django-utilitas 1.3.15

Django package with useful utility classes

Django Utilitas

Django Utilitas is a library that contains many ready-to-use utility classes so that you can just focus on the business logic of your API.

Utilitas itself is built upon many other great Django packages such as drf-ff, drf-yasg and the Django REST Framework itself.

Feature summary

Here is a summarized version of Utilitas' features:

• pagination
• sorting
• filtering (searching)
• swagger API support
• nested read serializers
• limiting the response' field
• bulk creation of models

Installation

pip install django-utilitas

Documentation

Docs are still in progress. I have assignments to do :(

Setting up a simple CRUD endpoint

Utilitas provides BaseListView, BaseDetails and BaseSearchView which are a more sophisticated version of DRF's default views. With them, you can set up a powerful API with about 9 lines of codes.

# views.py
class AccountListView(BaseListView):
    name = "Account list view" # optional
    model = Account # subclass of BaseModel
    serializer = AccountSerializer # subclass of BaseModelSerializer


class AccountDetailsView(BaseDetailsView):
    name = "Account details view"
    model = Account
    serializer = AccountSerializer


class AccountSearchView(BaseSearchView):
    name = "Account search view"
    model = Account
    serializer = AccountSerializer

Then, setup the urls like this. Please note that the argument must be named obj_id.

#urls.py
urlpatterns = [
    path("accounts/", views.AccountListView.as_view(), name="account-list"),
    path("accounts/<int:obj_id>", views.AccountDetailsView.as_view(), name="account-details"),
    path("accounts/search", views.AccountSearchView.as_view(), name="account-search"),
]

Then, when you go to your swagger endpoint, you should be able to see something like this:

Bulk Creation

When a POST request is made to list endpoints with the bulk query parameter set to a truthy value, a ListSerializer will be used to perform the creation process.

Request example

import requests

requests.post(
    "api/books/?bulk=True",
    {
        # need to set this 'objects' parameter in the request body
        "objects": [
            {"title": "The Hobbit", "author": "J. R. R. Tolkein"},
            {"title": "History of Burma", "author": "G. E. Harvey"}
        ]
    }   
)

Implementation

You can either inherit the Meta class from the BaseModelSerializer class

class BookSerializer(BaseModelSerializer):
    class Meta(BaseModelSerializer.Meta):
        model = Book
        fields = "__all__"

Or, set the list_serializer_class option to utilitas.serializers.BaseListSerializer. (more information here: https://www.django-rest-framework.org/api-guide/serializers/#listserializer)

from utilitas.serializers import BaseListSerializer

class BookSerializer(BaseModelSerializer):
    class Meta:
        model = Book
        fields = "__all__"
        list_serializer_class = BaseListSerializer

For more information about django-utilitas, please read the architecture document here

Getting a CSV response

Client just need to set a query parameter named csv to true. This works in BaseListView and BaseSearchView instances.

Request example

import requests

requests.get("api/books?csv=true")

Changelog

• 1.3.14

  • removed field validatoin for filter_param

• 1.3.13

  • bug fixes

• 1.3.12

  • added field validation in search endpoint's filter_param

• 1.3.11

  • made the '_translate_expand_params' function public

• 1.3.10

  • bulk-creation bug fixed. Dumb me didn't change anything in the views.py file D:

• 1.3.9

  • bug fixes

• 1.3.8

  • bug fixes

• 1.3.7

  • bug fixes

• 1.3.6

  • added an alias option for model fields. (user_friendly_field attribute)

• 1.3.5

  • bug fixes

• 1.3.4

  • bug fixes

• 1.3.3

  • forgot to add csv library. hehe.

• 1.3.2

  • added a csv-returning functionality.

• 1.3.1

  • now, the API won't return 500 if missing foreign keys are provided in the expand parameter.

• 1.3.0

  • removed the need for related_fields parameter. Prefetching related fields is now done automatically using the expand parameter provided by the client.

• 1.2.9

  • renamed file from migrations.py to middlewares.py (my stupid mistake in the first place)

• 1.2.8

  • bug fixes

• 1.2.7

  • bug fixes

• 1.2.6

  • bug fixes

• 1.2.5

  • added exclude_params option in search endpoints.

• 1.2.4

  • fixed bug in pagination

• 1.2.3

  • lowered required Python version

• 1.2.2

  • increased character limit for FilterParamSerializer's fields.

• 1.2.1

  • added a middleware for logging database queries in each request
  • bug fixes

• 1.2.0

  • added bulk creation

• 1.1.4

  • bug fixes

• 1.1.3

  • bug fixes

• 1.1.2 (the first usable version)

  • bug fixes