Simple JSON parsing and Requests

Tue 06 February 2018 by Patrick Pierson

The Requests package has an output format for JSON that makes things really easy to parse in Python.

Take for instance you were to pull data from Reddit's JSON api.

import requests
from pprint import pprint
headers = {'User-Agent': 'Pytesting'}
r = requests.get('https://reddit.com/r/python.json', headers=headers)
pprint(r.json())

You get back a ton of data:

{u'data': {u'after': u't3_7vpydk',
           u'before': None,
           u'children': [{u'data': {u'approved_at_utc': None,
                                    u'approved_by': None,
                                    u'archived': False,
                                    u'author': u'aphoenix',
                                    u'author_flair_css_class': u'',
                                    u'author_flair_text': u'reticulated',
                                    u'banned_at_utc': None,
                                    u'banned_by': None,
                                    u'brand_safe': True,
                                    u'can_gild': False,
                                    u'can_mod_post': False,

We can see that the 'data' block is a dictionary. If we dial down we can start to do things like printing the names of the posters.

import requests
from pprint import pprint
headers = {'User-Agent': 'Pytesting'}
r = requests.get('https://reddit.com/r/python.json', headers=headers)
for i in r.json()['data']['children']:
  pprint(i['data']['author'])

The result is as follows:

u'aphoenix'
u'AutoModerator'
u'dpack78'
u'lemurata'
u'dbader'
u'javi404'
u'dashee87'
u'gerryjenkinslb'
u'dan_kilomon'
u'dadmay96'
u'Toardo'
u'Demopathos'
u'mohi7solanki'
u'winner_godson'
u'Taconite_12'
u'Nathik27'
u'Delengowski'
u'masterjohn17'
u'AlSweigart'
u'junp1289'
u'marcrleonard'
u'AegisToast'
u'Blath716'
u'SerpentAI'
u'amirathi'
u'PostSeptimus'
u'Serialk'

This was just a short sample of how easy it is to parse JSON with python.