I have a problem. While parsing a page, a pop-up window appears and blocks access to html. I tried to click on the pop-up, but it re-directs me to a new page, and I have no clue how to get to the upper right area of the pop-up to close it. My question is how to determine the coordinates of that element? None of the related issues raised here at Stackoverflow helped me find a solutions, so assistance would be very much appreciated.
Why do you need the actual coordinates? That seems like a bad idea given it could move or have different results on different screen sizes (responsive designs). You'd be better off finding the class / id of the close button and calling browser.button(class: 'close').click or browser.link(class: 'close').click.
If you need help determining the class of the close button, open your browser and enable developer tools / console. They all have an inspector tool now to select the element and find all classes & ids.
In the unlikely event there are no classes or ids it may have a data attribute you could work off.
If the pop-up is a regular browser window you can close it by index. Your main window has index 0. So the pop-up has index 1.
if browser.window(:index => 1).exists?
browser.window(:index => 1).close
end
Related
I am no programmar nor tech savvy person myself.
I just wanna know if it's possible to achieve this goal:
The default behavior of the browser(e.g. chrome) is it sends one page up or page down depending on where I click(above or below scroll thumb).
But when clicking somewhere on the scroll track, I wish it navigates right through the target position in the web page.
for better understanding, here's the screenshot of cnn.com main page for reference:
(In first screnshot, I highlighted where I will click.
Second screenshot is the default behavior of the browser as a result.
Third screenshot is what I want realize for my own convenience. it's useful when navigating through long articles.)
I think it is almost impossible to do this on my part because maybe it's hardcoded in browser's engine level or something, but I wanted to make sure it really is.
or any workaround like making autohotkey script or something?
thank you so much!
There seems to exist a trick with Shift, but it doesn't work in browsers.
However, you are manipulating a browser, and browsers have a built-in scripting language — JavaScript. Here's the setup you need to do:
Go to the place on the page you want to scroll to.
Open the developer console (Ctrl+Shift+I and then click on the "Console" tab). This shouldn't interfere with the scroll position.
Type window.scrollY and press Enter.
After that, the console will output the amount of vertical scrolling. In the following AutoHotkey code I'll use 12345 as an example; replace it with the value you have. Note that it also scrolls to the left margin.
Send {F6}javascript:window.scrollTo(0,12345)`%3Bvoid`%200{Enter}
It appears that, due to a bug, this doesn't work in Firefox.
I know your problem, I think chrome must have this config, you can type in the address bar "chrome://chrome-urls" for searching
I am running a test on a web app using Ruby-Watir-Rspec. It is very simple since I'm a beginner.
I open a form and enter the required information, but the "Create" button is not in the visible area, so I get the message:
Failure/Error: #browser.button(class: xxx).click
Watir::Exception::UnknownObjectException:
element located, but timed out after 30 seconds, waiting for #<Watir::Button: located: true; {:class=>xxx, :tag_name=>"button"}> to be present
Caused by:
# Selenium::WebDriver::Error::ElementNotVisibleError:
# element not interactable
If I scroll while the script is running, it clicks on the button and the test is successful.
I tried scroll.to, wait_until_present, scroll to coordinates, scroll_into_view, none of them worked.
The only way to make it work was to put " #browser.send_keys :tab" several times until it reaches the button at the bottom of the form.
I believe the problem is the button being inside the form which does not take the entire page (behind the form is the map so that part of the page doesn't have the scrolling option)...so is there some way to scroll inside the form? Or do you know some other approach to finding this button? Any hint is appreciated.
Btw, the page is maximized.
Here is the code snippet, just simple:
it 'should create the place' do
#browser.button(class: xxx).click
end
My guess is that this is a custom scrollable element that hides the content with the overflow: hidden style. Elements in the overflow are not considered visible/present. When you manually scroll, you're bringing the element out of the overflow so that it's present.
I've seen a couple of these in the past. Each one needed a different approach for scrolling. Without the exact HTML/CSS, it's hard to say how to scroll the element.
However, if you're not trying to test the scrolling, you could manually fire the click event. This will bypass the visibility requirements:
#browser.button(class: xxx).click!
Try using the Watir Scroll gem: https://github.com/p0deje/watir-scroll and scrolling the element to the center of the viewport: button.scroll.to :center.
You can also submit the form directly #browser.form.submit
In the image above you can see two notifications on OS X. The first one is from my app and the second is from Apple's Reminders.app. In the image you can see the otherButtonTitle 'Complete' and the actionButtonTitle 'Later'.
The second notification, i.e. the one from Reminders.app behaves quite differently. It gets this little arrow pointing downwards on mouse over indicating that there are more actions when clicked. And indeed, you just need to click once on 'Later' and it will give you a couple more options to choose from.
However, I can't get the same behavior to work for my notification. I don't get the little arrow on mouse over and I don't get more options displayed from a single click on 'Later' (notification just gets dismissed). More options only get displayed when holding down the mouse button on 'Later' which is not obvious.
Am I missing something obvious here? How can I get my notification to have exactly the same as the ones from the Reminders.app?
While trying to find a solution for the same problem I found this nice explanation for the NSUserNotificationPrivate class that explains how the Reminders app does it.
https://github.com/indragiek/NSUserNotificationPrivate
If the notification type is set to "Alert", the alternateActionButtonTitles property lets you set an array of additional menu item titles to be shown in an action menu that can be accessed by hovering on the Action button and clicking on the arrow.
Once a notification is handled, the index of the action can be retrieved using the _alternateActionIndex property.
So they are using a private API. As the site's disclaimer say using any of this will result in your app being rejected from the MAS and potentially breaking if the APIs change.
I've added some code as found here Big problems with MFC/WinAPI to colour tab titles the same as the reset of the dialog, which works, but unfortunately all the tabs end up with the same name. This doesn't surprise me all that much as GetCurSel() is used to grab the text to use, and only one tab can currently be selected, but I'm struggling to see how you access the correct tab index from OnDrawItem().
I've googled and had a look on MSDN but don't see how anything passed to OnDrawItem lets you know which tab is currently being drawn, rather all the examples I've seen assume you're only interested in the one currently selected. All I want to do is something along the lines of GetWindowText() on the child window and redraw with that. I'm also unsure of the parent/child/sibling relationship between the sheet, tab control and page - it depends who you listen to.
I should probably add that I'm also unsure why all the tabs are redrawn when I select one. I don't know if this is normal or something specific to this implementation (that's something I'm looking at, but like seemingly everything else in this code base it's multiply inherited several times over ...).
Cheers for any help.
Not to worry, I now realise lpDrawItemStruct->itemID holds the tab index so I can get a handle to the tab using that.
i am using code right out of the examples in yahoo maps api, but when i click on a smart window i get the close button (x) in the upper left. the only thing i am changing is the html in the smart window but i dont understand why that would affect the location of the close button. also, i dont see any documentation on how to change this on the yahoo maps api web site.
You may want to look in your page CSS for anything that floats img tags left or positions them absolutely (think along those lines).
One debugging method would be to temporarily turn off your page's CSS to see the result.