autotyper.py

#!/usr/bin/python3
#
#
###############################################################
#	
#	Auto typing bot for typeracer.com made by CABREX
#	Takes screenshot of the region with the text to type,
#	converts it into text using Computer Vision and tesseract 
# 	engine. 
#	
#	Make sure that the starting border of the leaderboard table 
#	under the typing box is 1cm / 39 pixels from the bottom of 
#	the screen.
#	
###############################################################
#
#



from pynput.keyboard import Controller as KeyboardController
from pynput.mouse import Listener as MouseListener
from pynput.mouse import Controller as MouseController
from pynput import keyboard as Keyboard
import time
import pyscreenshot as ImageGrab
import sys
import os

import argparse

import cv2
import pytesseract

keyboard = KeyboardController()
mouse = MouseController()

delay = 0.01
iterations_count = 0
char_count = 0
debug_mode = False

# Weird variables
a = 0

# Default coords for laptops (1920x1080)
x1 = 450
# y1 = 716
y1 = 616
x2 = 1449
# y2 = 889
y2 = 789

# Command line argument options
description = "A bot that automates inputs for typeracer.com. Uses the OCR Engine to convert text in the image to text."


# Sends keyboard inputs
def typer(string):
	global iterations_count, char_count

	for c in string:
		char_count = char_count + len(c)
		if (not "change display format" in c.lower()):
			for ch in c:

				if ch == '|':
					keyboard.press("I")
					keyboard.release("I")

				elif ch == "[":
					pass

				elif ch == "\n":
					keyboard.press(" ")
					keyboard.release(" ")

				elif ch != "|":
					keyboard.type(ch)
				
				try:
					time.sleep(delay)
				except KeyboardInterrupt:
					print("keyboard interrupt detected")
					stats()
					exit(0)



# Gets the cropped screenshot of the place
def screenshawn():


	# part of the screen
	im=ImageGrab.grab(bbox=(x1, y1, x2, y2))

	# Saving cropped image to file
	if not os.path.isdir('assets/'):
		os.mkdir('assets/')
	im.save('assets/sample.png')


# Converting sample.png to text by using computer vision and tesseract OCR engine
def refined_message_to_string():
	pytesseract.pytesseract.tesseract_cmd = r"C:\Program Files\Tesseract-OCR\tesseract.exe"


	img = cv2.imread('assets/sample.png')
	 
	gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
	 

	ret, thresh1 = cv2.threshold(gray, 0, 255, cv2.THRESH_OTSU | cv2.THRESH_BINARY_INV)

	rect_kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (18, 18))

	dilation = cv2.dilate(thresh1, rect_kernel, iterations = 1)

	contours, hierarchy = cv2.findContours(dilation, cv2.RETR_EXTERNAL,
	                                                 cv2.CHAIN_APPROX_NONE)

	im2 = img.copy()

	for cnt in contours:
		x, y, w, h = cv2.boundingRect(cnt)
	     
	    # Drawing a rectangle on copied image
		rect = cv2.rectangle(im2, (x, y), (x + w, y + h), (0, 255, 0), 2)
	     
	    # Cropping the text block for giving input to OCR
		cropped = im2[y:y + h, x:x + w]
	     
		# Open a clean file
		if debug_mode:
			if not os.path.isdir('assets/'):
				os.mkdir('assets')

			with open("assets/recognized.txt", "w") as file:
				file.write("")
	     
		# Apply OCR on the cropped image
		try:
			text = pytesseract.image_to_string(cropped)
		except pytesseract.pytesseract.TesseractNotFoundError:
			print("[ERROR]- Pytesseract not found. Please install")
			exit(0)

	     
		# Appending the text into file
		
		if debug_mode:
			with open("assets/recognized.txt", "a") as file_handle:
				file_handle.write(text)

		yield text


def on_press(key):

	if key == Keyboard.Key.ctrl_r:
		return False



def fetch_coords(key):
	global a, x1, x2, y1, y1

	if key == Keyboard.Key.ctrl_r:
		coords = mouse.position
		print(f"x: {coords[0]}; y: {coords[1]}")
		print(coords)
		a += 1
		if a == 1:
			x1 = coords[0]
			y1 = coords[1]
		if a == 2:
			x2 = coords[0]
			y2 = coords[1]
			return False

def callibrate():

	with Keyboard.Listener(on_click=fetch_coords) as listener:
		try:
			listener.join()
		except KeyboardInterrupt:
			listener.stop()
			exit(0)


def stats():
	print("Stats for the run:")
	print(f"1) Number of iterations: {iterations_count}")
	print(f"2) Total number of characters typed: {char_count}")




if __name__ == "__main__":

	print("Auto typer bot by CABREX (https://github.com/0xcabrex)")

	# Initialize parser
	parser = argparse.ArgumentParser(description = description)
	parser.add_argument('-d', '--delay', help = "Changes the delay per character (default = 0.01)")
	parser.add_argument('-s', '--setdebug',action="store_true", help = "Toggles debug mode, outputs the cropped image and converted code to file (default False)")
	parser.add_argument('-c', '--callibrate', help = "Callibrates screenshot coordinates")
	args = parser.parse_args()

	if args.delay:
		try:
			delay = float(args.delay)
		except ValueError:
			print(f"Delay value '{args.delay}' is invalid, please enter a number")
			exit(0)

	if args.setdebug:
		debug_mode = args.setdebug

	if args.callibrate:
		print("callibrating")
		with Keyboard.Listener(on_press=fetch_coords) as listener:
			try:
				listener.join()
			except KeyboardInterrupt:
				listener.stop()
				exit(0)

	print(f"Debug mode is set to {debug_mode}")
	print(f"Delay has been set to {delay}")
	print("Bot is running...\n")
	while True:
		print("to exit, ctrl+c and then right control")

		with Keyboard.Listener(on_press=on_press) as listener:
		    try:
		        listener.join()
		        print(f'right control was pressed, image has been captured')
		    except KeyboardInterrupt:
		        print(f"Keyboard interrupt detected")
		        print("\n\n")
		        stats()
		        exit(0)


		screenshawn()


		string = refined_message_to_string()



		with Keyboard.Listener(on_press=on_press) as listener:
			try:
				listener.join()
				print(f'right click was pressed. Auto-typing...')
			except KeyboardInterrupt:
				print(f"Keyboard interrupt detected")
				stats()
				exit(0)

		typer(string)
		print("Done\n")
		if not debug_mode:
			os.remove('assets/sample.png')
		iterations_count += 1