fix: reject partial-zero classification dates (prof-5qy)#42
Merged
Conversation
Classification dates are either the "00000000" no-date sentinel or a real YYYYMMDD calendar date with ASCII digits, year >= 1, month 01-12, and a day valid for that month with leap years honored. Partial-zero components ("20240900" day 00, "20240015" month 00, "00000901" year 0000), impossible dates (e.g. "20240230"), and non-ASCII digit characters collapse to the sentinel.
drover.dates exposes is_valid_classification_date() and normalize_classification_date(). The model boundary (RawClassification.date, ClassificationResult.date) carries a non-raising mode="before" field_validator that normalizes the LLM-supplied date before any downstream consumer (naming policy, tag actions, eval comparison, JSON export, on-disk reloads) reads it. NARA naming delegates to the shared normalizer. GroundTruthEntry.date raises on bad values; _load_ground_truth catches the resulting ValidationError and logs the offending line. The synthetic-sample generator validates against the same rule. The classification prompt instructs the model to emit the sentinel rather than zero-fill a single component.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Classification dates are either the "00000000" no-date sentinel or a real YYYYMMDD calendar date with ASCII digits, year >= 1, month 01-12, and a day valid for that month with leap years honored. Partial-zero components ("20240900" day 00, "20240015" month 00, "00000901" year 0000), impossible dates (e.g. "20240230"), and non-ASCII digit characters collapse to the sentinel.
drover.dates exposes is_valid_classification_date() and normalize_classification_date(). The model boundary (RawClassification.date, ClassificationResult.date) carries a non-raising mode="before" field_validator that normalizes the LLM-supplied date before any downstream consumer (naming policy, tag actions, eval comparison, JSON export, on-disk reloads) reads it. NARA naming delegates to the shared normalizer. GroundTruthEntry.date raises on bad values; _load_ground_truth catches the resulting ValidationError and logs the offending line. The synthetic-sample generator validates against the same rule. The classification prompt instructs the model to emit the sentinel rather than zero-fill a single component.