Cache postcodes

finder/main.py

@@ -5,10 +5,12 @@ import tempfile
 import time
 from pathlib import Path
 
+import shutdown
 from constants import DATA_DIR, REPO_DIR
 
 
-SOURCE_CHOICES = ("rightmove", "onthemarket", "zoopla", "all")
+SOURCES = ("rightmove", "onthemarket", "zoopla")
+SOURCE_CHOICES = (*SOURCES, "all")
 TEST_MAX_PROPERTIES_PER_SOURCE = 100
 TEST_OUTCODES = (
     "E1",
@@ -47,9 +49,13 @@ def parse_args() -> argparse.Namespace:
     )
     parser.add_argument(
         "--source",
-        choices=SOURCE_CHOICES,
         default="all",
-        help="Portal to scrape. 'all' runs Rightmove, OnTheMarket, and Zoopla.",
+        metavar="SOURCES",
+        help=(
+            "Comma-separated portal(s) to scrape: any of "
+            f"{', '.join(SOURCES)}, or 'all' (default). "
+            "E.g. --source rightmove,onthemarket."
+        ),
     )
     parser.add_argument(
         "--output-dir",
@@ -100,15 +106,35 @@ def configure_logging() -> None:
 
 
 def selected_sources(source: str) -> list[str]:
-    if source == "all":
-        return ["rightmove", "onthemarket", "zoopla"]
-    return [source]
+    """Resolve --source: one or more comma-separated portals, or 'all'.
+
+    Accepts e.g. ``rightmove,onthemarket`` (whitespace and case tolerant).
+    Unknown values are rejected; the result is deduplicated and returned in the
+    canonical ``SOURCES`` order so downstream merge/dedup stays deterministic."""
+    requested = [part.strip().lower() for part in source.split(",")]
+    requested = [part for part in requested if part]
+    if not requested:
+        raise SystemExit("--source was empty")
+
+    unknown = sorted(set(requested) - set(SOURCE_CHOICES))
+    if unknown:
+        raise SystemExit(
+            f"Unknown --source value(s): {', '.join(unknown)}. "
+            f"Choose from {', '.join(SOURCE_CHOICES)} (comma-separated)."
+        )
+    if "all" in requested:
+        return list(SOURCES)
+    return [portal for portal in SOURCES if portal in requested]
 
 
 def main() -> int:
     args = parse_args()
     configure_standalone_runtime()
     configure_logging()
+    # Ctrl+C (and SIGTERM, e.g. `docker stop`) asks the scrapers to wind down
+    # gracefully — each source stops at its next outcode boundary and the run
+    # still persists detail caches and writes the listings collected so far.
+    shutdown.install_signal_handlers()
 
     if args.limit_outcodes is not None and args.limit_outcodes < 1:
         raise SystemExit("--limit-outcodes must be greater than zero")
@@ -182,6 +208,14 @@ def main() -> int:
     )
 
     elapsed = time.monotonic() - started
+    if shutdown.stop_requested():
+        log.warning(
+            "Scrape interrupted after %.1fs; partial results written to %s",
+            elapsed,
+            result.get("path"),
+        )
+        log.info("Result: %s", result)
+        return 130  # 128 + SIGINT, the conventional Ctrl+C exit code.
     log.info("Scrape finished in %.1fs", elapsed)
     log.info("Result: %s", result)
     if args.test and result.get("errors"):