muskanmahajan37 / cat-query-bot

The code that powers the @CatQueryBot Twitter bot

Cat Query Bot

This is a Twitter bot (@CatQueryBot) written in Python that receives words, queries the Google Custom Search Engine API, and then tweets the top 4 photos to an account if any are found, alone with the search query and the total amount of results.

Setup

So, you want to clone it and create your own? Sure thing!

• Clone this repo
• Make a copy of the local_settings_example.py file and name it local_settings.py - instructions for each are listed below

Getting Google API keys & Custom Search Engine ID

First things first, visit Google's Custom Search Engine console to get started. Click "Add" and follow the instructions. Then, when you're done, you can set it to search the entire web with the following excerpt from this Google Support topic:

To set an existing search engine to search the entire web:

• From the control panel, select the search engine you want to edit.
• Click Setup from the menu on the left and then click the Basics tab.
• Go to the Search the Entire Web section and toggle this feature On or Off.
• Selecting On will augment your results with general Web Search results. Selecting Off will limit your results only to the sites you specified.

You will also need to enable image search. Optional: adjust SafeSearch settings.

Twitter Developer API

Head on over to the Twitter Developer Platform website and fill out an application. Feel free to contact me if you're having issues getting your application accepted - I had to message back and forth for a little while before I got in with this particular bot.

Once you're in, create a new API project and generate some keys. Consult Google if you need help with this step.

Configuring

The bot is written to read in from a local_settings.py file that stores API keys, links to feeds, and debug variables.

SAFE_SEARCH = True

Set safe search to True/False depending on if you're looking to keep NSFW images out of your feed.

GOOGLE_API_KEY = 'Your Google API Key'
GOOGLE_CUSTOM_SEARCH_ID = 'Your Google Custom Search Engine ID'

Plug in your Google API keys from earlier!

MY_CONSUMER_KEY = '57axlM8Ti5ntN8aOiGXu7lx48'
MY_CONSUMER_SECRET = 'mgK4pyeCwUmI6So3XxQIEmKqE70w7wKvRxKYrAJthPr7J1396O'
MY_ACCESS_TOKEN_KEY = '3171134888-3fLCSwWWP8j3PAQkflR0OfjVklLr5QN1VadOLVL'
MY_ACCESS_TOKEN_SECRET = 'lvROpdTFXViRJHE02es3vgabaMxAqJ19gAD3SuyPh8hrM'

Plug in your Twitter API keys from earlier!

WORDS_URL = ""

This URL is something you'll need to set up on your own. I simply created a PHP script on my website that allows me to send a GET request, and it sends back a JSON response that looks like one of the following:

{ success: true, word: 'sniper' }
{ success: false, word: null }

When you design this script, it must return a JSON string matching that format.

TEST_SCRAPE_URL = False

Since the Google Custom Search Engine has a rate limit, you probably don't want to query it hundreds of times while testing. Because of this, you can set TEST_SCRAPE_URL to a static page containing the results of a search (which you can copy/paste from the console after running!)

Debug = True

This ensures that the bot does not tweet while testing to make sure everything works. Set this to False when you're ready to deploy your bot!

Credit

Thanks to both Google and Twitter for creating and maintaining great and simple to use APIs. Without them, this completely pointless project would have never seen the light of day.

I first got started with fiddling with Twitter bots thanks to Tom Meagher's heroku_ebooks. Check that repository out, and feel free to message me if you want some tips on how to implement some of the modifications I've made to it for @bret_ebooks.