Exclure les pièces de minifigs et intégrer les visuels de rareté

scripts/compute_part_rarity.py

@@ -0,0 +1,34 @@
+"""Calcule les pièces rares en comparant les sets filtrés au reste du catalogue."""
+
+from pathlib import Path
+
+from lib.rebrickable.part_rarity import build_part_rarity, select_until_reused, write_part_rarity
+
+
+PARTS_FILTERED_PATH = Path("data/intermediate/parts_filtered.csv")
+INVENTORIES_PATH = Path("data/raw/inventories.csv")
+INVENTORY_PARTS_PATH = Path("data/raw/inventory_parts.csv")
+PARTS_CATALOG_PATH = Path("data/raw/parts.csv")
+PART_CATEGORIES_PATH = Path("data/raw/part_categories.csv")
+FILTERED_SETS_PATH = Path("data/intermediate/sets_enriched.csv")
+DESTINATION_PATH = Path("data/intermediate/part_rarity.csv")
+TOP_DESTINATION_PATH = Path("data/intermediate/part_rarity_exclusive.csv")
+
+
+def main() -> None:
+    """Construit les CSV complets et top 10 des pièces les plus rares."""
+    rows = build_part_rarity(
+        PARTS_FILTERED_PATH,
+        INVENTORIES_PATH,
+        INVENTORY_PARTS_PATH,
+        PARTS_CATALOG_PATH,
+        PART_CATEGORIES_PATH,
+        FILTERED_SETS_PATH,
+    )
+    write_part_rarity(DESTINATION_PATH, rows)
+    top_rows = select_until_reused(rows)
+    write_part_rarity(TOP_DESTINATION_PATH, top_rows)
+
+
+if __name__ == "__main__":
+    main()

scripts/download_part_rarity_resources.py

@@ -0,0 +1,63 @@
+"""Télécharge les visuels des pièces les plus rares identifiées à l'étape 34."""
+
+import os
+from pathlib import Path
+
+import requests
+from dotenv import load_dotenv
+
+from lib.rebrickable.part_rarity import load_part_rarity
+from lib.rebrickable.resources import (
+    build_part_img_lookup,
+    download_binary,
+    download_resources,
+    fetch_part_img_url,
+    load_part_img_cache,
+    persist_part_img_cache,
+)
+
+
+PART_RARITY_TOP_PATH = Path("data/intermediate/part_rarity_exclusive.csv")
+RESOURCES_DIR = Path("figures/rebrickable")
+PART_IMG_CACHE_PATH = Path("data/intermediate/part_img_cache.csv")
+DOWNLOAD_LOG_PATH = Path("data/intermediate/part_rarity_download_log.csv")
+REQUEST_DELAY_SECONDS_IMAGES = 0.35
+REQUEST_DELAY_SECONDS_LOOKUP = 0.6
+
+
+def main() -> None:
+    """Construit les URLs d'images des pièces rares et les télécharge."""
+    load_dotenv()
+    token = os.environ["REBRICKABLE_TOKEN"]
+    session = requests.Session()
+
+    rows = load_part_rarity(PART_RARITY_TOP_PATH)
+    cache = load_part_img_cache(PART_IMG_CACHE_PATH)
+    part_img_lookup = build_part_img_lookup(
+        {row["part_num"] for row in rows},
+        fetcher=lambda part_num: fetch_part_img_url(part_num, token, session),
+        cache_path=PART_IMG_CACHE_PATH,
+        existing_cache=cache,
+        delay_seconds=REQUEST_DELAY_SECONDS_LOOKUP,
+    )
+    if cache:
+        part_img_lookup.update(cache)
+    persist_part_img_cache(PART_IMG_CACHE_PATH, part_img_lookup)
+
+    plan = [
+        {
+            "url": part_img_lookup[row["part_num"]],
+            "path": RESOURCES_DIR / row["sample_set_id"] / "rare_parts" / f"{row['part_num']}.jpg",
+        }
+        for row in rows
+    ]
+    download_resources(
+        plan,
+        downloader=lambda url, path: download_binary(url, path, session),
+        delay_seconds=REQUEST_DELAY_SECONDS_IMAGES,
+        log_path=DOWNLOAD_LOG_PATH,
+    )
+
+
+if __name__ == "__main__":
+    main()

scripts/plot_part_rarity.py

@@ -0,0 +1,19 @@
+"""Trace le diagramme des pièces rares pour l'étape 34."""
+
+from pathlib import Path
+
+from lib.plots.part_rarity import plot_part_rarity
+
+
+PART_RARITY_TOP_PATH = Path("data/intermediate/part_rarity_exclusive.csv")
+DESTINATION_PATH = Path("figures/step34/part_rarity.png")
+RESOURCES_DIR = Path("figures/rebrickable")
+
+
+def main() -> None:
+    """Charge le top des pièces rares et produit le graphique illustré."""
+    plot_part_rarity(PART_RARITY_TOP_PATH, DESTINATION_PATH, resources_dir=RESOURCES_DIR)
+
+
+if __name__ == "__main__":
+    main()