Skip to main content

build Django REST Framework API from jsonschema

Project description


djangorestframework-jsonschema


Overview

This package provides a management command that builds a Django-REST-Framework solution from a JSONSchema specification.

Requirements

  • Python (3.8+)
  • Django (3.0+)
  • Django-Rest-Framework (3.11)
  • JSONSchema (3.2)
  • Jinja2 (2.11)

Installation

Install using pip...

$ pip install djangorestframework-jsonschema

This app needs to included it in the INSTALLED_APPS of your project, it should come after rest_framework and django_filters, both of which are required, but before anything specific to your project.

INSTALLED_APPS = [
    "django.contrib.admin",
    ...
    "rest_framework",
    "django_filters",
    "jsonschema2dj",
    ...
]

Example

Save the following as schema.json in a new DRF app example_app

{
  "definitions": {
    "identifiers": {
      "properties": {
        "ISBN": {
          "type": "string",
          "pattern": "[0-9]*[-| ][0-9]*[-| ][0-9]*[-| ][0-9]*[-| ][0-9]*"
        },
        "Dewey Decimal Classification": {
          "type": "string",
          "pattern": "\\d{3}|\\d{3}\\.\\d+|[12456]--\\d+|3[ABC]?--\\d+"
        }
      }
    },
    "Book": {
      "properties": {
        "title": {
          "type": "string"
        },
        "other": {"$ref": "#/definitions/identifiers"},
        "genre": {
          "enum": [
            "celebrity-nonsense",
            "military-tat",
            "other"
          ]
        },
        "author": {
          "$ref": "#/definitions/Author"
        }
      }
    }
  },
  "properties":{
    "Book": {"$ref":  "#/definitions/Book"},
    "Author": {
      "properties": {
        "name": {
          "type": "string"
        },
        "date_of_birth": {
          "type": "string",
          "format": "date"
        }
      }
    }
  }
}

now run

$ python manage.py jsonschema2dj example_app

and the following are produced:

  • models.py
  • views.py
  • serializers.py
  • urls.py
  • filters.py
  • admin.py

These files are built using jinja templates using sensible default choices where possible. It is expected that the user will modify these files to suit their needs.

eg. models.py would be:

class Book(models.Model):

    title = models.CharField(null=True, max_length=255)
    other = JSONField(validators=[JSONSchemaValidator({'$ref': '#/definitions/identifiers'}, DEFINITIONS)])
    genre = models.CharField(null=True, max_length=25, choices=[
        ('celebrity_nonsense', 'celebrity nonsense'),
        ('military-tat', 'military-tat'),
        ('other', 'other')
    ])
    author = models.ForeignKey("Author", null=True, on_delete=models.CASCADE)


class Author(models.Model):

    name = models.CharField(null=True, max_length=255)
    date_of_birth = models.DateField(null=True)

This is intended to be:

  • accessible to anyone with knowledge of JSONSchema
  • extensible by anyone with a rudimentary understanding of Django

Rules

Models are objects at the root level of the properties of the schema.json.

A model's fields are its root level properties.

The definitions maybe used freely without constraint.

There are three types of field:

simple fields

If a field has a jsonschema:type that is anything other than an object or items or this can be inferred then this field is mapped to a djngo field approximately as:

  • "string" -> CharField
  • "integer" -> IntegerField
  • "number" -> DecimalField
  • "boolean" -> BooleanField

If the field's name is id then this wil be the primary key.

reference-fields

If a field has a jsonschema:type that is an object or items and this references an object (model) with a schema also defined in the top level properties then this is modeled as relationship to that object (model) like:

  • "$ref": "Model-X" -> one-to-one or one-to-many
  • "items": {"$ref": "Model-X"} -> many-to-one or many-to-many

Note that the exact cardinality can only be determined by comparing both sides of the relationship. If only one side is specified the it is assumed that it is one-to-many or many-to-many respectively.

json-fields

Lastly if a field has a jsonschema:type that is an object or items and this references an object (model) with a schema that is not defined in the top level properties then this is modeled as a jsons object in its own right, i.e.:

  • "object" -> JSONField

Its schema will be enfoced by the serializer.

Testing

$ ./runtests.py

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

djangorestframework-jsonschema-0.1.1.tar.gz (13.1 kB view details)

Uploaded Source

File details

Details for the file djangorestframework-jsonschema-0.1.1.tar.gz.

File metadata

  • Download URL: djangorestframework-jsonschema-0.1.1.tar.gz
  • Upload date:
  • Size: 13.1 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.1.1 pkginfo/1.4.2 requests/2.22.0 setuptools/45.2.0 requests-toolbelt/0.8.0 tqdm/4.30.0 CPython/3.8.2

File hashes

Hashes for djangorestframework-jsonschema-0.1.1.tar.gz
Algorithm Hash digest
SHA256 73b2209ef1c72ce03e8255223476762eb8463dd21046bd8a6fa4f4c641e5d923
MD5 f879341f4f91494ae07f0067bd5a42c7
BLAKE2b-256 ff2f2c439578c0a81ae5f4ddd3f21f1c1812935733c8a344492910d5ed26fb2c

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