Complète l’étape 26 avec l’évolution minifigs/set

lib/rebrickable/minifig_parts_correlation.py

@@ -80,6 +80,62 @@ def build_correlation_rows(
     return rows
 
 
+def build_minifigs_per_year(
+    filtered_counts_path: Path,
+    all_sets_path: Path,
+    inventories_path: Path,
+    inventory_minifigs_path: Path,
+) -> List[dict]:
+    """Calcule le nombre moyen de minifigs par set et par année (filtré vs catalogue)."""
+    filtered_totals: Dict[int, Dict[str, int]] = {}
+    with filtered_counts_path.open() as csv_file:
+        reader = csv.DictReader(csv_file)
+        for row in reader:
+            year = int(row["year"])
+            current = filtered_totals.get(year)
+            if current is None:
+                filtered_totals[year] = {"minifigs": int(row["minifig_count"]), "sets": 1}
+            else:
+                current["minifigs"] += int(row["minifig_count"])
+                current["sets"] += 1
+    global_minifigs = build_global_minifig_counts(inventories_path, inventory_minifigs_path)
+    catalog_totals: Dict[int, Dict[str, int]] = {}
+    with all_sets_path.open() as csv_file:
+        reader = csv.DictReader(csv_file)
+        for row in reader:
+            year = int(row["year"])
+            current = catalog_totals.get(year)
+            if current is None:
+                catalog_totals[year] = {"minifigs": global_minifigs.get(row["set_num"], 0), "sets": 1}
+            else:
+                current["minifigs"] += global_minifigs.get(row["set_num"], 0)
+                current["sets"] += 1
+    rows: List[dict] = []
+    for year in sorted(filtered_totals.keys()):
+        totals = filtered_totals[year]
+        average = totals["minifigs"] / totals["sets"]
+        rows.append(
+            {
+                "scope": "filtered",
+                "year": str(year),
+                "average_minifigs_per_set": f"{average:.3f}",
+                "set_count": str(totals["sets"]),
+            }
+        )
+    for year in sorted(catalog_totals.keys()):
+        totals = catalog_totals[year]
+        average = totals["minifigs"] / totals["sets"]
+        rows.append(
+            {
+                "scope": "catalog",
+                "year": str(year),
+                "average_minifigs_per_set": f"{average:.3f}",
+                "set_count": str(totals["sets"]),
+            }
+        )
+    return rows
+
+
 def write_correlation_rows(path: Path, rows: Sequence[dict]) -> None:
     """Écrit les lignes de corrélation pièces/minifigs."""
     ensure_parent_dir(path)
@@ -91,6 +147,17 @@ def write_correlation_rows(path: Path, rows: Sequence[dict]) -> None:
             writer.writerow(row)
 
 
+def write_minifigs_per_year(path: Path, rows: Sequence[dict]) -> None:
+    """Écrit le CSV annuel minifigs / set."""
+    ensure_parent_dir(path)
+    fieldnames = ["scope", "year", "average_minifigs_per_set", "set_count"]
+    with path.open("w", newline="") as csv_file:
+        writer = csv.DictWriter(csv_file, fieldnames=fieldnames)
+        writer.writeheader()
+        for row in rows:
+            writer.writerow(row)
+
+
 def load_correlation_rows(path: Path) -> List[dict]:
     """Charge le CSV de corrélation pièces/minifigs."""
     return read_rows(path)