# Django Rest Framework Python API Package
[](https://travis-ci.org/dkarchmer/django-rest-framework-client)
[](https://pypi.python.org/pypi/django-rest-framework-client)
A python library for interacting with any Django web server base on django-rest-framework
Package is based on https://github.com/samgiles/slumber, but enhanced to support tokens and other features.
## Features
* Support for tokens. Both
* django-rest-framework's own tokens: `rest_framework.authentication.TokenAuthentication`
* JWT tokens: `rest_framework_jwt.authentication.JSONWebTokenAuthentication`
* Support for arguments (e.g. `?name1=val1&name2=val2`)
* Support for custom methods (e.g. ``/ap1/v1/object/custom/`)
## Requirements
restframeworkclient requires the following modules.
* Python 3.7+
* requests
## Installation
```bash
python3 -m venv ~/.virtualenv/drf_client
source ~/.virtualenv/drf_client/bin/activate
pip install django-rest-framework-client
```
## Usage Guide
Example
```
from drf_client.connection import Api as RestApi
options = {
'DOMAIN': 'http://127.0.0.1:8000',
'API_PREFIX': 'api/v1',
'TOKEN_TYPE': 'jwt',
'TOKEN_FORMAT': 'JWT {token}',
'USERNAME_KEY': 'username',
'LOGIN': 'auth/login/',
'LOGOUT': 'auth/logout/',
'USE_DASHES': False, # Set to True to tell API to replace undercore ("_") with dashes ("-")
}
c = RestApi(options)
ok = c.login(username=username, password=password)
if ok:
# GET some data
my_object = c.myresourcename.get()
for obj in my_object['results']:
pprint(obj)
logger.info('------------------------------')
payload = {
'data1': 'val1',
'data2': 'val2',
}
resp = c.myresourcename.post(data=payload)
# If the URL includes "-", add under parenthesis:
# GET: /api/v1/someresource/some-path/
my_object = c.someresource('some-path').get()
```
## Django Setup
Client assumes by default that all urls should end with a slash (tested with the default
router: `routers.DefaultRouter()`)
Apart from the regular Django and Rest Framework setup, this package currently relies on the following custom
login and logout API functions:
```
class AccountSerializer(serializers.ModelSerializer):
class Meta:
model = User
fields = ('id', 'email', 'username')
class APILogoutViewSet(APIView):
permission_classes = (permissions.IsAuthenticated,)
def post(self, request, format=None):
logout(request)
return Response({}, status=status.HTTP_204_NO_CONTENT)
urlpatterns = [
url(r'^auth/logout/$', APILogoutViewSet.as_view(), name='api-logout'),
```
## Helpers
### BaseMain Helper
This class helps write a script with a flexible template that helps avoid having to duplicate
boiler plate code from script to script.
The class assumes that most scripts include the basic folliwing flow:
```
# Parse arguments
# Setup LOG configuration
# Login
# Do something after logging in
```
The opinionated class will execute the basic main flow:
```python
# Initialize arguments and LOG in the init function
# Add additional arguments by implemenenting self.add_extra_args()
self.domain = self.get_domain()
self.api = Api(self.domain)
self.before_login()
ok = self.login()
if ok:
self.after_login()
```
Any of the above functions can be overwritten by derving from this class.
Here is a sample script:
```python
from drf_client.helper.base_main import BaseMain
class MyScript(BaseMain):
def add_extra_args(self):
# Add extra positional argument (as example)
self.parser.add_argument('foo', metavar='foo', type=str, help='RTFM')
def before_login(self):
logger.info('-----------')
def after_login(self):
# Main function to OVERWITE and do real work
do_some_real_work(self.api, self.args)
if __name__ == '__main__':
work = MyScript()
work.main()
```
Given the above script, you will run it with
```bash
python myscript.py -u <USERNAME> --foo bar
```
## Development
To test, run python setup.py test or to run coverage analysis:
```bash
python3 -m venv .virtualenv/drf_client
source .virtualenv/drf_client/bin/activate
pip install -r requirements-test.txt
pip install -e .
coverage run --source=iotile_cloud setup.py test
coverage report -m
```
You can also use py.test:
```bash
py.test
```
## CI Deployment
1. Update `setup.py` with new version
2. Update `CHANGELOG.md` with description of new version
2. Create new tag with same version
```
git tag v0.4.1 -m "v0.4.1"
git push --tags
```
3. Create new release using GitHub Web Site. Github action will run automatically to deploy to PyPi.
## Manual Deployment
```bash
pip install -r requirements-build.txt
python setup.py sdist bdist_wheel
twine check dist/*
# Publish
twine upload dist/*
```