Sending SMS using 3rd party API by integrating with custom-built BOT in Python

Hi Guys!

Today, We’re going to discuss the way to send SMS through popular 3rd-party API (Twilio) using Python 3.7.

Before that, you need to register with Twilio. By default, they will give you some credit in order to explore their API.

And, then you can get a virtual number from them, which will be used to exchange SMS between your trusted numbers for trial Account.

The basic architecture can be depicted are as follows –

How to get a verified number for your trial account?

Here is the way, you have to do that –

You can create your own trial account by using this link.

Apart from that, you need to download & install Ngrok. This is available for multi-platform. For our case, we’re using Windows.

The purpose is to run your local web service through a global API like interface. I’ll explain that later.

You need to register & install that on your computer –

Once, you download & install you need to use the global link of any running local server application like this –

This is the dummy link. I’ll hide the original link. However, every time when you restart the application, you’ll get a new link. So, you will be safe anyway. 🙂

Once, you get the link, you have to update that global link under the messaging section. Remember that, you have to keep the “/sms” part after that.

Let’s see our sample code. here, I would be integrating my custom developed BOT developed in Python. However, I’ll be only calling that library. We’re not going post any script or explain that over here.

1. serverSms.py ( This script is a server script, which is using flask framework & it will respond to the user’s text message by my custom developed BOT using Python)

# /usr/bin/env python
##############################################
#### Written By: SATYAKI DE               ####
#### Written On: 03-Nov-2019              ####
####                                      ####
#### Objective: This script will respond  ####
#### by BOT created by me. And, reply to  ####
#### sender about their queries.          ####
#### We're using Twillio API for this.    ####
####                                      ####
##############################################

from flask import Flask, request, redirect
from twilio import twiml
from twilio.twiml.messaging_response import Message, MessagingResponse
import logging
from flask import request
from SDChatbots.clsTalk2Bot import clsTalk2Bot

app = Flask(__name__)

@app.route("/sms", methods=['GET', 'POST'])
def sms_ahoy_reply():
    """Respond to incoming messages with a friendly SMS."""
    # Start our response
    # resp = twiml.Response()
    message_body = request.form['Body']

    print(message_body)
    logging.info(message_body)

    y = clsTalk2Bot()
    ret_val = y.TalkNow(message_body)
    zMsg = str(ret_val)
    print('Response: ', str(zMsg))

    resp = MessagingResponse()

    # Add a message
    resp.message(zMsg)

    return str(resp)

if __name__ == "__main__":
    app.run(debug=True)

Key lines from the above scripts are –

@app.route("/sms", methods=['GET', 'POST'])

The route is a way to let your application understand to trigger the appropriate functionalities inside your API.

message_body = request.form['Body']

Here, the application is capturing the incoming SMS & print that in your server log. We’ll see that when we run our application.

y = clsTalk2Bot()
ret_val = y.TalkNow(message_body)
zMsg = str(ret_val)

Now, the application is calling my developed python BOT & retrieve the response & convert it as a string before pushing the response SMS to the user, who originally send the SMS.

resp = MessagingResponse() --This is for Python 3.7 +

# Add a message
resp.message(zMsg)

return str(resp)

Finally, you are preparing the return SMS & send it back to the user.

For the old version, the following line might work –

resp = twiml.Response()

But, just check with the Twilio API.

Let’s run our server application. You will see the following screen –

Let’s see, if one someone ask some question. How the application will respond –

And, let’s explore how our server application is receiving it & the response from the server –

Note that, we’ll be only sending the text to SMS, not the statistics sent by my BOT marked in RED.  😀

Let’s check the response from the BOT –

Yes! We did it. 😀

But, make sure you are regularly checking your billing as this will cost you money. Always, check the current balance –

You can check the usage from the following tab –

You can create a billing alarm to monitor your usage –

Let me know, how do you like it.

So, we’ll come out with another exciting post in the coming days!

N.B.: This is demonstrated for RnD/study purposes. All the data posted here are representational data & available over the internet.