Ranorex Forum

When working with Internet Explorer 11 using Ranorex 5.0, I sometimes see WebDocuments (both in Spy and when tests are running) that have Domain as null or containing an empty string. These WebDocuments later have their state change to invalid, and eventually disappear. There may be multiple of these at any time.

My guess is that objects are being cloned and garbage collected. I haven’t seen this behaviour with other browsers, although do spend the majority of my time working with IE, so am not 100% sure that other browsers do not do the same. Are there any plans to have the Ranorex libraries filter out WebDocuments that have State=invalid or that have content that is clearly unexpected, such as a null Domain?

Thanks,

John H.

Hi John,

Unfortunately I cannot see this behavior on my machine. Can you please update Ranorex to the current version and check if this can be reproduced with this version.
Do you also have any problems when executing the test?
Can you reproduce the issue also with other applications or only with your application under test.

Regards,
Bernhard

Hi Bernhard,

I am now using Ranorex 5.1.1.18898 with Internet Explorer 11.0.9600.17207 on 64 bit Windows 7 (although I was using Ranorex 5.0 when I first saw this).

Our Microsoft liaison person has confirmed that IE has this behaviour. However, I would have expected Ranorex libraries to filter out these invalid objects.

The problem is seen both using Ranorex Spy and during running of tests, although only if RxPaths do not include a check for a non-null/non-empty domain.

Thanks,

John H.

Hi John,

I am not exactly sure why this is a problem during the test run because I don't know how the path to the dom in your repository looks like. If the path is unique this should not be a problem.
I suggest to continue the communication via email because this seems to be a feature request.
Is it possible to send a Ranorex Snapshot file and your repository to [email protected].
Thank you.

Regards,
Bernhard