Testcomplete doesn't recognize browser objects, but recognizes non-browser objects.
When I use the spy tool TestComplete can only recognize the whole browser page, I tried on chrome, edge and mozilla. However when I switch to a virtual machine it works properly and recognizes objects. Is there a way to fix that on my laptop, because I'd rather not switch to the virtual machine?
Are your browser versions the same on your laptop and virtual machine?
The latest Chrome, Edge browsers have been causing issues for the ObjectSpy.
Be careful here in case you do have older browser version son your VM versus a newer version on your laptop. Checking the About in the browser auto updates and I do not want you to lose your VM working browsers.
Are your browser versions the same on your laptop and virtual machine?
The latest Chrome, Edge browsers have been causing issues for the ObjectSpy.
Be careful here in case you do have older browser version son your VM versus a newer version on your laptop. Checking the About in the browser auto updates and I do not want you to lose your VM working browsers.
this is giving me fits. I was able to spy (get the red box around) a couple of objects in my browser by doing the Windows Key+P and having everything just on my laptop screen instead of multi-monitors but even that isn't working now. Spy is putting the red box around the entire browser page rather than just the element over which my mouse is hovering. I have the latest version of chrome and the latest version of testcomplete. what else can i try?
Hi, in the object browser, check if chrome appears as "browser(chrome)" or "process(chrome)". When the browser is in process mode, you can't access to objects. To force the "browser" mode, ensure that the browser is part of your tested apps, and sometimes, it can help to add chrome in the process filter. I don't know why, cause it's a total nonsense, but it works. (Current Project properties > open applications > process filter).
because i created this project suite as a browser based application it did not even put a TestedApps module in the project but chrome is still identified as a browser in the Object Browser.
if i were to add a TestedApps module what do you suggest i select from this screen when prompted?
my application is not a mobile app or anything of the others displayed on this list.
i did add Chrome to the process filter as you recommended
I will let you know if that improves anything about object recognition. thanks!
When I add an app to the TestedApps module I always choose Generic App for any desktop application.
When the browser is identified as "Browser(name_of_browser)" you can expand the object tree, and you should see an object with : Page(some_url). And if you expand this object you can see what TC sees.
One more thing, TC cannot handle chrome's internal pages (new tab, settings, flags etc...).
If you are launching the browser via TestedApps, you might need to include the command line parameter to include your browser profile. Otherwise the browser extensions will be excluded.
do i have those processes running? yes, although they look different on my task manager that what your screenshot shows
is the smartbear extension installed? Yes. it looks like this (Version 14.0.0).
I'm launching the browser with Run Browser in the Item column and Chrome in the Operation column and my applications credentials URL in the Value column. i did as eykxas suggested and added chrome as a generic tested app. however, when i expand the object tree for a very brief time i can start to expand the Page (some url) entry as shown in this screenshot
but a few seconds later it vanishes down to just this
and all i'm left with is the BrowserWindow(0) in the tree.
i don't know if i mentioned this before but I began having to "fix" these automation scripts because all of a sudden all of the originally recorded ClickItem operations related to selecting something from a drop down box began failing and the only solution I could come up with was to replace them with the Keys operation with a value that reflected the first letter of the desired selection. of course, if there were multiple sections starting with the same letter my keys would have to look something like this:
Keys("SSS")
to get to the third selection starting with "S". In one case where i needed to select a prior to the current year (which appeared as the default value in the drop down) I had to use Keys("[Up]") because trying to enter Keys("22222.....") would have just gotten me eventually to the year 3000 before looping back to 2000 and going up again. As i was trying to fix those i started seeing a bunch of other object mappings that were no longer recognized from the original keyword test recordings
so if anyone know anything about why that might have started happening several weeks ago (and what would make ClickItem work again) that might be a better solution to my issue. Anyone?
I've never seen AppRoot(0) in the tree item of Browser("chrome") before. If you were to create a project, navigate to https://www.w3schools.com/csS/css3_buttons.asp and then use the Object Spy on the buttons, is TC able to identify each of the buttons?
A version of TC will work with a version of a Browser. If the Browser was updated, then that version of TC may fail to work. The same applies with all technologies.
You should find out if your web application, the technology it uses, has been updated? And if it uses Angular, then check your web application meets the requirements of Supported AngularUI Controls
If it hasn't been updated, then it could be an issue with the Browser which TC doesn't support!
