From 5dbb0cb8af6d17ba18a9a4a794ca820c8606d673 Mon Sep 17 00:00:00 2001
From: Stacy Brock <brock@oregonstate.edu>
Date: Mon, 29 Jul 2024 09:19:57 -0700
Subject: [PATCH] Normalize all inputs to is_in_message() function

---
 filter-rules.py | 29 ++++++++++++++++-------------
 1 file changed, 16 insertions(+), 13 deletions(-)

diff --git a/filter-rules.py b/filter-rules.py
index 82b15a5..29c8720 100644
--- a/filter-rules.py
+++ b/filter-rules.py
@@ -1,3 +1,5 @@
+import re
+
 BLOCK_EMAIL = [
     'support-noreply@status.duosecurity.com',
     'alerts@alerts.mail.hpe.com',
@@ -24,6 +26,7 @@ BLOCK_KEYWORDS = [
     'no longer would like to be contacted',
     'opt-out',
     'opt out',
+    'piano',
     'pmp exam',
     'prefer not to receive',
     'prefer not to see',
@@ -41,6 +44,7 @@ BLOCK_KEYWORDS = [
     'want to receive',
     'webcast',
     'webinar',
+    'welder',
     'white paper',
     'whitepaper',
     'wish to be contacted',
@@ -311,7 +315,7 @@ def filter_message(self, message):
         return
 
     # delete other spam
-    if is_in_message(normalized['BLOCK_ANNOYING'], message):
+    if is_in_message(BLOCK_ANNOYING, message):
         self._log_result(message, 'junking spam containing annoying content')
         move_message(message, 'Junk Email')
         return
@@ -336,7 +340,7 @@ def filter_message(self, message):
         return
 
     # junk messages with blocked keywords
-    if is_in_message(normalized['BLOCK_KEYWORDS'], message):
+    if is_in_message(BLOCK_KEYWORDS, message):
         self._log_result(message, 'junking spam containing blocked keyword')
         move_message(message, 'Junk Email')
         return
@@ -381,14 +385,17 @@ def is_in_message(list_, message):
     Returns True if any string in the list is found in the message
     """
     is_found = False
-    message_body = message.body.lower()
-    message_subject = message.subject.lower()
-    message_from = message.sender.address.lower()
 
-    for t in list_:
-        term = t.lower()
-        if (term in message_subject or term in message_from
-                or term in message_body):
+    # normalize inputs
+    search_terms = [x.lower() for x in list_]
+    message_body = message.body.lower().replace('\ufeff', '')
+    message_subject = message.subject.lower().replace('\ufeff', '')
+    message_from = message.sender.address.lower().replace('\ufeff', '')
+
+    for term in search_terms:
+        if (re.search(term, message_subject)
+                or re.search(term, message_from)
+                or re.search(term, message_body)):
             is_found = True
             break
     return is_found
@@ -413,10 +420,6 @@ def init_filters(self):
     for k,v in self._categories.items():
         category_cache[k] = v
 
-    # normalize filter keyword lists
-    normalized['BLOCK_KEYWORDS'] = [x.lower() for x in BLOCK_KEYWORDS]
-    normalized['BLOCK_ANNOYING'] = [x.lower() for x in BLOCK_ANNOYING]
-
 def get_header(header_key, headers):
     vals = []
     for header in headers: