Web browser automation with Selenium

There are two ways to automate pages in web browser (click links, fill web forms, extract data, etc):

• UI element functions (elm). See recipe Web automation.
• Selenium WebDriver. It has more features, for example wait for web page loaded, extract text, execute JavaScript. It can be faster and more reliable. But not so easy to use.

At first install NuGet packages Selenium.Support and Selenium.WebDriver.ChromeDriver. Update them for each new Chrome version.

This script starts new Chrome instance and then calls various automation functions.

/*/ nuget -\Selenium.Support; nuget -\Selenium.WebDriver.ChromeDriver; /*/
using OpenQA.Selenium;
using OpenQA.Selenium.Chrome;
using OpenQA.Selenium.Interactions;
using OpenQA.Selenium.Support.Extensions;
using OpenQA.Selenium.Support.UI;

//Set Chrome options.
ChromeOptions options = new();
//Enable and maybe edit this if want to use an existing profile. To get profile path, in Chrome open URL "chrome://version/".
//	Then before starting this script also may need to close existing Chrome instances that use this profile.
//options.AddArguments($"user-data-dir={folders.LocalAppData + @"Google\Chrome\User Data"}", "profile-directory=Profile 1");

//Start new Chrome instance.
var service = ChromeDriverService.CreateDefaultService(pathname.getDirectory(typeof(ChromeDriver).Assembly.Location));
service.HideCommandPromptWindow = true;
using var driver = new ChromeDriver(service, options);
driver.Manage().Window.Maximize();

//Open web page.
driver.Navigate().GoToUrl("https://www.libreautomate.com/forum/");

//Get page title, URL, HTML source. Extract all text.
print.it(driver.Title);
print.it(driver.Url);
//print.it(driver.PageSource);
var body = driver.FindElement(By.TagName("body"));
print.it(body.Text);

//Find and click a link. It opens new page and waits util it is loaded.
var e = driver.FindElement(By.LinkText("Shared C# code"));
e.Click();

//Find a text input field and enter text. To find it use XPath copied from Chrome Developer Tools.
var e2 = driver.FindElement(By.XPath("//*[@id='search']/input[1]"));
e2.SendKeys("find");

//How to invoke mouse or keyboard actions.
Actions action = new Actions(driver);
action.ContextClick(e2).Perform();

//Wait. Then the script will close the web browser.
3.s();
dialog.show("Close web browser", x: ^1);

This script connects to an existing Chrome instance and opens a web page.

/*/ nuget -\Selenium.Support; nuget -\Selenium.WebDriver.ChromeDriver; /*/
using OpenQA.Selenium;
using OpenQA.Selenium.Chrome;
using OpenQA.Selenium.Interactions;
using OpenQA.Selenium.Support.Extensions;
using OpenQA.Selenium.Support.UI;

var w = wnd.find(0, "*- Google Chrome", "Chrome_WidgetWin_1");
using var driver = Selenium.ConnectToChrome(w);
driver.SwitchTo().NewWindow(WindowType.Tab);
driver.Navigate().GoToUrl("https://www.libreautomate.com");

/// <summary>
/// Misc helper functions for web browser automation using Selenium.
/// </summary>
static class Selenium {
	/// <summary>
	/// Connects to a Chrome process, which must be started with command line like <c>--remote-debugging-port=9222</c>.
	/// </summary>
	/// <exception cref="Exception">Failed.</exception>
	/// <example>
	/// <code><![CDATA[
	/// var w = wnd.find(0, "*- Google Chrome", "Chrome_WidgetWin_1");
	/// using var driver = Selenium.ConnectToChrome(w);
	/// driver.Navigate().GoToUrl("https://www.libreautomate.com");
	/// ]]></code>
	/// Starting Chrome.
	/// <code><![CDATA[
	/// run.it("chrome.exe", "--remote-debugging-port=9222");
	/// ]]></code>
	/// </example>
	public static ChromeDriver ConnectToChrome(wnd w) {
		var s = process.getCommandLine(w.ProcessId, true) ?? throw new AuException("Failed to get Chrome command line");
		var da = StringUtil.CommandLineToArray(s).FirstOrDefault(o => o.Starts("--remote-debugging-port=")) ?? throw new AuException("Chrome must be started with command line like --remote-debugging-port=9222");
		ChromeOptions options = new() { DebuggerAddress = "127.0.0.1:" + da[24..] };
		var service = ChromeDriverService.CreateDefaultService(pathname.getDirectory(typeof(ChromeDriver).Assembly.Location));
		service.HideCommandPromptWindow = true;
		return new ChromeDriver(service, options);
	}
}

This code can be used to run Chrome so that the above script can connect to it.

run.it("chrome.exe", "--remote-debugging-port=9222");

More info on the internet.