| Summary: | Array not unfolded in strategy when using statevars and pointvars | ||
|---|---|---|---|
| Product: | UPPAAL Stratego | Reporter: | Martin Kristjansen <mk> |
| Component: | Engine | Assignee: | Peter Gjøl Jensen <pgj> |
| Status: | NEW --- | ||
| Severity: | major | ||
| Priority: | P5 | ||
| Version: | unspecified | ||
| Hardware: | PC | ||
| OS: | Linux | ||
| Architecture: | |||
| Attachments: | Model using pointvars and statevars | ||
Hi Peter, Can I assign this report to you? |
Created attachment 343 [details] Model using pointvars and statevars This by using the Strategy-8-beta3 branch. I make use of pointvars and statevars in the strategy synthesis. I have an array called T_learn. This is a double array with 11 elements. The idea is that I can write {statevars}->{T_learn} as part of the strategy synthesis query, such that the array T_learn is unfolded behind the scenes. Otherwise, I should write {T_learn[0],...,T_learn[10]}. However, the resulting strategy only has 1 pointvar instead of 11 pointvars. Whether or not it affects the strategy learned I cannot tell. But if I unfold the array myself in the query, I get a strategy with different behaviour. Model is attached with both queries.