scripts/cull_principles.py

@@ -50,11 +50,16 @@ _JUDGE_CONCURRENCY = 4
 async def _decisions(limit, sample):
     """case_law ids that have 'original' principles, with source metadata."""
     pool = await db.get_pool()
+    # Resumable: skip decisions already culled (their survivors carry reviewer
+    # 'cull:%'), so a continuation run after a ceiling-stop never re-judges them.
+    # (Dry-run never sets that reviewer, so nothing is skipped in dry-run.)
     rows = await pool.fetch(
         "SELECT cl.id, cl.case_number, cl.source_kind, cl.is_binding, "
         "       count(*) AS n "
         "FROM halachot h JOIN case_law cl ON cl.id = h.case_law_id "
         "WHERE h.instance_type = 'original' AND h.review_status <> 'rejected' "
+        "AND NOT EXISTS (SELECT 1 FROM halachot h2 WHERE h2.case_law_id = cl.id "
+        "  AND h2.instance_type = 'original' AND h2.reviewer LIKE 'cull:%') "
         "GROUP BY cl.id, cl.case_number, cl.source_kind, cl.is_binding "
         "ORDER BY n DESC",
     )