Scrape Google images

google-images-downloader.py

# ■こちらのYouTube動画で作成したコードです
# https://www.youtube.com/watch?v=qBQ7MNPDq7I
#
# ■下記のコードを写経しながら、スクレイピングについて学習した際のコードとなります
# https://gist.github.com/genekogan/ebd77196e4bf0705db51f86431099e57#gistcomment-3465381
#
# コードの実行には下記のインストールが必要です
# Selenium, chromedriver-binary

from selenium import webdriver
import chromedriver_binary
import requests
import argparse
import os
import time
import shutil


def save_img(inp, img, i, directory):
    try:
        filename = inp + str(i) + ".jpg"
        response = requests.get(img, stream=True)
        image_path = os.path.join(directory, filename)
        print(image_path)
        with open(image_path, "wb") as file:
            shutil.copyfileobj(response.raw, file)
    except Exception:
        print("ERROR: save_img")
        pass


def find_urls(inp, url, driver, directory):
    driver.get(url)
    for _ in range(5):
        driver.execute_script("window.scrollBy(0, 10000)")
        try:
            driver.find_element_by_css_selector('.mye4qd').click()
        except:
            continue
    for j, imgurl, in enumerate(driver.find_elements_by_xpath('//img[contains(@class, "rg_i Q4LuWd")]')):
        try:
            imgurl.click()
            img = driver.find_element_by_xpath('//body/div[2]/c-wiz/div[3]/div[2]/div[3]/div/div/div[3]/div[2]/c-wiz/div[1]/div[1]/div/div[2]/a/img').get_attribute("src")
            save_img(inp, img, j, directory)
            time.sleep(3)
        except:
            print("ERROR: find_urls")
            pass


if __name__ == "__main__":
    parser = argparse.ArgumentParser(description="Scrape Google images")
    parser.add_argument("-s", "--search", default="bananas", type=str, help="search term")
    parser.add_argument("-d", "--directory", default="./downloads", type=str, help="Save directory")
    args = parser.parse_args()
    driver = webdriver.Chrome()
    directory = args.directory
    inp = args.search
    if not os.path.isdir(directory):
        os.makedirs(directory)
    url = "https://www.google.com/search?q=" + str(inp) + "&source=lnms&tbm=isch&sa=X&ved=2ahUKEwie44_AnqLpAhUhBWMBHUFGD90Q_AUoAXoECBUQAw&biw=1920&bih=947"
    find_urls(inp, url, driver, directory)