Repository navigation
Agreement test set: the reference interpreter's outcomes for the forbid/require/permit subset - #85
Merged
Merged
Conversation
…id/require/permit subset scripts/gen_agreement_corpus.py runs each program in agreement_corpus.txt through the 0.18.4 reference interpreter (--quiet --test) and records exit, outcome, crossed rules with their because text, and stdout. The TypeScript and Swift agreement evaluators are held to this file. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Qvgf1CGxnJtF5ULbA1tBhJ
This file contains hidden or 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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What this is
Phase 1 of the NowLookHere build: a generated test set that pins what Liminate's reference interpreter (Python, 0.18.4) does with the
forbid/require/permitsubset, so that other evaluators can be held to it. The TypeScript evaluator (liminate-ts-validator) and the Swift evaluator (liminate-swift) both read this file and must match every case.Files
scripts/gen_agreement_corpus.py(new): readstests/fixtures/agreement_corpus.txt, runs each case through the real interpreter, and writes the outcomes. Followsscripts/gen_conformance_corpus.py.tests/fixtures/agreement_corpus.txt(new): the cases (76).tests/fixtures/agreement-0.18.4.json(new, generated): the interpreter's outcome for each case.tests/test_agreement_corpus.py(new): regenerates in memory and checks the committed JSON is current.A decision recorded while building the set: a program that mixes
andandorin one condition is outside the subset, because the reference interpreter's result for it is amber (a parse that depends on grouping), not a pass or a hold.Verification
pytest tests/test_agreement_corpus.py: 6 passed.liminate-ts-validatorandliminate-swift(checked byscripts/verify-views-safety-identity.shin the strands repository, group A).Untracked files in the working tree (
.playwright-cli/,audit/00_grounding_manifest.md) are not part of this change.🤖 Generated with Claude Code
https://claude.ai/code/session_01Qvgf1CGxnJtF5ULbA1tBhJ