Malin/DomGod
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: bind exception variable in ConnectError handler to prevent NameError

Browse Source 
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
Malin
2026-04-20 18:48:04 +02:00
parent 6657e6ea1f
commit b53545b7dd
1 changed files with 19 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

29
app/validator.py
View File

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