readme-callback-py.md

callback

A simple small module for Pythonista (http://omz-software.com/pythonista) which aims to make an easy callback URL handling & argument passing system.

Usage

1. import callback
2. Create a handler instance handler = callback.InfoHandler(sys.argv)
3. Decorate command handlers with @handler.cmd(cmdname)
4. Call handler.handle() (and if it returns False, there was nothing to handle!)
5. To create a callback url for a certain command use callback.url(cmdname[,script][,a=1,b=2,....) (defaults to current script) and assign this to any x-callback-url compliant URL argument!

How It Works

The resulting callback URL will have the first argv set to 'callback', and the second argv to a JSON object string. The JSON object has a key called 'cmd' for the command name, and other keys are named arguments for the command.

The InfoHandler class checks if argv[0] is 'callback' and if so, parses the JSON object in argv[1]. Each function decorated with @handler.cmd() gets added to a command map and when handler.handle() is called, the appropriate function is called with the arguments from the JSON object (except for 'cmd').

Examples

###tweetprocessor.py

This uses the TweetBot module (https://gist.github.com/djbouche/5079739). tweetprocessor.py when normally run, will read tweets.txt and process each line as a separate tweet via TweetBot.

import TweetBot
import callback
import sys

handler = callback.InfoHandler(sys.argv)

@handler.cmd("process_next")
def process_next(tweets):
    if len(tweets):
        t = tweets.pop() #process next tweet on the list
        #tweet, and send the list of remaining tweets down
        TweetBot.tweet(t,callback.url('process_next',tweets=tweets))
    else: #no more tweets
        print 'Finished processing tweets!'

if name == "__main__" and not handler.handle():
    #first entry point
    f = open("tweets.txt","r") #read tweets from a file called 'tweets.txt'
    a = f.read()
    f.close()
    tweets = a.splitlines() #one line per tweet
    tweets.reverse() #so we can pop them off
    process_next(tweets) #start processing!

###listtweet.py

Of course, you can use the handler command from another script as normal.

import tweetprocessor

tweetprocessor.process_next(['tweet 1','tweet 2','tweet 3'].reverse)

###facetweet.py

You could also use it in a callback in another script. facetweet.py will have the posted text to tweetprocessor.py process_next() after the post to Facebook is completed. This uses the Drafts module (https://gist.github.com/djbouche/5079537).

import Drafts
import callback

sometext = "Hello, world!"

# Use 'tweetprocessor' as the callback script
Drafts.do_action('Post to Facebook',sometext,
    callback.url('process_next','tweetprocessor',tweets=[sometext]))