基于XCode开发macOS原生浏览器交互应用的技术求助

Hey there! Let's work through your problem step by step—since you're new to macOS, Xcode, and Cocoa, I'll break this down into clear, actionable parts to help you build a native macOS app that interacts with browsers like Chrome or Firefox.

First: Fixing That C# NuGet Error on macOS #

The error you're seeing (Could not resolve type with token 010000b3...) stems from mismatched .NET runtime references in your Visual Studio for Mac project. Here's how to fix it:

• Target the correct .NET framework: Ditch .NET Standard 2.0 and use a macOS-specific .NET SDK (6.0 or newer). Edit your .csproj file to set this:

  <PropertyGroup>
    <TargetFramework>net6.0-macos</TargetFramework>
    <OutputType>Application</OutputType>
    <UseMacOSSdk>true</UseMacOSSdk>
  </PropertyGroup>
  

  This ensures your project uses the runtime optimized for macOS, not a generic standard library.
• Clean NuGet caches: In Visual Studio for Mac, go to Tools > NuGet Package Manager > Manage NuGet Packages for Solution, click the gear icon, and select Clear All NuGet Cache(s). Then reinstall your packages to resolve version conflicts.
• Generate a native .app bundle: Stop targeting console output—macOS native apps are .app packages. In your project properties, set the Output Type to Application instead of Console Application.

Second: Building a Native macOS App to Interact with Browsers #

You have two solid paths here: using Swift (the official macOS native language) or adjusting your C# setup. I'll cover both.

Option 1: Swift + AppKit + Selenium (Recommended for True Native) #

Selenium is the most widely used tool for browser automation, and it works seamlessly with Swift on macOS. Here's how to set it up:

1. Create an Xcode Project:
   • Open Xcode, select macOS > App, name it something like BrowserAutomator, and choose Swift as the language (stick with AppKit for simpler UI basics if you're new).
2. Install Dependencies:
   • Install Homebrew (if you don't have it) via the official script, then run these commands in Terminal to get browser drivers:

     brew install chromedriver  # For Chrome
     brew install geckodriver   # For Firefox
     

   • Add Selenium to your project via Swift Package Manager: Go to File > Add Packages, enter https://github.com/SeleniumHQ/selenium, select the latest version, and add it to your app target.
3. Sample Automation Code:
   Add a button to your ViewController.swift that launches Chrome and runs a simple search:

   import Cocoa
   import Selenium
   
   class ViewController: NSViewController {
   
       override func viewDidLoad() {
           super.viewDidLoad()
           // Add a clickable button to trigger automation
           let launchButton = NSButton(title: "Launch Chrome", target: self, action: #selector(launchChromeAndSearch))
           launchButton.frame = CGRect(x: 50, y: 50, width: 180, height: 40)
           view.addSubview(launchButton)
       }
   
       @objc func launchChromeAndSearch() {
           do {
               // Initialize Chrome driver
               let driver = try ChromeDriver()
               // Navigate to Google
               try driver.get("https://www.google.com")
               // Find the search box and type a query
               let searchBox = try driver.findElement(By.name("q"))
               try searchBox.sendKeys("macOS native browser automation")
               // Submit the search
               try searchBox.submit()
               // Wait 5 seconds to see the result
               Thread.sleep(forTimeInterval: 5)
               // Clean up: Quit the driver
               try driver.quit()
           } catch {
               print("Automation error: \(error.localizedDescription)")
           }
       }
   }
   

   • Note: Make sure your ChromeDriver version matches your installed Chrome version—run brew upgrade chromedriver to update if you get a version mismatch error.

Option 2: C#/.NET macOS App with Selenium #

If you prefer sticking with C#, here's how to adjust your project to work with browser automation:

1. Set up your project: Follow the NuFix fixes above to target net6.0-macos as an Application.
2. Install NuGet packages: Add Selenium.WebDriver, Selenium.WebDriver.ChromeDriver, and Selenium.WebDriver.GeckoDriver to your project.
3. Sample Code:
   Add this to your window controller to create a button that launches Chrome:

   using System;
   using OpenQA.Selenium;
   using OpenQA.Selenium.Chrome;
   using AppKit;
   
   namespace BrowserAutomatorCS
   {
       public partial class MainWindowController : NSWindowController
       {
           public MainWindowController(IntPtr handle) : base(handle) { }
   
           [Export("launchChrome:")]
           public void LaunchChrome(NSObject sender)
           {
               var chromeOptions = new ChromeOptions();
               // Optional: Run in headless mode (no visible window)
               // chromeOptions.AddArgument("--headless=new");
               
               using var driver = new ChromeDriver(chromeOptions);
               driver.Navigate().GoToUrl("https://www.google.com");
               var searchBox = driver.FindElement(By.Name("q"));
               searchBox.SendKeys("C# macOS browser automation");
               searchBox.Submit();
               System.Threading.Thread.Sleep(5000);
               driver.Quit();
           }
       }
   }
   

   • Bind this method to a button in your window (use Xcode's Interface Builder to link the button action to launchChrome:).

Third: Alternative Native Approach (Apple Accessibility API) #

If you want to avoid third-party tools like Selenium, you can use Apple's built-in Accessibility API to control browsers directly. This is more complex but offers deeper native integration. For example, to control a running Chrome instance:

import Cocoa
import ApplicationServices

func controlChromeNative() {
    // Get Chrome's process ID if it's running
    guard let chromePID = NSWorkspace.shared.runningApplications.first(where: { $0.bundleIdentifier == "com.google.Chrome" })?.processIdentifier else {
        print("Chrome is not running!")
        return
    }
    
    let chromeApp = AXUIElementCreateApplication(chromePID)
    var windows: AnyObject?
    // Get all Chrome windows
    AXUIElementCopyAttributeValue(chromeApp, kAXWindowsAttribute as CFString, &windows)
    
    if let windowList = windows as? [AXUIElement] {
        for window in windowList {
            // Example: Zoom the window
            AXUIElementPerformAction(window, kAXZoomAction as CFString)
            // You can further interact with elements like the address bar using accessibility attributes
        }
    }
}

• Important: You'll need to enable accessibility permissions for your app in System Settings > Privacy & Security > Accessibility to use this API.

内容的提问来源于stack exchange，提问作者Miguel Mesquita Alfaiate