WhatsApp-Selenium

What is this

This is fully automated version of Whatsapp on Selenium. Cause Meta wants you to pay for their own.

This is a selenium based python framework, that will allow you extract or send messages from whatsapp, which will be saved in a .CSV file. You can later run data analysis on it or train a model.

The App uses Web.Whatsapp.com to extract messages in an automated fashion. You will be provided with a QR-Code on first run to login. After that you can start extracting in Headless mode. Browser Data/Session is saved in UserData folder in same directory as your script.

How to use ?

Setup

1. First clone the prjoect then install requirements from the requirements.txt

pip install -r requirements.txt

2. Install Chrome web driver from here https://chromedriver.chromium.org/downloads and Chrome browser from here https://www.google.com/chrome. Make sure to add your web driver to your system path, although it is not necessary.

3. You can import the WhatsExtract.py into your main python file.

from Whatsapp import Whatsapp

Usage

Now you can use the object of the class to extract message. Example 1:

bot = Whatsapp(silent=True, headless=False)
bot.login()
bot.getMessages("Alex", scroll=10)

Options

Default Constructor

For the default constructor you have three options

1. executable_path You can specifiy the path from your Chrome driver if not already in your system PATH. The Format should be like C:/Driver/chromedriver.exe.
2. silent has two options True or False. If set to True, it will not display selenium logs.
3. headless has two options True or False. If set to True, the browser will be headless (not visible).

Function getMessages

Has 4 arguments

1. chatName - This will be the chatname you want to extract messages from. Could be part of the chatName. It utilizes WhatsApp chat search to find the chat.
2. all has two options True or False. If set to True, will true to extract all available chats. Although it may stop before that, see OPTION 4.
3. scroll You can set to any integer for scolling some of the chat, to extract more messages.
4. manualSync has two options True or False. If set to True, script will wait for your input before it start extracting messages. Its so can manually scroll to where you want the messages extraction to begin from.

Function replyTo

Has 2 arguments

1. element - This is message element, you want to reply to. It is a selenium object
2. msg Its the message you want to send.

Function hookIncomming

Has 2 arguments. Return last incomming message element and parsed message.

1. chatName - This is chat name you want to listen to.
2. func This is a async function you set to do things with the message returned.

Example:

responses = {"hello": "Hello, Nice to meet you !", "How are you ?": "I'm fine, thank you !",
                "Who are you ?": "My name is RevBot, I'm a chatbot made by AbdulRahman Nadeem."}

async def func(element, msg):
    print(msg)
    if (msg[2] == "EXIT!"):
        bot.browser.close()
        exit()

    if (msg[2].lower() == "help"):
        bot.replyTo(
            element, str("My commands are : " + str(responses)))
    else:
        try:
            bot.replyTo(element, responses[msg[2].lower()])
        except:
            bot.replyTo(element, "I don't understand !")

bot.hookIncomming("AbdulRahman", func)

BUGS

• Sometimes the sender name will be mangled.
• Unable to diffrentiate between VoiceNote in a replied message.
• Too many workarounds done in script, which might fail.

Features to be added

• Add support for Emojis
• Add support to diffrentiate between different media.
• Better Sync controll, to download all messages.