WebDriverBiDi.Extensions 0.0.63

WebDriverBiDi.Extensions

Conveniences for the WebDriverBiDi .NET client library, for code that works with the protocol directly.

• One-call extension methods for common commands, returning natural .NET types: a screenshot as byte[], located nodes as a list of references, a script's result as the RemoteValue type you ask for.
• InputBuilder, which builds input.performActions sequences tick by tick, with helpers for clicking, typing, key chords, drag-and-drop, and scrolling.
• Works on .NET Standard 2.0, .NET 8 and later, and with native AOT.

Every type and extension method is in the namespace of the module it belongs with (WebDriverBiDi.BrowsingContext, WebDriverBiDi.Input, WebDriverBiDi.Network, WebDriverBiDi.Script, WebDriverBiDi.Session), so the using directives you already have bring them into scope. Each extension method sends one command, or a short fixed sequence, and takes the same timeout override and CancellationToken as the command it wraps.

Installation

dotnet add package WebDriverBiDi.Extensions

Quick Start

With a connected BiDiDriver:

using WebDriverBiDi.BrowsingContext;
using WebDriverBiDi.Script;

IReadOnlyList<BrowsingContextInfo> tabs = await driver.BrowsingContext.GetTopLevelBrowsingContextsAsync();
string contextId = tabs[0].BrowsingContextId;

await driver.BrowsingContext.NavigateAsync(contextId, "https://example.com", ReadinessState.Complete);
StringRemoteValue title = await driver.Script.CallFunctionAsync<StringRemoteValue>(contextId, "() => document.title");
byte[] png = await driver.BrowsingContext.CaptureScreenshotAsync(contextId);
File.WriteAllBytes("example.png", png);

A script that throws is reported as a ScriptException, whose Details carry the browser's description of the JavaScript exception.

Input

using WebDriverBiDi.BrowsingContext;
using WebDriverBiDi.Input;
using WebDriverBiDi.Script;

IReadOnlyList<SharedReference> searchBoxes = await driver.BrowsingContext.LocateNodesByCssSelectorAsync(contextId, "input[name=q]");

InputBuilder builder = new InputBuilder()
    .AddClickOnElementAction(searchBoxes[0])
    .AddSendKeysToActiveElementAction("WebDriver BiDi")
    .AddSendKeysToActiveElementAction(Keys.Enter);
await driver.Input.PerformActionsAsync(contextId, builder);

// Releases any keys or buttons an action sequence left pressed.
await driver.Input.ReleaseActionsAsync(contextId);

Each action added is one tick, in which every other input source pauses; AddActions puts several sources' actions in one tick. Text is typed one user-perceived character at a time, so emoji and combining sequences arrive whole, on .NET Framework too. Keys names the special keys (Enter, Shift, the arrows, and so on). For finer control, create sources with CreateKeyInputSource, CreatePointerInputSource (mouse, pen, or touch), CreateWheelInputSource, and CreateNoneInputSource, and add the actions they create.

Documentation

The WebDriverBiDi.Extensions guide covers the package in more detail, and the API reference lists every member.