python+playwright

一：用python+playwright对比网页图片和本地图片

百度图片.png

from playwright.sync_api import Playwright, sync_playwright
import base64

# 初始化Playwright
with sync_playwright() as p:
    browser = p.chromium.launch(headless=False)
    page = browser.new_page()

    # 打开网页
    page.goto('https://www.baidu.com')

    # 获取图片url
    image_url = page.locator('#s_lg_img_new')
    image_url = image_url.get_attribute('src')

    #将图片转成base64
    image_data_base64 = page.evaluate('''(image_url) => {
        return fetch(image_url)
            .then(response => response.arrayBuffer())
            .then(buffer => {
                const byteArray = new Uint8Array(buffer);
                return btoa(String.fromCharCode.apply(null, byteArray));
            });
    }''', image_url)

    # Base64エンコードされた画像データをデコードしてファイルに保存または他の操作を行うことができます
    # 把base64的数据转成dyte数组
    image_data_bytes = base64.b64decode(image_data_base64)
    def test_1():
        # 与本地图片比较，判断是否一致
        with open('baidu.png', 'rb') as localFile:
            assert localFile.read() == image_data_bytes
    browser.close()

二：用page.on('response', data => {});方法监听，获取图片url，比对图片

import base64

from playwright.sync_api import sync_playwright


def image_compare(response):#image_compare(response)方法，遍历response的url，查看是否存在图片
    if '1771365ca98ig9er706.jpg' in response.url:
        with open(r'./1771365ca98ig9er706.jpeg', 'rb') as localFile:
            print(localFile.read() == response.body())

# Playwrightの初期化
with sync_playwright() as p:
    browser = p.chromium.launch(headless=False)
    page = browser.new_page()

    # 监听网页的request
    #page.on('request', lambda request: print(request.url, request))
    # 监听和遍历网页的response，每次遍历都调用image_compare()方法，将response作为参数传入方法
    page.on('response', lambda response: image_compare(response))

    # 打开网页
    page.goto('https://www.douban.com/')

三：用page.on()和open()方法下载图片

from playwright.sync_api import sync_playwright
import base64

# image_compare(response)方法，遍历response的url，查看是否存在图片
def image_download(response):
    if '1771365ca98ig9er706.jpg' in response.url:
        # with open(r'./1771365ca98ig9er706.jpeg', 'rb') as localFile:
        #     print(localFile.read() == response.body())
        with open('./downloaded_doubanban.png', 'wb') as image_file:
            image_file.write(response.body())
            print('画像データを取得しました')


# Playwrightの初期化
with sync_playwright() as p:
    browser = p.chromium.launch(headless=False)
    page = browser.new_page()

    # 监听网页的request
    # page.on('request', lambda request: print(request.url, request))
    # 监听和遍历网页的response，每次遍历都调用image_compare()方法，将response作为参数传入方法
    page.on('response', lambda response: image_download(response))

    # 打开网页
    page.goto('https://www.douban.com/')

四：用base46转成字节文件

from playwright.sync_api import sync_playwright
import base64

# Playwrightの初期化
with sync_playwright() as p:
    browser = p.chromium.launch()
    page = browser.new_page()

    # ウェブページにアクセス
    page.goto('https://www.example.com')  # 画像を取得したいページのURLに変更

    # 画像のURLを指定
    image_url = 'https://www.example.com/path/to/hokkaido.png'  # 画像のURLに変更

    # ウェブページ内でJavaScriptを実行して画像データを取得
    image_data_base64 = page.evaluate('''(image_url) => {
        return fetch(image_url)
            .then(response => response.arrayBuffer())
            .then(buffer => {
                const byteArray = new Uint8Array(buffer);
                return btoa(String.fromCharCode.apply(null, byteArray));
            });
    }''', image_url)

    if image_data_base64:
        # Base64エンコードされた画像データをデコードしてファイルに保存または他の操作を行うことができます
        image_data_bytes = base64.b64decode(image_data_base64)
        with open('./downloaded_image.png', 'wb') as image_file:
            image_file.write(image_data_bytes)

        print('画像データを取得しました')
    else:
        print('画像データの取得に失敗しました')

    # ブラウザを閉じる
    browser.close()