Skip to main content

Put your functions to REST

Project description


Deploy a REST API with one line by decorating your functions.


The package is available for all Mac, Linux, and Windows on my conda channel. Python 2 is not supported.

> conda install -c conda-forge tranquilizer

Quick start

Tranquilizer can be used with either Jupyter Notebooks (.ipynb) or Python script files (.py).

The decorated function below will be served as an end point called cheese with the GET method. The function must return a JSON serializable object.

See the complete description of @tranquilize() below.

from tranquilizer import tranquilize

def order(cheese):
    '''I'd like to buy some cheese!'''
    return "I'm afraid we're fresh out of {}, Sir.".format(cheese)

The REST API is served by Flask and Flask-RESTX using the tranquilizer command.

> tranquilizer cheese_shop.ipynb
 * Serving Flask app "tranquilizer.application" (lazy loading)
 * Environment: production
   WARNING: Do not use the development server in a production environment.
   Use a production WSGI server instead.
 * Debug mode: off
 * Running on (Press CTRL+C to quit)

Let's see if there is any Red Leicester.

> curl -G http://localhost:8086/order --data-urlencode "cheese=Red Leicester"
"I'm afraid we're fresh out of Red Leicester, Sir."

or using the Requests library in Python.

In [1]: import requests

In [2]: response = requests.get('http://localhost:8086/order', params={'cheese':'Red Leicester'})

In [3]: response.text
Out[3]: '"I\'m afraid we\'re fresh out of cheddar, Sir."\n'

The tranquilized API is documented with Swagger and is accessible in your web browser at http://localhost:8086.

Tranquilize Decorator

The @tranqulize decorator will assign the GET method by default. POST is also supported with method='post'. Other methods are under consideration.

By default a tranquilized function will receive all inputs as strings. This behavior can be modified by using type hints. When data is received by the Flask server it will use the provided type function to transform the string to the requested data type. This avoids having to perform the conversion in your tranquilized function.

Supported source formats

Tranquilizer can serve functions written in Python source (.py) files or Jupyter Notebooks (.ipynb).

When working interactively in Jupyter Notebooks the decorated functions will continue to operate as normal. Note that all calls to Jupyter Magic and Shell (!) commands will be ignored when the REST API is served. Only those lines will be ignored, the rest of the cell will continue to run.

Data Types

In addition to builtin types Tranquilizer provides specialized support for Lists, date/datetime, and files.

Type Description or datetime.datetime Converts string with dateutil.parser.parse and returns specified type.
list Converts repeated arguments to a list of strings.
typing.List[<type>] Converts repeated arguments to a list; each value is converted to <type>.

List arguments are constructed using the action='append' argument described in the Flask RESTX documentation. Any valid type can be used in List[].

The following file-like types are handled by werkzeug FileStorage. FileStorage is a file-like object that supports methods like .read() and .readlines(). These types support sending files with cURL using -F.

Type Description
typing.BinaryIO File-like object to read binary data.
typing.TextIO Converts FileStorage type to io.StringIO().

Further, specific support for Image and NumPy files are provided. The binary contents of the file are automatically converted.

Type Description
PIL.Image.Image Converts FileStorage type to PIL Image.
numpy.ndarray Converts FileStorage type to NumPy array using np.load().

Custom types

Custom type classes can be built...

Type hints example

The example below uses int, datetime.datetime, and typing.List. datetime.datetime support has been built with datetutil and will convert any compatible datetime string to a datetime.datetime object. typing.List supports specialization with [] and will transform all repeated arguments passed to the REST API into a list and convert the type of each element.

Finally, tranquilizer supports default arguments.

from tranquilizer import tranquilize
from datetime import date
from typing import List

def convert(string: str, date: date, items: List[float], factor: int = 10):
    '''Let's convert strings to something useful'''

    new_items = [i * factor for i in items]

    response = {
            'string': string.upper(),
            'date'  : date.strftime('%c'),
            'items' : new_items

    return response

Let's see what happens when I POST to this REST API.

In [1]: data = {'string':'hello, world!', 'date':'4th July 1776', 'items':range(5)}

In [2]: import requests

In [3]: response ='http://localhost:8086/convert', data=data)

In [4]: response.json()
{'date': 'Thu Jul  4 00:00:00 1776',
 'items': [0.0, 10.0, 20.0, 30.0, 40.0],
 'string': 'HELLO, WORLD!'}

In [5]:

Project details

Download files

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

Files for tranquilizer, version 0.6.0
Filename, size File type Python version Upload date Hashes
Filename, size tranquilizer-0.6.0.tar.gz (33.7 kB) File type Source Python version None Upload date Hashes View

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring DigiCert DigiCert EV certificate Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page