached_instance(instance) to erase one instance from the cache

References

David Cramer’s django-idmapper

Original code and concept: http://code.djangoproject.com/ticket/17

Project details


Download files

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

Source Distribution

django-idmap-0.4.1.zip (14.3 kB view details)

Uploaded Source

File details

Details for the file django-idmap-0.4.1.zip.

File metadata

  • Download URL: django-idmap-0.4.1.zip
  • Upload date:
  • Size: 14.3 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No

File hashes

Hashes for django-idmap-0.4.1.zip
Algorithm Hash digest
SHA256 e54adc9976e0c8517e02ff3e1d451f3b4c7a0fc47043562c8e10f9603a1d6ecf
MD5 e2171208f8e10465fb36eca4e8408327
BLAKE2b-256 054635c7af903806f75dbfa0c9b2d0f26aa1093fef9711f3cabc14e8a3c8b659

See more details on using hashes here.

Supported by

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