Rerun failed test once

Nothing built in because the definition of "failure" is variable from user to user.  That's something that would be handled most likely from a Test Management software application.  For that matter, you would have to have a way of classifying types of failures to re-run.  Timing issues you would... issues where a checkpoint failed you might not.  That's something that would have to be user defined in some sort of code.

ok so let's say that this is what i want



if (test script that just ran reported an error (errcount>0)) and Run_count < 2 then

rerun the test

else continue on with the rest of the suite.





This is the basic pseudocode, but i'm not sure were i would put it. Is this a valid set of code to go in an "onEvent" method?

What test?  How is the test written?  Is it a granular test in a Keyword or single Script code function?  Or is it some sort of combination of multiple functions?  Are you using a data driven/table driven framework to define your tests or is everything procedural based?



All these questions add varying degrees to what you're asking.  You could call it in an event, but you would have to have some way of passing to the event, either through a global variable or something, to indicate what you defines as "the test" that failed.  Depending upon the architecture of what you've constructed for your automation, an "OnErrorEvent" might not be the best place.  For that matter, is a OnLogWarning, a valid place for such a thing?  is a warning a failure?  Or are errors the only "failure" condition?



As for run count, you would have to track that somewhere, too, because that's not a native item for many different kinds of test construction in TestComplete.  



So.,... my best answer... it depends.  Again, there's nothing "native" to TestComplete to do that, it's something you would have to construct within your particular automation framework/architecture.



In my case, the framework I've constructed, each "test" is read in from a set of CSV files as a list of steps.  I could wrap my iterative code in such a way that, if the any step in that list returns a "failed" status (basically a boolean "false" if a function fails to complete with a pre-determined indicator of success), to go back to the beginning of the loop and try again.  It would all be code and it would be a bit complex to make sure it runs in the framework so yes, it's possible to do what you're asking, but it all depends...