import requests
import xlsxwriter
import os
from bs4 import BeautifulSoup

headers = {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36'}

workbook = xlsxwriter.Workbook('parsed_data.xlsx')
worksheet = workbook.add_worksheet()

worksheet.write(0, 0, 'Название товара')
worksheet.write(0, 1, 'Название изображения')
worksheet.write(0, 2, 'Ссылка на изображение')
worksheet.write(0, 3, 'Item code')

row = 1

if not os.path.exists('image'):
    os.makedirs('image')

for page_num in range(1, 22):  
    url = f'https://www.scmgroupspareparts.com/en/index.php/m6.html?mode=grid&p={str(page_num).zfill(2)}'
    response = requests.get(url, headers=headers)
    soup = BeautifulSoup(response.content, 'html.parser')
    
    items = soup.find_all(class_='col-lg-3 col-md-3 col-sm-6 col-xs-12 item')

    for item in items:
        try:
            product_name = item.find(class_='product-name').a.text.strip()
            
            code_element = item.find('p', string='Item code:')
            if code_element:
                item_code = code_element.find_next('b').text.strip()
            else:
                item_code = 'Item code not found'
            
            image_url = item.find(class_='product-image').img['src']
            
            # Extracting image name from the URL
            image_name = image_url.split('/')[-1]
            image_name = image_name.replace('.jpg', f'_{row}.jpg')  # Append row number to make the image name unique

            print(f'Обрабатывается товар: {product_name}')
            print(f'Ссылка на изображение: {image_url}')

            worksheet.write(row, 0, product_name)
            worksheet.write(row, 1, image_name)
            worksheet.write(row, 2, image_url)
            worksheet.write(row, 3, item_code)

            image_response = requests.get(image_url)

            with open(f'image/{image_name}', 'wb') as image_file:
                image_file.write(image_response.content)

            print(f'Изображение сохранено как: {image_name}')
        except Exception as e:
            print('Ошибка при парсинге данных:', e)

        row += 1

workbook.close()