편집

다음을 통해 공유


Call web-side code from native-side code

Using JavaScript in WebView2 controls allows you to customize native apps to meet your requirements. This article explores how to use JavaScript in WebView2, and reviews how to develop using advanced WebView2 features and functions.

Before you begin

This article assumes that you already have a working project. If you don't have a project, and want to follow along, see Get started with WebView2.

Basic WebView2 functions

Use the following functions to begin embedding JavaScript in your WebView2 app.

API Description
ExecuteScriptAsync Run JavaScript in a WebView2 control. Call this method after the page Document Object Model (DOM) content is loaded or the navigation is completed. See Get started with WebView2.
AddScriptToExecuteOnDocumentCreatedAsync Runs on every page when the DOM is created. Call this method after the CoreWebView2 is initialized.

Scenario: ExecuteScript JSON-encoded results

Because the result of ExecuteScriptAsync is JSON-encoded, if the result of evaluating the JavaScript is a string, you will receive a JSON-encoded string and not the value of the string.

For example, the following code executes script that results in a string. The resulting string includes a quote at the start, a quote at the end, and escaping slashes:

string result = await coreWebView2.ExecuteScriptAsync(@"'example'");
Debug.Assert(result == "\"example\"");

The script returns a string that ExecuteScript JSON-encodes for you. If you call JSON.stringify from your script, then the result is doubly encoded as a JSON string the value of which is a JSON string.

Only the properties that are directly in the result are included in the JSON-encoded object; inherited properties aren't included in the JSON-encoded object. Most DOM objects inherit all properties, so you'll need to explicitly copy their values into another object to return. For example:

Script Result
performance.memory {}
(() => { const {totalJSHeapSize, usedJSHeapSize} = performance.memory; return {totalJSHeapSize, usedJSHeapSize}; })(); {"totalJSHeapSize":4434368,"usedJSHeapSize":2832912}

When we return just performance.memory we don't see any of its properties in the result because all properties are inherited. If instead, we copy particular property values from performance.memory into our own new object to return, then we see those properties in the result.

When executing script via ExecuteScriptAsync that script is run in the global context. It helps to have your script in an anonymous function so that any variables you define aren't polluting the global context.

For example:

  • If you run the script const example = 10; more than once, the subsequent times you run the script will throw an exception, because example was defined the first time you ran it.

  • If you instead run the script (() => { const example = 10; })(); the example variable is defined in the context of that anonymous function. That way, it's not polluting the global context, and can be run more than once.

Scenario: Running a dedicated script file

In this section, you access a dedicated JavaScript file from your WebView2 control.

Note

Although writing JavaScript inline may be efficient for quick JavaScript commands, you lose JavaScript color themes and line formatting that makes it difficult to write large sections of code in Visual Studio.

To solve the problem, create a separate JavaScript file with your code, and then pass a reference to that file using the ExecuteScriptAsync parameters.

  1. Create a .js file in your project, and add the JavaScript code that you want to run. For example, create a file called script.js.

  2. Convert the JavaScript file to a string that is passed to ExecuteScriptAsync, by pasting the following code after the page is done navigating:

    string text = System.IO.File.ReadAllText(@"C:\PATH_TO_YOUR_FILE\script.js");
    
  3. Pass your text variable using ExecuteScriptAsync:

    await webView.CoreWebView2.ExecuteScriptAsync(text);
    

Scenario: Removing drag-and-drop functionality

In this section, you use JavaScript to remove the drag-and-drop functionality from your WebView2 control.

To begin, explore the current drag-and-drop functionality:

  1. Create a .txt file in order to drag-and-drop. For example, create a file named contoso.txt and add text to it.

  2. Press F5 to build and run the project.

  3. Drag-and-drop the contoso.txt file into the WebView2 control. A new window opens, which is the result of the code in your sample project:

    Result of dragging and dropping contoso.txt

  4. Next, add code to remove the drag-and-drop functionality from the WebView2 control. Paste the following code after the CoreWebView2 object is initialized in your code:

    await webView.CoreWebView2.AddScriptToExecuteOnDocumentCreatedAsync(
       "window.addEventListener('dragover',function(e){e.preventDefault();},false);" +
       "window.addEventListener('drop',function(e){" +
          "e.preventDefault();" +
          "console.log(e.dataTransfer);" +
          "console.log(e.dataTransfer.files[0])" +
       "}, false);");
    
  5. Press F5 to build and run the project.

  6. Try to drag and drop contoso.txt into the WebView2 control. Confirm that you can't drag and drop.

Scenario: Removing the context menu

In this section, you remove the right-click menu from your WebView2 control.

To begin, explore the current functionality of the right-click menu:

  1. Press F5 to build and run the project.

  2. Right-click anywhere on the WebView2 control. The context menu displays the default right-click menu commands:

    The right-click menu, showing the default commands

    Next, add code to remove the right-click menu functionality from the WebView2 control.

  3. Paste the following code after the CoreWebView2 object is initialized in your code:

    await webView.CoreWebView2.ExecuteScriptAsync("window.addEventListener('contextmenu', window => {window.preventDefault();});");
    
  4. Press F5 to build and run the project. Confirm that you can't open a right-click menu.

See also