Consuming APIs with Python
Download as PPTX, PDF2 likes608 views
This document discusses consuming APIs with Python. It introduces the requests library for making HTTP requests to APIs from Python. It shows examples of making GET, POST, PUT, and DELETE requests to a sample TODO list API to perform operations like retrieving, adding, updating, and deleting tasks. It then builds out these requests into reusable functions within a todo.py library module to easily interact with the API. Finally, it demonstrates using the library functions to add a task and retrieve all tasks from the API.
![import requests
resp = requests.get('http://apiathens.herokuapp.com/tasks/')
if resp.status_code != 200:
# This means something went wrong.
raise ApiError('GET /tasks/ {}'.format(resp.status_code))
for todo_item in resp.json():
print('{} {}'.format(todo_item['id'],todo_item[title]))
GET /Tasks](https://image.slidesharecdn.com/pru7ypnbr0ub3ifuqumy-signature-3022ced01d98cd890637460a6d2ab4c5af5cba6b9ebf9c6798f14c83b7cd3415-poli-150427170136-conversion-gate02/85/Consuming-APIs-with-Python-14-320.jpg)
![task = {"summary": "Take out trash", "description": "" }
resp =
requests.post('http://apiathens.herokuapp.com/tasks/',
json=task)
if resp.status_code != 201:
raise ApiError('POST /tasks/
{}'.format(resp.status_code))
print('Created task. ID: {}'.format(resp.json()["id"]))
POST /Task](https://image.slidesharecdn.com/pru7ypnbr0ub3ifuqumy-signature-3022ced01d98cd890637460a6d2ab4c5af5cba6b9ebf9c6798f14c83b7cd3415-poli-150427170136-conversion-gate02/85/Consuming-APIs-with-Python-15-320.jpg)
![import todo
resp = todo.add_task("Take out trash")
if resp.status_code != 201:
raise ApiError('Cannot create task: {}'.format(resp.status_code))
print('Created task. ID: {}'.format(resp.json()["id"]))
resp = todo.get_tasks()
if resp.status_code != 200:
raise ApiError('Cannot fetch all tasks: {}'.format(resp.status_code))
for todo_item in resp.json():
print('{} {}'.format(todo_item['id'], todo_item[title]))
Let’s Use our Library!](https://image.slidesharecdn.com/pru7ypnbr0ub3ifuqumy-signature-3022ced01d98cd890637460a6d2ab4c5af5cba6b9ebf9c6798f14c83b7cd3415-poli-150427170136-conversion-gate02/85/Consuming-APIs-with-Python-19-320.jpg)
Consuming APIs with Python
- 1. Consuming your APIs with Python Michael Petychakis, Ntua
- 2. { "@context":{ "schema":"http://schema.org/", "as":"http://www.w3.org/ns/activitystreams#" }, "@type": "schema:Person", "schema:name": "Michael Petychakis", "schema:jobTitle": "Phd Student", "schema:telephone": "(425) 123-4567", "schema:email":"mpetyx@epu.ntua.gr", "schema:url": "http://www.mpetyx.com", "schema:nationality":"Greece", "as:actor":{ "@id":"http://www.twitter.com/mpetyx" } } About @mpetyx
- 7. 958 million websites 60 trillion web pages
- 8. Each API is Unique
- 10. +optional parameters (useful for pagination, partial responses, filtering,…) /snakes?limit=3&offset=14 /snakes?fields=name,age,description Design accepted representations to be sent by potential clients
- 11. consider existing standardized data formats e.g., HTML, Atom, JSON,… Accept: application/json (Digg) /venue.json (Foursquare) http://api.sciencemuseum.org.uk/exhibitions/ ?output=json Choose the representation(s) of the response to be sent to client
- 13. # Step one for every Python app that talks over the web. $ pip install requests Requirements
- 14. import requests resp = requests.get('http://apiathens.herokuapp.com/tasks/') if resp.status_code != 200: # This means something went wrong. raise ApiError('GET /tasks/ {}'.format(resp.status_code)) for todo_item in resp.json(): print('{} {}'.format(todo_item['id'],todo_item[title])) GET /Tasks
- 15. task = {"summary": "Take out trash", "description": "" } resp = requests.post('http://apiathens.herokuapp.com/tasks/', json=task) if resp.status_code != 201: raise ApiError('POST /tasks/ {}'.format(resp.status_code)) print('Created task. ID: {}'.format(resp.json()["id"])) POST /Task
- 16. # todo.py def get_tasks(): pass def describe_task(task_id): pass def add_task(summary, description=""): pass def task_done(task_id): pass def update_task(task_id, summary, description): pass #todo.py
- 17. def _url(path): return 'http://apiathens.herokuapp.com' + path Helper Function
- 18. import requests def get_tasks(): return requests.get(_url('/tasks/')) def describe_task(task_id): return requests.get(_url('/tasks/{:d}/'.format(task_id))) def add_task(summary, description=""): return requests.post(_url('/tasks/'), json={ 'summary': summary, 'description': description, }) def task_done(task_id): return requests.delete(_url('/tasks/{:d}/'.format(task_id))) def update_task(task_id, summary, description): url = _url('/tasks/{:d}/'.format(task_id)) return requests.put(url, json={ 'summary': summary, 'description': description,}) Integrated Code
- 19. import todo resp = todo.add_task("Take out trash") if resp.status_code != 201: raise ApiError('Cannot create task: {}'.format(resp.status_code)) print('Created task. ID: {}'.format(resp.json()["id"])) resp = todo.get_tasks() if resp.status_code != 200: raise ApiError('Cannot fetch all tasks: {}'.format(resp.status_code)) for todo_item in resp.json(): print('{} {}'.format(todo_item['id'], todo_item[title])) Let’s Use our Library!
- 20. Thank you Michael Petychakis @mpetyx <a href="mailto:mpetyx@epu.ntua.gr?Subject= Hello" target="_top">Drop me an e-mail</a>