added gap validation.

gap_validation/whittaker_compare.py

@@ -2,6 +2,7 @@
 
 from __future__ import annotations
 
+from datetime import date, datetime
 from pathlib import Path
 
 from metrics_stats import (
@@ -12,32 +13,48 @@ from metrics_stats import (
 )
 
 
+def _date_in_window(dk: str, start: date, end: date) -> bool:
+    try:
+        d = datetime.strptime(dk[:10], "%Y-%m-%d").date()
+    except ValueError:
+        return False
+    return start <= d <= end
+
+
 def whittaker_gcc_on_gap_masked_series(
     base: Path,
     strategy: str,
     prediction_iso: str,
     withheld_iso: str,
+    *,
+    window_start_iso: str | None = None,
+    window_end_iso: str | None = None,
     lam: float = WHITTAKER_LAMBDA_DAYS_SQ,
 ) -> float | None:
-    """Whittaker smooth on cloud-screened S2 GCC **excluding** the withheld acquisition day.
-
-    Comparator aligned with ``baseline.s2_whittaker_lambda400`` in ``metrics_stats`` (same λ,
-    same preselection GCC), but the withheld date is removed so the smoother does not see
-    the target acquisition. Value at ``prediction_iso`` (YYYY-MM-DD) is returned.
-    """
+    """Whittaker on cloud-screened S2 GCC excluding gap-window dates and withheld day."""
     pred_k = _norm_date_key(prediction_iso)
     wh_k = _norm_date_key(withheld_iso)
     if not pred_k or not wh_k:
         return None
+    w0 = w1 = None
+    if window_start_iso and window_end_iso:
+        w0 = datetime.strptime(window_start_iso[:10], "%Y-%m-%d").date()
+        w1 = datetime.strptime(window_end_iso[:10], "%Y-%m-%d").date()
     all_gcc, flags = _s2_gcc_series_from_preselection(base)
     if not all_gcc:
         return None
     idx = 0 if strategy == "aggressive" else 1
-    kept = sorted(
-        (d, g)
-        for d, g in all_gcc.items()
-        if d in flags and not flags[d][idx] and _norm_date_key(d) != wh_k
-    )
+    kept = []
+    for d, g in all_gcc.items():
+        if d not in flags or flags[d][idx]:
+            continue
+        dk = _norm_date_key(d)
+        if not dk or dk == wh_k:
+            continue
+        if w0 is not None and w1 is not None and _date_in_window(dk, w0, w1):
+            continue
+        kept.append((d, g))
+    kept.sort(key=lambda t: t[0])
     if len(kept) < 2:
         return None
     obs_d, obs_v = zip(*kept)
@@ -57,7 +74,7 @@ def first_gap_where_fusion_below_whittaker(
         for r in rows
         if r.get(fusion_key) is not None and r.get(whittaker_key) is not None
     ]
-    eligible.sort(key=lambda r: r["gap_days"])
+    eligible.sort(key=lambda r: (r.get("transition") or "", r["gap_days"]))
     for r in eligible:
         if r[fusion_key] < r[whittaker_key]:
             return int(r["gap_days"])