From 46999debdb623fec36248128cc95059830403ad3 Mon Sep 17 00:00:00 2001
From: Afonso Coutinho <afonso@omelhorsite.pt>
Date: Wed, 3 Jun 2026 05:45:33 +0100
Subject: [PATCH] Decode email headers without injected spaces

Use email.header.make_header for MIME header decoding so adjacent encoded/plain header parts preserve RFC spacing, with regression coverage.
---
 mcp_servers/email_server.py                  | 27 +++++++++++-----
 tests/test_mcp_email_decode_header_spaces.py | 34 ++++++++++++++++++++
 2 files changed, 53 insertions(+), 8 deletions(-)
 create mode 100644 tests/test_mcp_email_decode_header_spaces.py

diff --git a/mcp_servers/email_server.py b/mcp_servers/email_server.py
index 354e592..8438577 100644
--- a/mcp_servers/email_server.py
+++ b/mcp_servers/email_server.py
@@ -337,14 +337,25 @@ def _decode_header(raw):
     """Decode MIME encoded header."""
     if not raw:
         return ""
-    parts = email.header.decode_header(raw)
-    decoded = []
-    for data, charset in parts:
-        if isinstance(data, bytes):
-            decoded.append(data.decode(charset or "utf-8", errors="replace"))
-        else:
-            decoded.append(data)
-    return " ".join(decoded)
+    try:
+        # make_header concatenates per RFC 2047: no spurious space between an
+        # encoded-word and adjacent plain text (plain runs keep their own
+        # whitespace), and whitespace between two adjacent encoded-words is
+        # dropped. The old " ".join produced "Re:  Jose" style double spaces
+        # on every non-ASCII subject or sender.
+        return str(email.header.make_header(email.header.decode_header(raw)))
+    except Exception:
+        # Malformed header or unknown charset: lossy per-part decode
+        decoded = []
+        for data, charset in email.header.decode_header(raw):
+            if isinstance(data, bytes):
+                try:
+                    decoded.append(data.decode(charset or "utf-8", errors="replace"))
+                except LookupError:
+                    decoded.append(data.decode("utf-8", errors="replace"))
+            else:
+                decoded.append(data)
+        return "".join(decoded)
 
 
 def _extract_text(msg):
diff --git a/tests/test_mcp_email_decode_header_spaces.py b/tests/test_mcp_email_decode_header_spaces.py
new file mode 100644
index 0000000..5ce7009
--- /dev/null
+++ b/tests/test_mcp_email_decode_header_spaces.py
@@ -0,0 +1,34 @@
+"""mcp email server _decode_header must not inject spaces between parts.
+
+email.header.decode_header returns plain-text runs WITH their surrounding
+whitespace (e.g. (b"Re: ", None)), so joining parts with " " produced a
+double space after "Re:" on every non-ASCII subject, a spurious space in
+"Name <addr>" senders, and violated RFC 2047 6.2 which requires whitespace
+between two adjacent encoded-words to be dropped.
+"""
+import pytest
+
+pytest.importorskip("mcp")
+
+import mcp_servers.email_server as es
+
+
+def test_prefix_then_encoded_word_single_space():
+    assert es._decode_header("Re: =?utf-8?b?SsOzc2U=?=") == "Re: J\u00f3se"
+
+
+def test_encoded_word_then_plain_text():
+    assert es._decode_header("=?utf-8?b?SsOzc2U=?= Smith") == "J\u00f3se Smith"
+
+
+def test_adjacent_encoded_words_join_without_space():
+    out = es._decode_header("=?iso-8859-1?q?Caf=E9?= =?utf-8?b?5pel5pys?=")
+    assert out == "Caf\u00e9\u65e5\u672c"
+
+
+def test_plain_ascii_header_unchanged():
+    assert es._decode_header("Weekly report") == "Weekly report"
+
+
+def test_empty_header():
+    assert es._decode_header("") == ""