fix: retry transient official source failures

scripts/provision_mol_municipality_workspace.py

@@ -17,10 +17,12 @@ from pathlib import Path
 from typing import Any, Iterable
 
 import requests
+from requests.adapters import HTTPAdapter
 from pyproj import Transformer
 from shapely.geometry import GeometryCollection, MultiPolygon, Polygon, mapping, shape
 from shapely.ops import transform, unary_union
 from shapely.validation import make_valid
+from urllib3.util.retry import Retry
 
 
 MUNICIPALITY_NAME = "Mol"
@@ -40,6 +42,7 @@ DEFAULT_API_URL = "http://127.0.0.1:8000"
 DEFAULT_PAGE_LIMIT = 1000
 DEFAULT_MAX_FEATURES = 100000
 GEOJSON_CRS = {"type": "name", "properties": {"name": "EPSG:4326"}}
+RETRYABLE_SOURCE_STATUS_CODES = (429, 500, 502, 503, 504)
 
 
 def parse_args() -> argparse.Namespace:
@@ -116,6 +119,25 @@ def write_json(path: Path, payload: dict[str, Any], *, pretty: bool = False) ->
     )
 
 
+def build_source_session() -> requests.Session:
+    retry = Retry(
+        total=5,
+        connect=5,
+        read=5,
+        status=5,
+        backoff_factor=1.0,
+        status_forcelist=RETRYABLE_SOURCE_STATUS_CODES,
+        allowed_methods=frozenset({"GET"}),
+        raise_on_status=True,
+    )
+    adapter = HTTPAdapter(max_retries=retry)
+    session = requests.Session()
+    session.headers.update({"User-Agent": "GeoIntel-Mol-Municipality-Operator/1.0"})
+    session.mount("https://", adapter)
+    session.mount("http://", adapter)
+    return session
+
+
 def next_page_url(payload: dict[str, Any]) -> str | None:
     links = payload.get("links") or []
     for link in links:
@@ -319,8 +341,7 @@ def prepare_artifacts(args: argparse.Namespace) -> tuple[Path, Path, dict[str, A
     if args.page_limit <= 0 or args.max_features <= 0:
         raise RuntimeError("--page-limit and --max-features must be positive integers")
 
-    with requests.Session() as session:
-        session.headers.update({"User-Agent": "GeoIntel-Mol-Municipality-Operator/1.0"})
+    with build_source_session() as session:
         boundary_feature, boundary, boundary_source_url = fetch_mol_boundary(session, args.request_timeout)
         buildings, building_summary = build_municipality_buildings(
             iter_grb_pages(