feat: expand RF4 media inventory

tests/test_community_cli.py

@@ -36,6 +36,7 @@ def test_fetch_site_key_normalizes_common_subdomains() -> None:
     assert fetch_site_key("https://www.rf4db.com/") == "rf4db.com"
     # Base domain stays the same
     assert fetch_site_key("https://rf4db.com/") == "rf4db.com"
+    assert fetch_site_key("https://gw.rf4map.ru/public/images/fish.webp") == "rf4map.ru"
     # Non-normalized subdomains stay as-is
     assert fetch_site_key("https://rf4map.ru/point/1") == "rf4map.ru"
     assert fetch_site_key("https://rf4-posts.com/spots/") == "rf4-posts.com"

tests/test_media_assets.py

@@ -34,6 +34,14 @@ def test_gallery_page_provenance_classifies_generic_asset_urls() -> None:
     assert chrome[0].entity_type == "reference"
 
 
+def test_rf4map_gateway_fish_filename_is_classified_as_fish() -> None:
+    items = extract_media_candidates(
+        '<img src="https://gw.rf4map.ru/public/images/fish_123.webp" alt="Ерш">',
+        source_page="https://rf4map.ru/fishes",
+    )
+    assert items[0].entity_type == "fish"
+
+
 def test_manifest_merges_and_binary_store_is_content_addressed(tmp_path: Path) -> None:
     item = extract_media_candidates('<img src="/bait/worm.png" alt="Bait worm">', source_page="https://example.test")[0]
     first = merge_manifest(tmp_path / "manifest.json", [item])