Closed antoinepouille closed 9 months ago
Does this change anything besides adding the "lesser than" testing capacity?
I maintain a parser/analyzer at https://github.com/hmedina/KaSaAn which I'll need to update, so I'm just wondering if there's more to it or just this.
Does this change anything besides adding the "lesser than" testing capacity?
I maintain a parser/analyzer at https://github.com/hmedina/KaSaAn which I'll need to update, so I'm just wondering if there's more to it or just this.
Hey :-)
Not sure how you manage counters there. If there are no <=
tests in the model, everything will behave the same.
It seems like KaSaAn is based on parsing snapshots ?
As for now <=
is syntaxic sugar for adding a new counter, a new counter ending in __inverted
will appear in the snapshots if <=
is used.
It is likely we add a new logic that will remove this new counter for the rule condition to work.
Does that answer your question?
Hey there!
Ok, if snapshots won't contain this new construct, only rules, then I don't need to update my parsing yet, as the "real" thing in the snapshot should parse fine. Will contact maps have the rule's definition (<=
), or the generated new agent (__inverted
)?
In this PR, after the parsing of the .ka
, the <=
are gone and replaced by the use of the __inverted
agent, which will then show up everywhere, including the contact maps.
Everything should work fine on your end.
Hello @hmedina , after discussion with @feret , we actually are hiding these counters now in snapshots since this recent commit
Sorry for the worry about this!
Allows
<=
tests in a agent counter.Using
<=
will enable in parallel a hidden inverted counter where the>=
test will be made. Counter logic is planned to be revamped to allow this naturally.Example of use:
Here, a counter
c__inverted
is added onto agentA
, which manages the<=
tests.