基于 Selenium 的 Python 自动化测试框架 seleniumbase

https://seleniumbase.io/integrations/docker/ReadMe/

https://seleniumbase.io/

def mymain(url="https://hpo.jax.org/data/ontology"):
    '''
    SeleniumBase常用命令列表详见： SeleniumBase/help_docs/method_summary.md

    报下面这个错时，可以在/etc/hosts中添加  127.0.0.1 localhost试试
    selenium.common.exceptions.WebDriverException: Message: Can not connect to the Service /usr/local/lib/python3.10/dist-packages/seleniumbase/drivers/chromedriver

    找到元素后，单击元素
    robot_label = sb.find_element("label.altcha-label")
    sb.click(robot_label)
    '''
    # with SB(headless=True , binary_location="/usr/bin/google-chrome") as sb:
    with SB(headless=True) as sb:

        sb.open(url)
        print(f"firt url: {url}")
        # 下载目录
        download_folder = sb.get_downloads_folder()
        print(download_folder)

        '''
        保存当前html
        file_name = "wmba.html"
        sb.save_page_source(file_name)
        
        也可利用save_screenshot截图保存
        
        筛选超链接，按超链接在界面上显示的内容查找
        sb.wait_for_element("a:contains('Parent Directory')", timeout=1000)
        '''

        sb.wait_for_ready_state_complete()
        sb.sleep(2)  # 给动态内容一点加载时间
        items = sb.find_elements('h3[class*="download-banner-title"]')
        for item in items:
            print(item.text)
            item.click()
            print('11111', sb.get_current_url())
            sb.sleep(2)
        # 获取下载目录
        sfd = sb.get_downloaded_files()
        print(sfd)

        '''
        阻塞等待    抛出断言异常，测试失败
        '''
        sb.assert_downloaded_file("hp.json", timeout=60)  # hp.obo hp.owl hp.json

        # 3. 获取文件路径并进行后续处理
        file_path = sb.get_path_of_downloaded_file("hp.json")
        print(f"文件已保存至: {file_path}")

 

 

from seleniumbase import SB
 
def exec(url='https://pubmed.ncbi.nlm.nih.gov/39627211/'):
    with SB(headless=True) as sb:
        sb.open(url)
        print(f"firt url: {url}")
        # sb.wait_for_element('.heading-title')

        doi_selector1 = '//*[@id="full-view-identifiers"]/li[3]/span/a'
        _doi = sb.get_text(doi_selector1)
        print(_doi)
        # full_text_div = sb.find_elements('//*[@id="full-view-identifiers"]/li[3]/span/a')

        _title = sb.get_text(".heading-title")
        print(f"_title: {_title}")


if __name__ == "__main__":
    exec()

　　

制作镜像：

git clone http://github.com/seleniumbase/SeleniumBase.git

cd SeleniumBase

docker build -t seleniumbase .

 

SeleniumBase：功能全面的浏览器自动化框架。该项目是基于 Selenium 的 Python 自动化测试框架，集成了爬虫、自动化测试和生成报告等多种功能。它提供了丰富的示例，并且独特的 UC 模式，可以帮助开发者在进行浏览器自动化操作时避免被检测出来。

from seleniumbase import BaseCase
BaseCase.main(__name__, __file__)

class TestSimpleLogin(BaseCase):
    def test_simple_login(self):
        self.open("seleniumbase.io/simple/login")
        self.type("#username", "demo_user")
        self.type("#password", "secret_pass")
        self.click('a:contains("Sign in")')
        self.assert_exact_text("Welcome!", "h1")
        self.assert_element("img#image1")
        self.highlight("#image1")
        self.click_link("Sign out")
        self.assert_text("signed out", "#top_message")

地址：github.com/seleniumbase/SeleniumBase