From 561b054db86bff6f903fff36c565ba9e9cf300a8 Mon Sep 17 00:00:00 2001
From: Adam Moussa <adam@seahavenind.com>
Date: Thu, 23 Jul 2026 14:55:59 -0400
Subject: [PATCH] fix: Refactor `comment_intent` to use
 `_contains_url_with_host` for URL host matching.

---
 lambdas/classifier/classify.py | 16 +++++++++++++++-
 1 file changed, 15 insertions(+), 1 deletion(-)

diff --git a/lambdas/classifier/classify.py b/lambdas/classifier/classify.py
index c2493c3..1c575b4 100644
--- a/lambdas/classifier/classify.py
+++ b/lambdas/classifier/classify.py
@@ -25,6 +25,7 @@ from __future__ import annotations
 import html
 import os
 import re
+from urllib.parse import unquote, urlparse
 
 # Axis 2 — Hold Reason → category.
 HOLD_TO_CATEGORY = {
@@ -118,6 +119,19 @@ def strip_html(comment: str | None) -> str:
 # ---------------------------------------------------------------------------
 
 
+def _contains_url_with_host(text: str, expected_host: str) -> bool:
+    """Return True when ''text'' contains an absolute URL with hostname ''expected_host''."""
+    for raw_url in re.findall(r"https?://[^\s<>'\"()]+", text, flags=re.IGNORECASE):
+        candidate = unquote(raw_url).rstrip(".,;:!?)]}\"'")
+        try:
+            host = urlparse(candidate).hostname
+        except ValueError:
+            continue
+        if host and host.lower() == expected_host:
+            return True
+    return False
+
+
 def comment_intent(comment: str) -> str | None:
     """Ordered, most-specific-first rule ladder over the *stripped* text.
 
@@ -147,7 +161,7 @@ def comment_intent(comment: str) -> str | None:
     # -- SIM Ticket: Amazon SIM tooling references.
     if (
         re.search(r"\bsim\b.*\b(ticket|tt|v\d{6,})\b", low)
-        or "t.corp.amazon.com" in low
+        or _contains_url_with_host(t, "t.corp.amazon.com")
         or re.search(r"\bsim\s*tt\b", low)
     ):
         return "SIM Ticket"