I don't think TC have that ability to do that as it is kind of Custom requirement.
What you can do is, Instead of doing this thru TestItem you can create a Custom function that controls this execution.
'TestItem1 - Test1 - Set to execute 3 Iterations
'TestItem2 - Test2 - Set to execute 3 Iterations
Sub Test1
Log.Message "Pass"
Log.Error "Fail"
Log.Checkpoint "Success"
End Sub
Sub Test2
Log.Message "Pass"
End Sub
From the above two routines, when error occurs it should re-run the TestItem1 from the beginning instead of executing the
Log.Checkpoint "Success" statement
shankar_r Re-run should happen only for the given set of iterations set from 'Count' setting in the Test Items window
I don't think TC have that ability to do that as it is kind of Custom requirement.
What you can do is, Instead of doing this thru TestItem you can create a Custom function that controls this execution.
I had the same problem and I can present my solution:
All my testcases are defined in the testitems- view in the project. Additonaly I have created two additional Folders: "Retest1" and "Retest2". The tests are executed with "Testexecute".
I have created a powershell- script, that makes a checkout from GIT and starts the tesexecution. So far so good.
In a second step i have modified the event handlers. Whenever an error in a testcase is raised I set a value, that the testcase has failed. If it is passed no flag is setted. This flag is written to a csv- file, like: "TCSName","Result".
After the testrun is completed the powershell script analyses the csv. Each testcase that has the flag "failed" is copied into the ".mds" - file in the tag "Retest1". After all items are identified the exectuion will be restarted, but only the testitem "Retest1".
I don't know if this approach is the best, but it works very well for us.
I execute my tests with testexecute and not
