MarkHaysOccasional ContributorJoined 5 years ago25 Posts16 LikesLikes received1 SolutionView All Badges
ContributionsMost RecentMost LikesSolutionsRe: Difference in JavaScript support with TC 15.61 Started copying the Sales person on the requests for updates to the open ticket. Would go back every couple of weeks with another email requesting an update and complaining about lack of support. Sales did eventually get someone else involved, and a few months later finally got the response there is nothing they can do, and that was when they mentioned the Manifest version. I did have some other open issues as well, that were not getting addressed in a timely manner, so those were included in my complaints as well. Re: Difference in JavaScript support with TC 15.61 Sounds Similar to some of the issues we have experienced as well. In version 15.57 we were able to make a call using contentDocument.Eval() and it worked fine. Versions after that it do not work. After many months and asking for feedback on this issue, TestComplete Support finally, said that the eval function was deprecated by Chrome/Javascript. It is my understanding this was related to Chrome moving to Manifest version 2. So, the newer versions of TestComplete work with Manifest version 2. Older versions of TestComplete worked with manifest version 1 which allowed the call to the eval function. We are currently exploring an option of attempting to call the functions directly (and bypassing the eval function) by using contentDocument.Script.functionname(), but that is not working in version 15.64 of TestComplete. Re: Global Items in TestComplete Yes, makes sense. We have a lot of common Library files that should be in 1 place instead of for each project. Re: Unable to load content document in Object Browser We use the Script.eval functionality of contentDocument, and experienced the same issue where it stopped working in version 15.58 (and also 15.59). I opened a case over a week ago. Perhaps open a case with them as well. Re: Changing xpath trying to wildcard it possible? As indicated you cannot use a dynamic number in the namemap. You could use it in code and make a call to FindElement. However, it appears as if the data you are looking for is in a table that can change. Additionally any xpath with something such as //td/th[64] means that it is looking for the 64th instance of that object within the HTML. In my opinion, this xpath is very susceptible to changes (especially if it is in a table). I would not map objects in a table, but rather write code that can be used to find the items that you are looking for. Re: Bug with the "improve performance" message ? This feature has never worked for me since it was first introduced many months ago. If I have 2 options in my selectors (XPaths) and switch to what is suggested, I get the same or an opposite error. Here is a link to a thread that discusses this and also a work-around about how to suppress the messages: https://community.smartbear.com/t5/TestComplete-Questions/Remove-the-quot-improve-your-test-performance-quot-in-test-log/m-p/236940 IMO this feature has never worked and there should be an option to disable the checking for the performance improvement all together rather than test Complete to spend time identifying an improvement that is not correct and then having to ignore/suppress the messages to the log file. Re: Repeat Exists (Avoid Writing Log Errors) Python I think the call to FindElement is probably the one that is logging the error. You can use other alternatives such as FindElements and then get a list of elements (list would contain nothing in this case) and go from there. I do agree that when FindElement cannot find something it logging an error is problematic, and it would be good if they had an optional parameter to log an error or not. Re: Test compete does not communicating with Chrome. It might be related to the issue with https://community.smartbear.com/t5/TestComplete-Questions/TestComplete-issues-with-Chrome-113/m-p/249733#M108752 Re: Problems with TestComplete 15 and Chrome 113 We have been encountering Memory Issues since late January. With specific tests the memory issues happen probably about 80% of the time. Other tests they do not happen at all. The symptoms we see is that at a certain point the memory being used by Chrome increases rather rapidly until it crashes the system (or we kill Chrome). We currently have an open case, but have not heard much. And specifically related to this thread, I tried to see which version of Chrome was installed and as soon as I went into About Chrome I saw that it was 112, but it immediately started updating. Re: Chrome updated, now TC can't find browser I am getting similar issues with Chrome 113 as well.