I am currently building a test harness for the company I work at. I have experience both with C# and WatiN and have never encountered the issue I am now having.
Below, is a snippet of the markup for the page giving me the issue:
<div id="toggle1" class="NavLayout toggle">
<span onClick="toggleMenu(1, false);">
<span id="toggletext1">Quote Processing</span>
</span>
</div>
As you can see, I have a div, 2 spans and an image. I am using WatiN to try and click the image, that will then expand the menu, exposing yet another layer that I will need to click something else on. The problem I am having is in getting the 'Click' to happen. From what I can see in the snippet, it seems to me I need to be able to click the event, but cannot 'find' it with the code.
Any help out there to be had?
I have also had issues with clicking on certain elements.
I've run into issues where I could only click on an element if it was highlighted by mousing over the element.
Since I cannot see your code snippet, I can't tell if there is any javascript that deals with mouseover associated with the image, but if there is, you can try the following:
img.FireEvent("onmouseover");
img.FireEvent("onmousedown");
img.FireEvent("onMouseup");
You might also might want to try img.FireEvent("onclick") as well.
These are all guesses, since I can't see your code. It's also possible that rather than clicking on the image element itself, that you may want to try clicking on the parent object.
EDIT:
Ok, now that I can see your code, it appears that you should fire an onClick event against the span with the 'onclick' code in it.
I don't see an image listed in your code snippet, but this code should call the parent of the lowest level span.
Watin.Core.Span span = browserinstance
.Span(Find.By("innertext", "Quote Processing"));
span.Parent.FireEvent("onclick");
The DOM content that you intended to post is not visible. You might want to edit your post and check if it is visible.
In order to click on images
Watin.Core.Image img = browserinstance.Image(Find By Constraint);
if (img!=null and img.Exists)
img.ClickNoWait();
OR
img.FireEvent("onclick");
Related
I'm Writing a Selenium text case for an ASPX page. I want to click this ASP.NET asp:LinkButton element:
<asp:LinkButton runat="server" OnClick="Test_Click" Text="Just try to click me" ID="testtest123"></asp:LinkButton>
Which appears like this in the HTML page generated by .NET:
<a id="testtest123" href="javascript:__doPostBack('testtest123','')">Just try to click me</a>
But while Selenium has no problem finding the element, the Click() does nothing. I get no element missing exceptions or timeouts, the testcase just runs on as if the click fired properly.
Here's the simple line for clicking the element;
Browser.Driver.FindElement(By.Id("testtest123")).Click();
This code works fine with other ASP elements such as buttons and text boxes. I use C# to write my test cases using the 64 bit webdriver and IE 11.
I researched the solutions here: ASP.Net LinkButton Prevent Postback Not Working - JavaScript but they do not seem to apply to my situation:
Any help would be greatly appreciated!
EDIT: I posted another thread about the issue on this website:
https://code.google.com/p/selenium/issues/detail?id=7846&can=8&colspec=ID%20Stars%20Type%20Status%20Priority%20Milestone%20Owner%20Summary
I had a similar problem where some controls were rendered as espected and other don't, so are you sure the link ID is "testtest123"?
You have to set Client ID Mode to static in order to make the HTML ID be the same as in the .NET (the default is inherit, but in the web.config it is generlly set to be AutoID).
<asp:LinkButton runat="server" OnClick="Test_Click" Text="Just try to click me" ID="testtest123" ClientIDMode ="Static"></asp:LinkButton>
If the rendered HTML is actually <A id=testtest123 then that is not valid HTML since the attribute does not have quotation marks around the value. If this is actual value then the problem is further upstream than Selenium
One possibility is that selenium is correctly clicking the element, but that the linkbutton's javascript code is not executing quickly enough for the rest of your test. This is because when selenium encounters an <input type="submit"> tag (like asp.net button controls), it knows that it has to wait for the form submission before going to the next test step. However, a linkbutton is just an <a> tag, and worse, it has javascript as the href as opposed to a url. So it's possible that the postback javascript takes 50 milliseconds to execute or whatever, in that case, it's possible a test like this:
step n: click linkbutton - selenium does not know how long this takes so proceeds to next step
step n+1: click some button - if this happens before the js finishes executing, it would be like step n did not happen at all
This is plausible in the case of selenium because it operates at the speed of a program, whereas a human tester would never be able to click the next step quickly enough to cause such a problem. The fix for this is to add a delay step, or, to add a watch step on a javascript variable that you set on click of your linkbutton.
Edit:
Ok, so then this is not the issue. I'm almost positive it has something to do with the javascript in the href, though, so could you try the answer here: https://stackoverflow.com/a/18847713/1981387 ?
Background
Here's what I want to happen:
A user is on one page1.html (jsFiddle).
When they click on one of the <a href...> links, I want to navigate to page2.html (jsFiddle) and simulate the user entering the number into the textbox and clicking on the button.
Example: On page1.html, user clicks on display 2. Then we will navigate to page2.html and get an alert of 2 (as if user had entered 2 and clicked the button).
Question
How do I do this?
Is there a way to make a C# method with a specific URL to navigate to, such as page2.html/searchfor/2?
Or is there some way in JavaScript to manually go about doing other things after navigating to <a href="page2.html">?
Things I've tried
Using a <span> with an onclick function, but then it's not a true link like <a href> where I can middle click to open in new tab and right click to follow link
Wrapping my first attempt in <a href> tags, like <span>Display 2</span>. This still doesn't solve the problem of performing extra actions after navigation.
Note
I am building this webpage using Entity Framework, ASP.NET MVC, and C#.
I have simplified the problem for discussion purposes, but the concept is the same.
Try using the page2.html document's onload() function. You can pass parameters through the URL, then take that data and perform "other actions" as soon as the document is loaded.
My Question
I managed to answer myself, however the same set of functionality has another problem. For some reason the first postback of the save event of the posted file hits the Ol' Object not set to an instance of an object error, but on the second attempt of uploading a file and firing my save event (converts to byte[] an stored as SQL Server BLOB) it does everything is supposed to do.
Same problem here
There is a good suggestion of using the AJAX AsyncUpload control however I am a firm believer of removing the cause and not treating the problem. I will continue down this route to best my understanding of asp.net etc.
Would there be a wizrd amongst you that could help me identify why I get "object ref not set to inst of obj" error on first postback but on second it works fine. Content page has a master page which wraps content page in an update panel. Not my decision to do this. There is also an update panel with postback triggers targeting my save event.
What are your thoughts people?
The problem (as seen here http://forums.asp.net/t/1060363.aspx) seems to be when you use the visibility property on the surrounding panel (as it seems you are from the linked question).
The suggested workaround is to use CSS visibility instead so use this to make it invisible -
<asp:Panel ID="pnlUpload" runat="server" class="workerDetailsPanelLeft" style="display:none">
The explanation for this from the thread is
If your container is set to invisible, the upload control is not actually rendered as HTML, causing the form's enctype not to be set to enctype="multipart/form-data", causing the file upload control not to post the selected file back to the server. The workaround is either to make sure the FileUpload control is rendered to HTML (by setting its style to display:none in stead of Visible=false), or by manually setting the enctype
So another workaround would be to alter your form tag to this
<form id="form1" enctype="multipart/form-data" runat="server">
I think either one of those should solve your problem.
You could do the same thing via code on Page_Load event.. Enter this code and it will solve the issue.
Page.Form.Attributes.Add("enctype", "multipart/form-data");
I am working with a website that has javascript that does some changes on the page load. However, when I load the page and handle the DocumentCompleted event, this change isn't there. If I then continue paste the DocumentCompleted event, I can see the change happen. However I need this change to happen during DocumentCompleted so I can check some things.
Is there an other event I can subscribe to, or a way to cause the webBrowser to do all the javscript on page?
Edit: This is what I am talking about.
I loaded a sample page just to show you, and clicked the submit button with all fields empty to generate an the error.
Here is the result:
http://s8.postimage.org/zfv6stcar/sfsdfsdfds.jpg
Now if I take the HTML at that precise moment from that WebBrowser control, and render it somewhere else, those errors go away. The same thing happens when the server sends back those errors. If I handle the DocumentCompleted event and take the html, it isnt there. But after the event, it shows up in the control.
Hope you understand, it's hard to explain.
The problem seems to be that the DocumentCompleted event is being fired before the javascript. You should do some reading on how client side/server side things function.
One option is to make a separate method for the DocumentCompleted event and call it form the javascript after it has been completed. This would get the sequencing of these events working properly, but is not very ideal.
Alternatively, you could call the javascript code at the beginning of your DocumentCompleted event. The link below gives a pretty good explanation of how to go about that.
http://forums.asp.net/t/1117189.aspx/1
Personally, I would avoid using javascript and do the validation on the client side .NET, but I don't know enough about the website to really say.
EDIT:
This should be the script you are looking for. Alternatively here is a thread related to your issue. Sorry I don't have the exact code as I don't have a project to test this on.
http://msdn.microsoft.com/en-us/library/system.web.ui.clientscriptmanager.registerstartupscript.aspx
Calling JavaScript Function From CodeBehind
RE-EDIT:
What is happening on the link you provided in the comments, is that each textbox is calling some javascript as well as the submit button. The best way to examine this is using the "Inspect Element" in the right-click menu on Google Chrome. For example, doing this on the textbox would show that it is registered with a few events:
onfocus="$('f_tip_Username').style.display = 'inline'"
onblur="$('f_tip_Username').style.display = 'none'"
onchange="$('f_err_Username').style.display = 'none'"
The first the element with the ID 'f_tip_Username', sets the display style of that element to inline (visible).
The submit button calls the following:
onclick="return o_edit_profile_form.validate()"
Doing a find on "o_edit_profile_form" in the source code, you can find the exact javascript location that is being called. Enjoy!
FINAL EDIT (hopefully?):
Follow these steps: go to your site, right click and go view source. Do a find for "f_tip_Username". This is the ID of one of the div tags being used. The third entry of it, should be a "div tag" that is used under the first textbox to warn of "min 3 characters".
You'll notice above that in the code is a input type "text" with the Name "Username". Notice the three events it has registered in it:
onfocus="$('f_tip_Username').style.display = 'inline'"
onblur="$('f_tip_Username').style.display = 'none'"
onchange="$('f_err_Username').style.display = 'none'"
These either hide or make visible, the div tag we found (f_tip_username) and also a separate div tag (f_err_Username) which is the error message div tag. Let me know if you are not able to find these in the source. Follow the steps I provided and you will find it in the "view source" OR in the DocumentText.
I have a pretty simple web-form set up in .Net where I am leveraging jQuery for some of the functionality. I am using the DOMWindow portion for part of the presentation layer.
There is a login form in a div that is set to display:none. When a user clicks a button on the page, it displays the login form. However the .Net button for the login form will not fire it's event when display is set to none. If i take this out, it fires fine. I have also tried using the visibility attribute, but no luck.
the div code is:
<div id="Login" style="display:none;">
The launching code is:
click here to login.<br />
the jQuery code is:
function LaunchLoginWindow() {
$(document).append("#Login");
$.openDOMWindow({
loader: 1,
loaderImagePath: 'animationProcessing.gif',
loaderHeight: 7,
loaderWidth: 8,
windowSourceID: '#Login'
});
}
Any help or explanation that anyone can offer is appreciated.
I noticed i had some code in there defining a client-side function on the Login div. I removed this so as to eliminate it as a possible issue.
I can see in your code that you are appending the div #Login but not setting its style property back to normal like block so. Set it back to block and i am sure it will work
try adding somthing like:
$(document).append("#Login").show();
OK, after playing around with this using firebug, I found the issue: When the jQuery plug-in DOMWindow creates its display layer, it appends to the HTML node of the DOM, which places the control outside the asp.net form tag. Therefore the button and actions associated with it via the DOMWindow are not recognized by .Net. So i edited the DOMWindow source file to append to the DOM form node rather then the html node.
The drawback is that the source has now been customized and will have to be QA'd thoroughly, especially if any further changes are made. But I hope to manage this effectively via commenting in the file.
Hope this helps anyone else who hits this issue.
pbr