Recent changes to bugs

[](Fluent) is not working well

Leandro Nahabedian — Tue, 05 May 2015 03:26:06 -0000

When we make set a as a safety goal. The machine obtained is the same when the Fluent is initially 0 or 1. They should be different machines.

For example:

Fluent OnInitiallyFalse = <on,off> initially 0
Fluent OnInitiallyTrue = <on,off> initially 1

Both machines are the same.

Pessimistic is distributed if applied to a merge

German Sibay — Mon, 11 May 2009 13:48:08 -0000

pessimistic C = (A ++ B).
The tool is doing pessimistic A, then pessimistic B and after that trying to merge.
It should perform the merge and then get the pessimistic model.

Minimal - Bug and inconsistency

German Sibay — Wed, 01 Apr 2009 13:44:42 -0000

Check the attached file.
minimal ||It2_PostMin = (It1_Resulting_MTS).
minimal ||It2Min = (It1).
It1 = (OrderFood ++ UpdateStock).

It1_Resulting_MTS is just the result of the merge in It1 (copy and pasted from the Transition window). So technically they are the same.
It2_PostMin is minimised and the result is verified as correct. However It2Min is minimised but the minimised and the original are not comparable (they have a different alphabet). They should be comparable, and also, It2Min and It2_PostMin should lead to the same resulting MTS (or both raise an exception).

Cannot see model names in refinement dialog box

Sebastian Uchitel — Fri, 27 Mar 2009 17:08:46 -0000

On my mac, if you go to the refiement option in MTS menu, the dialog box that appears which lets you pick to models wich are to be compared uses drop down menus which only let me see the first few letters of the model names.

Invalid result with strong refinement

Sebastian Uchitel — Fri, 27 Mar 2009 17:06:36 -0000

Checking strong refinement, the tool does not abort if the models have different alphabets. Should say: Not applicable.

Index out of bound exception when minimising

German Sibay — Fri, 27 Feb 2009 12:19:03 -0000

When merging and then minimising like "minimal ||Itl_Aux1_P3 = (Itl_Aux1_Env ++ C_P3)." an IndexOutOfBoundException is thrown.
However if using minimal on the pre calculated merge there is no such exception.
See attached file.

Combining minimal and optimistic has no clear meaning

German Sibay — Tue, 17 Feb 2009 12:35:55 -0000

Combining minimal and optimistic behaves differently when applied to a process than when applied to a composition.
When used on a simple MTS processes the optimistic version is first calculated and then minimised (LTS minimisation).
When used on a composite MTS the MTS minimisation is used (and the optimistic is not calculated) (MTS minimisation).

For example:

minimal optimistic MTS1 = ... // (FSP process definition). Calculates the optimistic MTS and then minimises it (LTS minimisation)
minimal optimistic ||MTS2 = (MTS1). // minimises MTS1 (MTS minimisation). No optimistic version is calculated.
minimal optimistic ||MTS3 = (MTS1 ++ AnotherMTS). // minimises the merged MTS (MTS minimisation). No optimistic version is calculated.

Safety checks: unexplained requirement

Sebastian Uchitel — Mon, 16 Feb 2009 16:53:00 -0000

Why does there have to be a deadlock free implementation to do a safety checks? (Note that this is the check against an observer, it is not referring to checks against FLTL safety properties)

Suspect behaviour of minimization

Sebastian Uchitel — Mon, 16 Feb 2009 16:50:10 -0000

With all the changes regarding refinement, it would only be natural to have problems with minimization.
We should have at least a sanity check at the end of it where the resulting minimized MTS is checked for equivalence against the original.

Weird code in checkSafety()

Sebastian Uchitel — Mon, 16 Feb 2009 16:44:56 -0000

WHy does it say "there must be a property to check"