python的selenium学习

利用selenium获取图片并下载；基于pixabay，获取图片

# -*- coding:utf-8 -*-

import requests
from urllib.parse import quote
from time import sleep
import threading
from tqdm import tqdm
from selenium import webdriver
import time


def get_type(type):
    options = webdriver.ChromeOptions()
    "#配置忽略HTTPS安全证书提示"
    options.add_argument('--ignore-certificate-errors')
    # Chrome 正在受到自动化软件控制删除
    options.add_argument("--disable_infobars")

    # options.add_argument('--headless')

    driver = webdriver.Chrome(r"填写自己谷歌的驱动器的地址",
                              chrome_options=options)

    url = "https://pixabay.com/zh/photos/search/" + quote(type)  # 转为assic码
    driver.get(url)
    driver.maximize_window()
    # driver.refresh()
    sleep(2)
    photos_numbers = driver.find_element_by_xpath('//*[@id="content"]/div/div[3]/div/h1').text
    photos_numbers = str(photos_numbers).split(" ")
    print("一共检测照片数量", photos_numbers[0])
    pages = (int(photos_numbers[0]) // 100) + int(1)
    page = int(input("输入的页码:"))
    while page <= pages:
        photo_dic = {}
        for i in range(1, page + 1):
            print("正在获取页面信息......")
            # 执行这段代码，会获取到当前窗口总高度
            js = "return action=document.body.scrollHeight"
            # 初始化现在滚动条所在高度为0
            height = 0
            # 当前窗口总高度
            new_height = driver.execute_script(js)

            while height < new_height:
                # 将滚动条调整至页面底部
                for i in tqdm(range(height, new_height, 100)):  # 速度为没像素100的速度进行
                    driver.execute_script('window.scrollTo(0, {})'.format(i))
                    sleep(0.8)
                height = new_height
                sleep(1.5)
                new_height = driver.execute_script(js)

            photos_links = driver.find_elements_by_class_name('photo-result-image')
            print(len(photos_links))
            for photo_links in photos_links:
                photo_link = photo_links.get_attribute("src")
                photo_name = photo_links.get_attribute("alt")
                photo_dic[photo_link] = photo_name  # 字典存储信息，key存储下载地址，val存名称；可以考虑使用item迭代器，快一点
            driver.implicitly_wait(2)
            driver.find_element_by_xpath('//*[@id="content"]/div/a').click()
            # 利用按钮判断存在下一页，然后同样的方式获取图片信息
        return photo_dic


def download_photo(url, name):
    try:
        photo = requests.get(url, timeout=10)
    except requests.exceptions.ConnectTimeout:
        print("timeout")
    # path= input("")
    path = r"D:/python_code/python_test/photo_get/" + str(name) + ".jpg"  # 文件记录路径
    with open(path, "wb") as f:
        f.write(photo.content)


if __name__ == "__main__":
    photo_dict = get_type("夜空")
    print("正在下载.....")
    start = time.time()
    for url, name in tqdm(photo_dict.items()):
        # print(url)
        t = threading.Thread(target=download_photo, args=(url, name))
        t.start()
    end = time.time()
    print("下载总耗时: %s 分钟!" % (round((end - start) / 60, 1)))