stirno / fluentautomation Goto Github PK

I.Expect.Text("Add").In("User.AddButton") - returns 'Value assertion failed. Expected element [input[Value='Add']] to have a value of [Add] but actual value is [string.Empty]

Error message should be input[Text='Add']

I am unable to find a "I.Close(url)" to compliment the "I.Open(url)"

Not working properly in Selenium using Firefox or Chrome drivers. Tracking so I can tackle this later.

System.NullReferenceException: Object reference not set to an instance of an object.
at FluentAutomation.SeleniumWebDriver.AutomationProvider.HandleAlertDialog(String expectedMessage)
at FluentAutomation.API.ExpectManager.Alert(String alertMessage)

I.Use(BrowserType.InternetExplorer);

What if I want to run the same test against different browsers?

I.Use may not be necessary assuming this won't support opening multiple browser sessions during a single test then the browser can be opened automatically on the first call to I.Action

_browser can be referenced from a property and if null can open a new session

Options to set browser can be in config or I.Set.Browser(BrowserType.InternetExplorer)

If I expect count of 2 then I get an error 'Count assertion failed. Expected there to be [2] elements matching [somevalue]].Actual count is [1].

Count seems to be zero based and shouldn't be

FluentAutomation.API.AssertException: Value assertion failed. Expected element [#login-button] to have a value of [Login] but actual value is [string.Empty].

Trying to get this working with Xunit + Selenium WebDriver + Resharper Test Runner

All I get back is System.InvalidOperationException: No automation provider available.

I have the relevant Nuget packages installed:

Is something not copying correctly when the test runner runs?

Apparently there's some authentication logic within the AutomationProvider, but it's not clear how to expose that to a test.

Is this a case where I just create my own AutomationProvider/CommandManager and or what are your intentions with that basic auth logic?

(might not be a bug)

Add css selector used on an assert. Also place values in the assert in [] to separate the selector or text element used from the assert error sentance.

If a have a css selector such as: " .node a:eq({0})", I need to be able to pass in the index and have it return the text value of that css selector

Should be able to add logging output to a file or to test output.

Related to Issue #9 but limited to removing the functionality of I.Use to open the browser and instead setting the desired browser if different the default. This would open the browser on the first action and not on I.Use. Ability to change default browser in config file is separate issue

An alert dialog that doesn't get handled causes the test runner to wait for user interaction, which never comes.

I.Drag throws NotImplementedException with the Selenium provider.

FluentAutomation.API.AssertException: Value assertion failed. Expected element [div.error] to have a value of [User name or password is incorrect.] but actual value is [string.Empty].

Confirmed in jquery that when running:
$("div.error")
I get 1 result:
<div class="error">User name or password is incorrect.</div>

We have a select list and a list of items can be added to it. If an item is already there it should not be added to the list. We need a way to validate that a list does not have duplicates or specify that a specific item is not duplicated.

If a selector contains multiple elements the text content of each is mashed together for an I.Expect.Value. Improve on this by allowing a list of values to compare to the content of a group of elements.

Great project - would it be possible to add a property to access your current WebDriver, i.e. FirefoxDriver. I'm not sure how it would with I, something like

I.GetDriver().FindElement(By.Name("q"))

I need this as sometimes Sizzle break with jQuery, or doesn't give us the ability to do a text search of the innerHtml.

I.Expect.Class('.error').On('#field');

Since I have some problems with WebDriver when running tests in batch is it possible to update to the latest version? Thanks!

string.Format error when class is not found on element in Selenium using Chrome. Haven't tested in other browsers/WatiN.

It would be great to be able to do something like this:

I.Expect.Attribute("src").Value("tick.png").In("#my_img");

or is it already available?

Would be nice to be able to extend the ActionManager class to help encapsulate repeated test steps.

I.Expect.Url(Config.LoginPage);

public static string LoginPage
{
get
{
return string.Format("http://{1}/Account/LogOn?returnUrl=http%3a%2f%2f{0}.{1}%3a80", Customer, Site);
}
}

worked previously but now getting:
URL Assertion failed. Expected URL
http://domain.tld/Account/LogOn?returnUrl=http%3a%2f%2fcustomer.domain.tld%3a80 but actual URL is
http://domain.tld/Account/LogOn?returnUrl=http://customer.domain.tld:80.

because the window is now maximized, I can't visually verify the url

Doing and I.Expect.Url right after I.Click that triggers a page load fails due to race condition. I.Expect.Url can trigger before the page has loaded and therefore compares against the previous page.

In firebug, this works: $(".wizard select.It:eq(0)");

When using: $(".wizard select.It:eq(0)"); this error occurs:
SelectElement value assertion failed. Expected element [.wizard select.It :eq(0)] to have a selected value of [Select...] but actual selected value is [string.Empty].

I am now able to get it to work with [.wizard select.It :eq(0) option:selected]

In the SeleniumWebDriver.Element.OnChange method (line 121) has problems with IE, including IE9:

    public virtual void OnChange()
    {
        ((IJavaScriptExecutor)_driver).ExecuteScript(string.Format("if (typeof jQuery != 'undefined') {{ jQuery(\"{0}\").click(); }}", _fieldSelector.Replace("\"", "")));
    }

The answer appears to be to use click(), from this discussion: http://stackoverflow.com/questions/208471/getting-jquery-to-recognise-change-in-ie

Or if you're using jQuery 1.5 the problem is resolved but unfortunately we're using jQuery 1.3.

Once I setup Github I'll try to push a patch

Counts of items matching a selector..

I.Expect.Count(10).Of("selector");

If a row is entered on an UI, and the css selector is same for all items on the UI except for the :eq index number, I need to be able to get the index number of the customer identified by a title value.

For example, the UI consists of a sorted grid with two columns. In the first column it lists the customer in alpha order. The next column is a delete buttons which correspond to the customer that the row it is on.

Customer | Action |Index |
Dick's Bar and Grill | Delete |0 |
Don's Kwik Shop | Delete |1 |
Joe's Bait Store | Delete |2 |
The Gas Station | Delete |3 |

Now, if I add "Apple Pie Restaurant" to this, then all the indexes are going to change. If I want to click the Delete button for Joe's Bait Store, I need to be able to get the index for Joe's Bait Store, so I can click the correct Delete button.

Need the ability to sendkeys - tried to use:

System.Windows.Forms.SendKeys.SendWait("TAB");

but no luck

May seem obvious but is it possible to test and ensure a line of text is "not" found in the page once it has rendered?

When using the WaitUntil functionality with Selenium I kept running into issues where the automation was trying to run before the page was fully loaded.
This was causing JS errors.

My current work around look like is:
public void WaitUntil(Expression conditionAction, TimeSpan timeout)
{
this.Act(() =>
{
DateTime dateTimeTimeout = DateTime.Now.Add(timeout);
string source = this.webDriver.PageSource;

            while (DateTime.Now < dateTimeTimeout)
            {
                System.Threading.Thread.Sleep(Settings.DefaultWaitUntilThreadSleep);

                if (source.Equals(this.webDriver.PageSource))
                {
                    break;
                }
                else
                {
                    source = this.webDriver.PageSource;
                }
            }

            bool threwException = false;
            var compiledAction = conditionAction.Compile();

            FluentException lastFluentException = null;

            do
            {
                try
                {
                    threwException = false;
                    compiledAction();
                }
                catch (FluentException ex)
                {
                    threwException = true;
                    lastFluentException = ex;
                }
                catch (Exception ex)
                {
                    throw new FluentException("An unexpected exception was thrown inside WaitUntil(Action). See InnerException for details.\r\n{0}", ex, source);
                }

                if (!threwException)
                {
                    break;
                }

                System.Threading.Thread.Sleep(Settings.DefaultWaitUntilThreadSleep);
            }
            while (DateTime.Now < dateTimeTimeout);

            // If an exception was thrown the last loop, assume we hit the timeout
            if (threwException == true)
            {
                throw new FluentException("Conditional wait passed the timeout [{0}ms]. See InnerException for details of the last FluentException thrown.\r\n{1}", lastFluentException, timeout.TotalMilliseconds, source);
            }
        });
    }

We have a textbox for search. When you type in the textbox a list is displayed below it. This list closes when you lose focus. When trying to click on an item in the list I can see watin highlight the item that it clicked on but nothing seems to happen. This works repeating the steps manually. It seems that Fluent/Watin is causing some additional interaction from what a user would do that is causing this list to lose focus and close.

Allow for lambdas to support:

I.Expect.This(s => s.Contains("string")).In("selector");
I.Expect.Url(url => url.StartsWith("https://").In("selector");

Need ability to switch document in case of iframes in page. Or have a way of having access to the driver and do it classic way ie

driver.SwitchTo().Frame(someFrame).

Maybe this can be achieved in another way already , but I couldn't find it.

Thanks for your help

It is difficult to calculate a point from the corner of the screen which will not be consistent with screen resolution. Need the ability to specify an element in the browser to be a point of reference from which the point is calculated

While looking into another reported issue, noticed that the Selenium providers Focus implementation is.. incomplete..

public virtual void Focus()
{
}

Oops.

Assert.Fail failed. Value assertion failed. Expected element [input.email:eq(0)] to have a value of [] but actual value is [].

email = string.Empty.

Perhaps it is trying to compare a null to a string.Empty

With this code:

If I do an "I.Expect.This("add").In("some-node a:eq(0)"), it will not fail but should because "display: none"

I.Enter("Jason").Quickly.In("#name");
I.Expect.This("Jason").In("#name");

Gives error Assert.Fail failed. Value assertion failed. Expected value of Jason but actual value is .

Value assertion failed. Expected element [.success-message ] to have a value of [Successfully added] but actual value is [Successfully added ].

The expected value is stored in a resource file which trims values but the actual value is not trimmed. Actual value should be trimmed.

Hello,

we've tried out the version that is under development right now (brought down the codez, compiled and used it), and we're missing the Quickly-way of entering text. That is - no events firing.

Is this by design or has it slipped away

Comparison from a select using I.Expect.This().In() on a select box compares the entire list of options not the currently selected option.

IE window now maximizes and this should be user configurable

I.Hover('some-selector') isn't working properly on Selenium. The mouse moves, however, it doesn't appear to be scoping the hover in the context of the browser window.