fix: enforce community cooldown per site

rf4_research/community_cli.py

@@ -7,6 +7,7 @@ import sys
 import time
 from dataclasses import asdict
 from pathlib import Path
+from urllib.parse import urlsplit
 from urllib.request import Request, urlopen
 
 from .community_sources import (
@@ -32,6 +33,16 @@ MIN_FETCH_INTERVAL_SECONDS = 30 * 60
 DEFAULT_STATE_FILE = Path(".cache/community-fetch-state.json")
 
 
+def fetch_site_key(url: str) -> str:
+    """Return a stable cooldown key shared by all endpoints of one site."""
+    hostname = (urlsplit(url).hostname or "").lower()
+    if hostname.startswith("www."):
+        hostname = hostname[4:]
+    if not hostname:
+        raise ValueError("source URL must include a hostname")
+    return hostname
+
+
 def enforce_fetch_interval(
     source: str, *, state_file: Path, now: float | None = None,
 ) -> None:
@@ -83,9 +94,11 @@ def main(argv: list[str] | None = None) -> int:
     default_url, parse = SOURCES.get(args.source, (None, DETAIL_SOURCES.get(args.source)))
     url = args.url or default_url
     try:
-        enforce_fetch_interval(args.source, state_file=args.state_file)
+        site_key = fetch_site_key(url)
+        enforce_fetch_interval(site_key, state_file=args.state_file)
+        # Reserve before network I/O: failed attempts count toward the limit too.
+        mark_fetch(site_key, state_file=args.state_file)
         html = fetch_html(url)
-        mark_fetch(args.source, state_file=args.state_file)
         records = (parse(html, source_url=url) if args.source in DETAIL_SOURCES else parse(html))[:args.limit]
     except Exception as exc:
         print(f"community source failed: {exc}", file=sys.stderr)