rules: skip rules with utf8 decoding errors

If a rule can't be decoded as utf-8, we can't further parse it. So
skip it with a warning. The previous behavior was to raise an
exception and abort.

Ticket: #7812

Author: jasonish

suricata/update/rule.py

@@ -321,11 +321,16 @@ def parse_fileobj(fileobj, group=None):
     rules = []
     buf = ""
     for line in fileobj:
-        try:
-            if type(line) == type(b""):
-                line = line.decode()
-        except:
-            pass
+        if type(line) == type(b""):
+            try:
+                line = line.decode("utf-8", "strict")
+            except UnicodeDecodeError:
+                logger.warning("Skipping rule due to encoding issue: %s", repr(line))
+                # Skip this line and reset buffer if we were accumulating a multi-line rule
+                if buf:
+                    logger.warning("Discarding incomplete multi-line rule due to encoding issue")
+                    buf = ""
+                continue
         if line.rstrip().endswith("\\"):
             buf = "%s%s " % (buf, line.rstrip()[0:-1])
             continue

tests/rules-with-encoding-issues.rules

@@ -0,0 +1,11 @@
+alert tcp any any -> any any (msg:"Valid rule 1"; sid:1001; rev:1;)
+alert tcp any any -> any any (msg:"Bad encoding ��"; sid:1002; rev:1;)
+alert tcp any any -> any any (msg:"Latin1 ��"; sid:1003; rev:1;)
+alert tcp any any -> any any (msg:"Valid rule 2"; sid:1004; rev:1;)
+alert tcp any any -> any any (msg:"Valid multiline"; \
+    content:"test"; \
+    sid:2001; rev:1;)
+alert tcp any any -> any any (msg:"Bad multiline"; \
+    content:"bad��"; \
+    sid:2002; rev:1;)
+alert tcp any any -> any any (msg:"Valid after bad"; sid:2003; rev:1;)

tests/test_rule.py

@@ -262,3 +262,30 @@ def test_parse_var_lists(self):
         self.assertEqual(rule["source_port"], "any")
         self.assertEqual(rule["dest_addr"], "[any,![$EXTERNAL_IP,$REVERSE_PROXY_HOSTS,$ODD_HTTP_HOSTS]]")
         self.assertEqual(rule["dest_port"], "80")
+
+    def test_rule_with_encoding_issues(self):
+        """Test that rules with encoding issues are skipped (Issue #7812)"""
+        # Parse the file with encoding issues
+        with open('tests/rules-with-encoding-issues.rules', 'rb') as fileobj:
+            rules = suricata.update.rule.parse_fileobj(fileobj)
+
+        # Should have parsed:
+        # - Valid single-line rules (1001 and 1004)
+        # - Valid multiline rule (2001)
+        # - Valid rule after bad multiline (2003)
+        # Should have skipped:
+        # - Rules with bad encoding (1002, 1003)
+        # - Bad multiline rule (2002)
+        self.assertEqual(len(rules), 4)
+
+        # Check single-line rules
+        self.assertEqual(rules[0].sid, 1001)
+        self.assertEqual(rules[0].msg, "Valid rule 1")
+        self.assertEqual(rules[1].sid, 1004)
+        self.assertEqual(rules[1].msg, "Valid rule 2")
+
+        # Check multiline rules
+        self.assertEqual(rules[2].sid, 2001)
+        self.assertEqual(rules[2].msg, "Valid multiline")
+        self.assertEqual(rules[3].sid, 2003)
+        self.assertEqual(rules[3].msg, "Valid after bad")