Allow including lemmas dynamically into APRProver #4681
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…a functional rule
…s clause translation for functional rule
…ication rules to konvert
…unctional rules to translate
… tests to working version
…n concrete/symbolic attribute too
| @@ -198,45 +218,105 @@ def _inject(definition: KDefinition, term: KInner, sort: KSort) -> KInner: | |||
|
|
|||
| # 'krule' should have sorts on variables | |||
| def krule_to_kore(definition: KDefinition, krule: KRule) -> Axiom: | |||
|
|
|||
tothtamas28
approved these changes
Nov 14, 2024
| @@ -102,15 +102,15 @@ def fun(term: KInner) -> KInner: | |||
| return fun | |||
|
|
|||
|
|
|||
| def bool_to_ml_pred(kast: KInner) -> KInner: | |||
| def bool_to_ml_pred(kast: KInner, sort: str | KSort = GENERATED_TOP_CELL) -> KInner: | |||
| def _bool_constraint_to_ml(_kast: KInner) -> KInner: | |||
| if _kast == TRUE: | |||
| return mlTop() | |||
There was a problem hiding this comment.
Suggested change
| return mlTop() | |
| return mlTop(sort=sort) |
| @cached_property | ||
| def function_labels(self) -> tuple[str, ...]: | ||
| """Returns the label names of all the `KProduction` which are function symbols for all modules in this definition.""" | ||
| return tuple(func.klabel.name for func in self.functions if func.klabel is not None) |
There was a problem hiding this comment.
I don't think there's a function symbol without a klabel, consider asserting instead.
|
|
||
| axiom_vars: tuple[SortVar, ...] = () | ||
| kore_axiom: Pattern | ||
| if not is_functional: |
There was a problem hiding this comment.
Consider extracting two helpers: one for the functional case, one for a normal rewrite.
| @@ -249,6 +329,25 @@ def kflatmodule_to_kore(definition: KDefinition, kflatmodule: KFlatModule) -> Mo | |||
| return Module(name=kflatmodule.name, sentences=(imports + kore_axioms)) | |||
|
|
|||
|
|
|||
| def _krule_att_to_kore(att_entry: AttEntry, kast_rule_vars: dict[str, list[KVariable]]) -> App: | |||
There was a problem hiding this comment.
There's similar logic in _module_to_kore, maybe it's possible to extract that into it's own module and reuse it here.
| @@ -255,19 +255,19 @@ def prove_claim( | |||
| depth=depth, | |||
| ) | |||
|
|
|||
| def get_claim_modules( | |||
| def parse_modules( | |||
There was a problem hiding this comment.
Is the module parser exposed by kprove general enough to parse an arbitrary module?
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
In order to include lemmas dynamically in the Kontrol prover, we need an option for passing the extra module into the
APRProverso that it can be fed into the booster'sadd-moduleendpoint. This also requires being able to translate functional rules to Kore in pyk usingkrule_to_kore. This PR:extra_moduleparameter to theAPRProverinitialization, which allows passing a new module with lemmas into the prover at initialization time. It will pass this module on to theadd-moduleendpoint and make sure it's used on all future requests.symbolicandsmt-lemmaattributes to the list of attributes recognized by pyk.KProve.get_claims_modulestoKProve.parse_modules, and renames some of its parameters.bool_to_ml_predfor converting boolean predicates to matching logic ones._krule_to_koreto make it support translating some functional/simplification rules to Kore which were not supported before (and adds tests):_krule_att_to_koremethod for converting attributes ofKRuleto kore.krule_to_kore(eg. handleowise => priority(200)andsimplificationvspriority).axiomto differentiate between semantic and functional rules, and to convert functional rules as matching-logicimpliesaxioms.