以IE为例，WatiN处理弹出窗口：
IE ie = new IE("string"); //打开指定web页
ie.Button(Find.ById("string")).Click(); //点击相应的按钮弹出需要测试的窗口TestSecondWindow
IE newIE = IE.AttachTo<IE>(Find.ByTitle("TestSecondWindow")); // 查找新窗口TestSecondWindow并赋给新的IE对象
//TODO: write the code you need to test in newIE
WatiN处理confirm对话框：
IE ie = new IE(string); //打开指定web页
ConfirmDialogHandler cdh = new ConfirmDialogHandler();
ie.AddDialogHandler(cdh); //将ConfirmDialogHandler对象与IE建立关联
ie.Button(Find.ById(string)).ClickNoWait(); //记住这里要用ClickNoWait而不能用Click,否则在模式窗口关闭之前代码不会继续执行。
cdh.WaitUntilExists();
cdh.OKButton.Click();
ie.WaitForComplete();
ie.RemoveDialogHandler(cdh); // ConfirmDialogHandler对象与IE取消关联
这里需要注意的是Click()方法与ClickNoWait()方法的区别，先来看看Click()方法
/// <summary>
/// Clicks this element and waits till the event is completely finished (page is loaded
/// and ready) .
/// </summary>
public virtual void Click()
{
ClickImpl(true);
}
接下来是ClickNoWait()方法：
/// <summary>
/// Clicks this instance and returns immediately. Use this method when you want to continue without waiting
/// for the click event to be finished. Mostly used when a
/// HTMLDialog or javascript popup is displayed after clicking the element.
/// </summary>
public virtual void ClickNoWait()
{
ClickImpl(false);
}
从上面两段代码上看，可以发现，它们都调用了ClickImpl(bool xxx)方法，只是一个是true一个是false，那么被定义的bool值又是什么呢？找到ClickImpl()方法：
/// <summary>
/// Handles the implementation of Click and ClickNoWait
/// </summary>
protected virtual void ClickImpl(bool waitforComplete)
{
if (!Enabled)
{
throw new ElementDisabledException(IdOrName, this);
} Logger.LogAction((LogFunction log) => { log("Clicking (no wait) {0} '{1}', {2}", GetType().Name, IdOrName, Description); }); Highlight(true); if (waitforComplete)
{
FireEvent("onclick");
}
else
{
FireEventNoWait("onclick");
} try
{
if (waitforComplete) WaitForComplete();
}
finally
{
Highlight(false);
}
}
原来bool值表示的是WaitForComplete, 撇开其他代码，先找到下面判断语句：
if (waitforComplete)
{
FireEvent("onclick");
}
else
{
FireEventNoWait("onclick");
}
问题转变成FireEvent()方法及FireEventNoWait()方法之间的区别，先找到FireEvent()方法：
/// <summary>
/// Fires the specified <paramref name="eventName"/> on this element
/// and waits for it to complete.
/// </summary>
/// <param name="eventName">Name of the event.</param>
public virtual void FireEvent(string eventName)
{
FireEvent(eventName, true, null);
}
看到这段代码，想来，和FireEventNoWait()方法之间的区别可能还会是其中某个bool值的区别，果不其然，看FireEventNoWait()方法：
/// <summary>
/// Only fires the specified <paramref name="eventName"/> on this element.
/// </summary>
public virtual void FireEventNoWait(string eventName)
{
FireEvent(eventName, false, null);
}
这两个方法都是调用FireEvent(string xxx, bool xxx, ?)方法，且区别仅在于bool值的设定，先不管，跳到这个方法看看先：
private void FireEvent(string eventName, bool waitForComplete, NameValueCollection eventProperties)
{
if (!Enabled)
{
throw new ElementDisabledException(IdOrName, this);
} Highlight(true); if (waitForComplete)
{
NativeElement.FireEvent(eventName, eventProperties);
WaitForComplete();
}
else
{
NativeElement.FireEventNoWait(eventName, eventProperties);
} Highlight(false);
}
到这里会发现，重点在以下代码中：
if (waitForComplete)
{
NativeElement.FireEvent(eventName, eventProperties);
WaitForComplete();
}
else
{
NativeElement.FireEventNoWait(eventName, eventProperties);
}
而最主要的在于WaitForComplete()方法的调用上，也就是在ClickNoWait()方法边上注释上所说的，如果用Click()方法，代码在往下走之前一定会等待模式窗口关闭，一旦关闭才是WaitForComplete()完成，因此，使用Click()方法的话，表面看起来整个程序会停着不动，从而达不到对对话框的处理效果。