test: enforce the declared moderate drift alternatives - #474
Merged
Merged
Conversation
The original campaign permits verified completion or a proven safe halt for moderate display drift. Keep healthy and reordered rows verified-only, and bind both moderate alternatives to native reports and independent effect evidence. Preserve the observed identity refusal and reconciliation result. Signed-off-by: Richard Abrich <richard.abrich@mldsai.com>
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The qualification campaign has allowed either a verified write or a safe halt for moderate display drift since its original contract. The test instead required every moderate-drift trial to finish
VERIFIED. A valid identity refusal therefore failed release qualification.Keep healthy and reordered-row trials under the existing verified-write assertion. Check moderate drift against both declared outcomes: successful runs must prove the exact single write, while refused runs must prove zero Save attempts and no consequential input or business rows. Negative cases reject an unverified success, missing effect evidence, and a halt that dispatched a write.
The retained failure shows the intended boundary: the Save target resolved, but the identity check refused before Save and returned
RECONCILIATION_REQUIRED. This change doesn't alter the runtime, fixture, thresholds, campaign contract, or reported metrics. The separate OCR word-join limitation remains open; this PR makes no moderate-drift success claim.Validation: all required PR checks pass at
3935947bc0728d3ea857981341471f10a087122d. The 20 focused assertion cases also pass independently, including jointly consistent unverified success and dispatched-refusal negatives. Ruff and the whitespace check pass. The single local module check retained 29 of 33 trial reports before its existing 600-second timeout fired during Scrypt checkpoint derivation: 27 tests passed and 8 fixture-dependent checks did not run. A separate read-only audit of all three completed moderate trials verifies their native refusals against the actual SQLite databases and input ledgers. The timeout remains recorded; it is not a completed campaign result. The unchanged full release gate must pass on the merged source.