셀럽미: 닮은 아이돌 찾기 2024년 맞이 업데이트 (1/3) – 이미지 수집

from selenium import webdriver
from selenium.webdriver.common.keys import Keys
from selenium.webdriver.common.by import By
import time
import urllib3
import os
import io

from PIL import Image
import base64

os.environ["PATH"] = os.environ.get("PATH") + ":."

# 평판 https://brikorea.com/rk/ip2312


driver = webdriver.Chrome()
driver.set_window_position(0, 0)
driver.set_window_size(1400, 1000)
driver.get("https://www.google.co.kr/imghp?hl=ko&authuser=0&ogbl")


header = {
    "User-Agent": "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/43.0.2357.134 Safari/537.36"
}


http = urllib3.PoolManager()


celeb_list_men = [
    "BTS 정국",
    "비스트 용준형",
    "슈퍼주니어 신동",
    "인피니트 엘",
]


celeb_list = [
    "하이키 서이",
    "하이키 리이나",
    "하이키 휘서",
    "하이키 옐",
    "르세라핌 카즈하",
    "르세라핌 홍은채",
    "르세라핌 허윤진",
]


def crawl_image(celeb_name, gender):
    print(celeb_name)
    elem = driver.find_element(by=By.NAME, value="q")
    elem.clear()
    elem.send_keys(f'allintitle:"{celeb_name}" -닮은 -청량 -단체사진 -다이어트 -살 -레어리 -facebook')
    elem.send_keys(Keys.RETURN)
    idx = 0

    SCROLL_PAUSE_TIME = 3
    # Get scroll height
    last_height = driver.execute_script("return document.body.scrollHeight")
    while True:
        # Scroll down to bottom
        driver.execute_script("window.scrollTo(0, document.body.scrollHeight);")
        # Wait to load page
        time.sleep(SCROLL_PAUSE_TIME)
        # Calculate new scroll height and compare with last scroll height
        new_height = driver.execute_script("return document.body.scrollHeight")
        if new_height == last_height:
            try:
                driver.find_element(By.CSS_SELECTOR, ".LZ4I").click()
            except Exception:
                break
        last_height = new_height

    images = driver.find_elements(By.CSS_SELECTOR, ".rg_i.Q4LuWd")
    driver.execute_script("window.scrollTo(0, 0);")
    for image_elem in images[:500]:
        try:
            idx += 1
            image_elem.click()
            time.sleep(0.3)

            down_link = driver.find_element(
                by=By.XPATH,
                value='//*[@id="Sva75c"]/div[2]/div[2]/div[2]/div[2]/c-wiz/div/div/div/div/div[3]/div[1]/a/img[1]',
            ).get_attribute("src")
            dir_path = os.path.join("images", gender, celeb_name)
            if not os.path.exists(dir_path):
                os.makedirs(dir_path)

            try:
                r = http.request("GET", down_link, preload_content=False, headers=header)
                with open(os.path.join(dir_path, f"{idx:02d}.png"), "wb") as out:
                    out.write(r.read())
                r.close()
            except Exception:
                z = down_link[down_link.encode().find(b"/9") :]
                Image.open(io.BytesIO(base64.b64decode(z))).save(os.path.join(dir_path, f"{idx:02d}.png"))
        except Exception:
            print("Error: ", celeb_name, idx)


# celeb_list = ["아이즈원 장원영"]
for celeb_name in celeb_list:
    try:
        crawl_image(celeb_name, "여자")
    except BaseException as e:
        print(str(e))
        pass

for celeb_name in celeb_list_men:
    try:
        crawl_image(celeb_name, "남자")
    except BaseException as e:
        print(str(e))
        pass

elem.send_keys(f'allintitle:"{celeb_name}" -닮은 -청량 -단체사진 -다이어트 -살 -레어리 -facebook')