The first 2 commits cover analysis notes on invariants and one (at John T's request) capturing initial concerns about the wider system of interest. The 5th Oct commits add some additional tests, fix a big in AEO relating to intra job invariants and fix the order of job definitions so that the source extra job invariant is defined before the users.
6th Oct commits pulled in the latest p2j, started a new (and incomplete) regression script for testing planned failure conditions, updated KitchenSinkJob2 (KSJ2) so that it reuses all of the events from KSJ1 (which makes the test tougher), added a test for break as it was otherwise only present in the KSJ definitions and, finally, fixed a bug in HappyJob.AssessingConstraints where merge count was being checked even if the merge event was not seen at runtime.
The first 2 commits cover analysis notes on invariants and one (at John T's request) capturing initial concerns about the wider system of interest. The 5th Oct commits add some additional tests, fix a big in AEO relating to intra job invariants and fix the order of job definitions so that the source extra job invariant is defined before the users.
6th Oct commits pulled in the latest p2j, started a new (and incomplete) regression script for testing planned failure conditions, updated KitchenSinkJob2 (KSJ2) so that it reuses all of the events from KSJ1 (which makes the test tougher), added a test for break as it was otherwise only present in the KSJ definitions and, finally, fixed a bug in HappyJob.AssessingConstraints where merge count was being checked even if the merge event was not seen at runtime.