Python wrapper for official Hacker News API

Since Hacker News launched their official API, I started working on Python wrapper for it. API v0 is fairly simple, you send a GET request with hacker news story's ID or user id, it returns data in JSON.

Here I do the same using requests library and return specific object like Item for news story or comment and User for a user on Hacker News.

All source can be found here.

```
#!/usr/bin/env python

__all__ = [
'User',
'Item',
'HackerNews',
'InvalidAPIVersion',
'InvalidItemID',
'InvalidUserID']

"""
haxor
Unofficial Python wrapper for official Hacker News API
"""

import datetime
import json

import requests

from settings import supported_api_versions
# supported_api_versions = {'v0': 'https://hacker-news.firebaseio.com/v0/'}

class InvalidItemID(Exception):
pass

class InvalidUserID(Exception):
pass

class InvalidAPIVersion(Exception):
pass

class HackerNews(object):

def __init__(self, version='v0'):
try:
self.base_url = supported_api_versions[version]
except KeyError:
raise InvalidAPIVersion

def _get(self, url):
response = requests.get(url)
if response.status_code == requests.codes.ok:
return response
else:
raise Exception('HTTP Error: {}'.format(response.status_code))

def get_item(self, item_id):
response = self._get('{0}item/{1}.json'.format(self.base_url, item_id))

if not response.json():
raise InvalidItemID

return Item(response.json())

def get_user(self, user_id):
response = self._get('{0}user/{1}.json'.format(self.base_url, user_id))

if not response.json():
raise InvalidUserID

return User(response.json())

def top_stories(self, limit=None):
response = self._get('{}topstories.json'.format(self.base_url))
return response.json()[:limit]

def get_max_item(self):
response = self._get('{}maxitem.json'.forma

response.json() appears repeatedly here and in other methods:

def get_item(self, item_id):
    response = self._get('{0}item/{1}.json'.format(self.base_url, item_id))

    if not response.json():
        raise InvalidItemID

    return Item(response.json())

You could simplify by getting .json() in the first place:

def get_item(self, item_id):
    data = self._get('{0}item/{1}.json'.format(self.base_url, item_id)).json()

    if not data:
        raise InvalidItemID

    return Item(data)

This could be more efficient, because you can't really know for sure if the .json() is implemented lazily (evaluating things only once), or every time you call it. In fact, the speed difference is noticeable after this change, when you iterate over a list of items.

self.base_url appears in many methods, for example:

def get_item(self, item_id):
    response = self._get('{0}item/{1}.json'.format(self.base_url, item_id))
    # ...

def get_user(self, user_id):
    response = self._get('{0}user/{1}.json'.format(self.base_url, user_id))
    # ...

def top_stories(self, limit=None):
    response = self._get('{}topstories.json'.format(self.base_url))
    return response.json()[:limit]

def get_max_item(self):
    response = self._get('{}maxitem.json'.format(self.base_url))
    return response.json()

To reduce the duplication, you could introduce some more helper methods like ._get, for example:

def _get_page(self, page):
    return self._get('{0}{1}.json'.format(self.base_url, page))

def _get_page_param(self, page, param):
    return self._get('{0}{1}/{2}.json'.format(self.base_url, page, param))

Using these new methods, you could simplify the above methods:

def get_item(self, item_id):
    response = self._get_page_param('item', item_id)
    # ...

def get_user(self, user_id):
    response = self._get_page_param('user', user_id)
    # ...

def top_stories(self, limit=None):
    return self._get_page('topstories').json()[:limit]

def get_max_item(self):
    return self._get_page('maxitem').json()

StackExchange Code Review Q#65214, answer score: 2