feat(tht): sync-schema-linking projects F4 ledger into schema_linking.json

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
2026-07-06 23:25:00 +02:00
committed by Marco Pancotti
co-authored by Claude Opus 4.8
parent 1df5d40d05
commit 00365cc6e6
3 changed files with 122 additions and 0 deletions
+45
View File
@@ -181,6 +181,51 @@ def set_schema_linking(
return path
def sync_schema_linking(session_id: str, sessions_root: Path) -> Path:
"""Project the effective F4 ledger decisions into schema_linking.json.
candidates/excluded are rebuilt from table_promoted/table_excluded +
column_promoted/column_excluded (last decision per subject wins). question,
joins, concept_formulas and open_questions are preserved from the existing
file when present. The reviewer's curation is thus authoritative and
deterministic (no model transcription)."""
from tht.phase import effective_decisions
session_dir = sessions_root / session_id
manifest = load_session(session_id, sessions_root)
existing: dict = {}
sl_path = session_dir / "schema_linking.json"
if sl_path.exists():
existing = json.loads(sl_path.read_text())
# last decision per subject wins (handles a re-run of the gate).
latest: dict[str, str] = {}
for d in effective_decisions(session_dir):
if d.type in ("table_promoted", "table_excluded", "column_promoted", "column_excluded"):
latest[d.subject] = d.type
candidates: list[dict] = []
excluded: list[dict] = []
for subject, dtype in latest.items():
is_column = "." in subject
kind = "column" if is_column else "table"
if dtype in ("table_promoted", "column_promoted"):
candidates.append({"kind": kind, "name": subject, "decision": "promoted"})
else:
excluded.append({"kind": kind, "name": subject})
data = {
"question": existing.get("question") or manifest.question,
"candidates": candidates,
"joins": existing.get("joins", []),
"excluded": excluded,
"open_questions": existing.get("open_questions", []),
"concept_formulas": existing.get("concept_formulas", []),
}
return set_schema_linking(session_id, data, sessions_root)
def load_session(session_id: str, sessions_root: Path) -> SessionManifest:
path = sessions_root / session_id / MANIFEST
if not path.exists():