image2ocr2/sct_to_text.py

'''Created Feb 28, 2023 deeejas'''

import re
import os
import platform
import subprocess as sp
from collections import deque

import cv2
import keyboard as kb
import numpy as np
import pywinctl as gw
import pytesseract
import mss
from plyer import notification

container = deque(maxlen=500)
if platform.system() == 'Windows':
    pytesseract.pytesseract.tesseract_cmd = '{}/tesseract.exe'.format(os.getenv('TESSERACT'))

# Noise Removal
def noise_removal(image):
    kernel = np.ones((1, 1), np.uint8)
    image = cv2.dilate(image, kernel, iterations=1)
    kernel = np.ones((1, 1), np.uint8)
    image = cv2.erode(image, kernel, iterations=1)
    image = cv2.morphologyEx(image, cv2.MORPH_CLOSE, kernel)
    image = cv2.medianBlur(image, 3)
    return (image)

# Dilation and Erosion
def thin_font(image):
    image = cv2.bitwise_not(image)
    kernel = np.ones((2, 2), np.uint8)
    image = cv2.erode(image, kernel, iterations=1)
    image = cv2.bitwise_not(image)
    return (image)

def thick_font(image):
    image = cv2.bitwise_not(image)
    kernel = np.ones((2, 2), np.uint8)
    image = cv2.dilate(image, kernel, iterations=1)
    image = cv2.bitwise_not(image)
    return (image)

def copy_rec():
    # wh = gw.getScreenSize()._asdict()
    # mon = {'top': 0, 'left': 0, **wh}
    title = 'ANAF depunere declaratii'
    timeout = 3  # seconds
    app_name = 'e-guvernare'
    app_icon = 'not_icon.ico'

    try:
        with mss.mss() as sct:
            monitor = sct.monitors[1]
            left = monitor['left'] + monitor['width'] * 35 // 100  # 50% from the left
            top = monitor['top'] + monitor['height'] * 25 // 100  # 50% from the top
            right = left + 800  # 400px width
            lower = top + 100  # 400px height
            bbox = (left, top, right, lower)
            chrome_win = gw.getWindowsWithTitle('decl.anaf.mfinante.gov.ro/WAS6DUS/displayFile.do - Brave')[0]
            active_win = gw.getActiveWindow()
            if chrome_win.title in active_win.title:
                sct_img = sct.grab(bbox)
                # mss.tools.to_png(sct_img.rgb, sct_img.size, output='screenshot.png')
                im = np.asarray(sct_img)  # -> good approch with cv2
                # im = Image.open('recipisa.png').convert('RGB')
                # im = np.array(im)
                # img = Image.frombytes("RGB", sct_img.size, sct_img.bgra, "raw", "BGRX")
                # gray = cv2.cvtColor(im, cv2.COLOR_BGR2GRAY)
                im_rgb = cv2.cvtColor(im, cv2.COLOR_BGR2RGB)
                _thresh, im_bw = cv2.threshold(im_rgb, 210, 230, cv2.THRESH_BINARY)

                # df = pytesseract.image_to_data(gray, output_type=pytesseract.Output.DATAFRAME)
                text = pytesseract.image_to_string(im_bw)
                receipt_index = re.search(r'(?<=Indexul este )[0-9]+', text)
                if receipt_index:
                    receipt_index = receipt_index.group()
                    # filtered_df = df.loc[(df['word_num'] == 14) & (df['block_num'] == 11)]
                    # receipt_index = filtered_df.iloc[0]['text']
                    if receipt_index not in container:
                        item1 = sp.run(['copyq', 'add', receipt_index], capture_output=True, text=True)
                        print(item1.stdout)
                        container.appendleft(receipt_index)
                        titleb = f'{receipt_index}'
                        print(len(container), f'index <{receipt_index}> appended')
                        notification.notify(title=titleb,
                                            message='index',
                                            app_name=app_name,
                                            app_icon=app_icon,
                                            timeout=timeout)
                    else:
                        notification.notify(title=title,
                                            message=f'index <{receipt_index}> already in container',
                                            app_name=app_name,
                                            app_icon=app_icon,
                                            timeout=timeout)
                else:
                    notification.notify(title=title,
                                        message='index not found in re.search pattern',
                                        app_name=app_name,
                                        app_icon=app_icon,
                                        timeout=timeout)
            else:
                notification.notify(title=title,
                                    message='chrome window not found',
                                    app_name=app_name,
                                    app_icon=app_icon,
                                    timeout=timeout)

    except Exception as exc_:
        raise exc_(str(exc_))
        print('wndow not active')


def main():
    kb.add_hotkey('`+1', copy_rec, args=())
    kb.wait()


if __name__ == '__main__':
    main()
    # print(gw.getAllTitles())