From b53545b7ddf0f915dea447473174258b363c226c Mon Sep 17 00:00:00 2001
From: Malin <malin@cloudhost.es>
Date: Mon, 20 Apr 2026 18:48:04 +0200
Subject: [PATCH] fix: bind exception variable in ConnectError handler to
 prevent NameError

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
---
 app/validator.py | 29 +++++++++++++++++++----------
 1 file changed, 19 insertions(+), 10 deletions(-)

diff --git a/app/validator.py b/app/validator.py
index d5a88a0..9abb50a 100644
--- a/app/validator.py
+++ b/app/validator.py
@@ -104,8 +104,14 @@ async def _check_domain(domain: str) -> dict:
     # that accept TCP but never respond on port 80.  If http fails for ANY
     # reason (refused, timeout, protocol error, redirect loop…) we fall back
     # to https directly, which is what most modern sites actually serve.
-    timeouts = {"http": httpx.Timeout(connect=4, read=8, write=5, pool=10),
-                "https": httpx.Timeout(connect=7, read=12, write=5, pool=15)}
+    # Use a short connect timeout for http — if port 80 times out, port 443
+    # will too, so we don't bother retrying on timeout errors.
+    # We DO retry https on ConnectError/RemoteProtocolError (port 80 closed or
+    # speaking wrong protocol) because those servers are often https-only.
+    timeouts = {
+        "http":  httpx.Timeout(connect=4, read=6,  write=3, pool=5),
+        "https": httpx.Timeout(connect=6, read=10, write=3, pool=5),
+    }
 
     for scheme in ("http", "https"):
         try:
@@ -126,7 +132,6 @@ async def _check_domain(domain: str) -> dict:
             final_url  = str(resp.url)
             final_host = urlparse(final_url).netloc.lower().lstrip("www.")
 
-            # Redirected to a completely different domain
             if not _same_domain(domain, final_url):
                 for ph in PARKING_REDIRECT_HOSTS:
                     if ph in final_host:
@@ -135,9 +140,8 @@ async def _check_domain(domain: str) -> dict:
                 result["prescreen_status"] = "redirect"
                 return result
 
-            # Any response from the server = the domain is live.
-            # 4xx/5xx still means a working web server — only no-response = dead.
-            # Only check parking signals on 200 responses (2xx bodies are readable).
+            # Any HTTP response = server is alive (4xx/5xx still means live web server)
+            # Only check parking content on readable 200 responses
             if resp.status_code in (200, 203):
                 html_lc = resp.text[:20_000].lower()
                 for sig in PARKING_BODY_SIGNALS:
@@ -148,11 +152,16 @@ async def _check_domain(domain: str) -> dict:
             result["prescreen_status"] = "live"
             return result
 
+        except (httpx.ConnectError, httpx.RemoteProtocolError) as e:
+            # Port refused or bad HTTP response → server may be https-only, try it
+            if scheme == "http":
+                logger.debug("Validator %s: %s on http, trying https", domain, type(e).__name__)
+                continue
+            break
         except Exception as e:
-            # Any failure on http → try https. Any failure on https → dead.
-            logger.debug("Validator %s (%s): %s — %s", domain, scheme, type(e).__name__, e)
-            if scheme == "https":
-                break
+            # Timeout or other error → https won't help, mark dead
+            logger.debug("Validator %s (%s): %s", domain, scheme, type(e).__name__)
+            break
 
     result["load_time_ms"] = int((time.monotonic() - t0) * 1000)
     return result