From f951445e8214cb3088518bc73f67b1bc9cbf6764 Mon Sep 17 00:00:00 2001
From: s-aga-r <sagarsharma.s312@gmail.com>
Date: Mon, 16 Feb 2026 20:42:52 +0530
Subject: [PATCH] fix: properly decode encoded `Subject` header (#37016)

* fix: properly decode encoded `Subject` header

* test(email): add RFC2047 subject decoding tests for InboundMail
---
 .../email_account/test_email_account.py       | 62 +++++++++++++++++++
 frappe/email/receive.py                       | 28 +++++----
 2 files changed, 79 insertions(+), 11 deletions(-)

diff --git a/frappe/email/doctype/email_account/test_email_account.py b/frappe/email/doctype/email_account/test_email_account.py
index c6f88c2862..b50dd69b34 100644
--- a/frappe/email/doctype/email_account/test_email_account.py
+++ b/frappe/email/doctype/email_account/test_email_account.py
@@ -651,6 +651,68 @@ class TestInboundMail(IntegrationTestCase):
 		reference_doc = inbound_mail.reference_document()
 		self.assertEqual(todo.name, reference_doc.name)
 
+	def test_inbound_mail_decodes_rfc2047_subject(self):
+		subjects = [
+			# UTF-8 Quoted-Printable (English)
+			(
+				"=?UTF-8?Q?New_Notifications?=",
+				"RE: New Notifications",
+			),
+			# UTF-8 Base64 (English)
+			(
+				"=?UTF-8?B?TmV3IE5vdGlmaWNhdGlvbnM=?=",
+				"RE: New Notifications",
+			),
+			# FWD prefix + Base64 (Russian)
+			(
+				"FWD: =?UTF-8?B?0J/RgNC40LLQtdGCINC80LjRgA==?=",
+				"RE: FWD: Привет мир",
+			),
+			# RE prefix + Quoted-Printable (Russian)
+			(
+				"RE: =?UTF-8?Q?=D0=9E=D1=82=D1=87=D0=B5=D1=82_=D0=B3=D0=BE=D1=82=D0=BE=D0=B2?=",
+				"RE: RE: Отчет готов",
+			),
+			# Mixed plain + encoded (number symbol)
+			(
+				"Invoice =?UTF-8?Q?=E2=84=96_1234?=",
+				"RE: Invoice № 1234",
+			),
+			# Multiple encoded words (split header)
+			(
+				"=?UTF-8?B?TmV3?= =?UTF-8?B?IE5vdGlmaWNhdGlvbnM=?=",
+				"RE: New Notifications",
+			),
+			# Emoji (Quoted-Printable)
+			(
+				"=?UTF-8?Q?Deployment_complete_=F0=9F=9A=80?=",
+				"RE: Deployment complete 🚀",
+			),
+			# Lowercase encoding markers
+			(
+				"=?utf-8?b?TmV3IE5vdGlmaWNhdGlvbnM=?=",
+				"RE: New Notifications",
+			),
+			# ISO-8859-1 Quoted-Printable
+			(
+				"=?ISO-8859-1?Q?Ol=E1_Mundo?=",
+				"RE: Olá Mundo",
+			),
+			# Encoded word inside sentence
+			(
+				"Meeting about =?UTF-8?B?0L/RgNC+0LXQutGC?= tomorrow",
+				"RE: Meeting about проект tomorrow",
+			),
+		]
+
+		for subject, expected in subjects:
+			mail_content = self.get_test_mail(fname="incoming-subject-placeholder.raw").replace(
+				"{{ subject }}", subject
+			)
+			email_account = frappe.get_doc("Email Account", "_Test Email Account 1")
+			inbound_mail = InboundMail(mail_content, email_account, 12345, 1)
+			self.assertEqual(inbound_mail.subject, expected)
+
 	def test_create_communication_from_mail(self):
 		# Create email queue record
 		mail_content = self.get_test_mail(fname="incoming-2.raw")
diff --git a/frappe/email/receive.py b/frappe/email/receive.py
index 8d1c64217a..08157dc5f6 100644
--- a/frappe/email/receive.py
+++ b/frappe/email/receive.py
@@ -424,21 +424,27 @@ class Email:
 
 	def set_subject(self):
 		"""Parse and decode `Subject` header."""
-		_subject = decode_header(self.mail.get("Subject", "No Subject"))
-		self.subject = _subject[0][0] or ""
 
-		if charset := _subject[0][1]:
-			# Encoding is known by decode_header (might also be unknown-8bit)
-			self.subject = safe_decode(self.subject, charset, ALTERNATE_CHARSET_MAP)
+		raw_subject = self.mail.get("Subject")
+		if not raw_subject:
+			self.subject = "No Subject"
+			return
 
-		if isinstance(self.subject, bytes):
-			# Fall back to utf-8 if the charset is unknown or decoding fails
-			# Replace invalid characters with '<?>'
-			self.subject = self.subject.decode("utf-8", "replace")
+		decoded_fragments = []
+		for fragment, charset in decode_header(raw_subject):
+			if isinstance(fragment, bytes):
+				charset = charset or "utf-8"
+				try:
+					fragment = fragment.decode(charset, errors="replace")
+				except LookupError:
+					# Fallback to utf-8 if decoding fails
+					fragment = fragment.decode("utf-8", errors="replace")
+			decoded_fragments.append(fragment)
+
+		subject = "".join(decoded_fragments).strip()
 
-		# Convert non-string (e.g. None)
 		# Truncate to 140 chars (can be used as a document name)
-		self.subject = str(self.subject).strip()[:140] or "No Subject"
+		self.subject = subject[:140] if subject else "No Subject"
 
 	def set_from(self):
 		# gmail mailing-list compatibility