Reasoning modes fixes - #366
Conversation
kstrauch94
commented
Jul 29, 2026
kstrauch94
commented
- Fix various bugs with reasoning modes in the propagator
| # stage 2: get reasoning mode atoms for propagator | ||
| # stage 3: report results on first model found here | ||
| self.reasoning_mode_stage_lits: dict[Literal[1, 2], int] = {1: -1, 2: -1} | ||
| self.reasoning_stage: Literal[0, 1, 2] = 0 |
There was a problem hiding this comment.
Is self.reasoning_stage used anywhere? If yes, is it intended that it's declared as 0,1,2 while the comment above as well as self.optimization_stage_lits talk about stages 1,2,3?
There was a problem hiding this comment.
It is not used. It is probably some old code. I deleted it! Thanks for spotting it.
There was a problem hiding this comment.
Is it intentional that this nogood isn't locked?
There was a problem hiding this comment.
I guess there is no intention behind this decision. I will add the lock since it makes sense to have it. I will push it and merge if all tests pass.
There was a problem hiding this comment.
The performance test failed initially (6.04sec > 6sec). It isn't related to this PR, it's just that the test threshold is too tight. I usually just rerun the failed job and with a bit of luck the time goes under 6sec.
2 participants
