The first attachment is a has a test case where issuing a command with variable substitution results in no blue boxes but the RM is found as demonstrated by right-hand-mouse open RM from the response. Issuing the command explicitly does give the blue boxes. The second attachment is of a case that works both ways. I can't see the difference between these. Any ideas?
Greg
This is, arguably, a bug in iTest. Our test case editor is being "smart" because at design time when you select a step in the test case, it tries to automatically remap the most recent results of that step using the latest response map that is appropriate for the step. In this case, though, it is using the command as it appears in the test case step to find the appropriate response map -- which doesn't match the applicability data for the correct map. Only after substitution is performed will the correct response map applicability match.
We should be using the meta-data from the last execution to select the appropriate response map to use at design time when we can't find one. This also addresses another problem we have -- which happens when the session profile associated with a step cannot be determined at design time (because the session property contains substitutions, or it depends on the default session type from the procedure settings).
We'll take a look. In the meantime, a workaround is to explicitly set the appropriate response map on the step -- at least for the duration of adding analysis rules, etc. to the step.
