Merge pull request 'Sanitize folded email headers before SMTP send' (#3) from fix/sanitize-folded-headers into main

hearth/mailer.py

@@ -8,17 +8,21 @@ hearth.yml.
 from __future__ import annotations
 
 import logging
+import re
 import smtplib
-import uuid
 from dataclasses import dataclass
 from email.message import EmailMessage
 from email.utils import make_msgid
 
-from hearth.config import Config, ConfigError, Participant
+from hearth.config import Config, Participant
 from hearth.settings import settings
 
 log = logging.getLogger("hearth.mailer")
 
+# Real IMAP headers often arrive folded (CRLF + whitespace). EmailMessage
+# rejects raw newlines when assigning header values, so unfold first.
+_HEADER_WS = re.compile(r"[\r\n\t ]+")
+
 
 class SendBlocked(RuntimeError):
     pass
@@ -30,6 +34,23 @@ class SendResult:
     dry_run: bool
 
 
+def sanitize_header(value: str | None) -> str:
+    """Collapse folded whitespace so headers are safe for EmailMessage."""
+    if not value:
+        return ""
+    return _HEADER_WS.sub(" ", value).strip()
+
+
+def _reply_references(prior: str | None, in_reply_to: str | None) -> str:
+    """Build a single-line References chain ending with the parent Message-ID."""
+    refs = sanitize_header(prior)
+    parent = sanitize_header(in_reply_to)
+    if parent and parent not in refs:
+        refs = f"{refs} {parent}".strip() if refs else parent
+    return refs
+
+
+
 def _connect(sender: Participant) -> smtplib.SMTP:
     if sender.smtp_port == 465:
         smtp = smtplib.SMTP_SSL(sender.smtp_host, sender.smtp_port, timeout=20)
@@ -62,6 +83,9 @@ def send(
         raise SendBlocked(f"recipient email for '{recipient_key}' does not match allow-list")
 
     message_id = make_msgid(domain=sender.email.split("@", 1)[-1])
+    subject = sanitize_header(subject) or "(no subject)"
+    parent_id = sanitize_header(in_reply_to) or None
+    refs = _reply_references(references, parent_id) or None
 
     if settings.dry_run or not settings.enabled:
         log.info(
@@ -78,10 +102,10 @@ def send(
     msg["To"] = recipient.email
     msg["Subject"] = subject
     msg["Message-ID"] = message_id
-    if in_reply_to:
-        msg["In-Reply-To"] = in_reply_to
-    if references:
-        msg["References"] = references
+    if parent_id:
+        msg["In-Reply-To"] = parent_id
+    if refs:
+        msg["References"] = refs
     msg.set_content(body)
 
     smtp = _connect(sender)

hearth/watcher.py

@@ -64,15 +64,15 @@ def _handle_message(config: Config, store: Store, mailbox_key: str, raw: bytes,
     if sender_key == mailbox_key:
         return  # our own mailbox talking to itself, ignore
 
-    message_id = _decode(msg.get("Message-ID")).strip()
+    message_id = mailer.sanitize_header(_decode(msg.get("Message-ID")))
     if not message_id:
         return
     if store.already_seen(message_id) or store.is_reply_queued(message_id):
         return
 
-    in_reply_to = _decode(msg.get("In-Reply-To")).strip() or None
-    subject = _decode(msg.get("Subject")).strip() or "(no subject)"
-    references = _decode(msg.get("References")).strip() or in_reply_to
+    in_reply_to = mailer.sanitize_header(_decode(msg.get("In-Reply-To"))) or None
+    subject = mailer.sanitize_header(_decode(msg.get("Subject"))) or "(no subject)"
+    references = mailer.sanitize_header(_decode(msg.get("References"))) or in_reply_to
 
     thread_id = store.thread_id_for(in_reply_to)
     store.record_message(

tests/test_mailer_guard.py

@@ -85,3 +85,54 @@ def test_is_known_email_is_exact_match_only(config):
     # A lookalike/subdomain address must never resolve to a real participant.
     assert config.is_known_email("alerts@example.com.evil.com") is None
     assert config.is_known_email("not-friend@example.com") is None
+
+
+def test_sanitize_header_unfolds_folded_references():
+    folded = "<a@x>\r\n <b@y>\n\t<c@z>"
+    assert mailer.sanitize_header(folded) == "<a@x> <b@y> <c@z>"
+    assert "\n" not in mailer.sanitize_header(folded)
+    assert mailer.sanitize_header(None) == ""
+    assert mailer.sanitize_header("  hi  ") == "hi"
+
+
+def test_reply_references_appends_parent_once():
+    prior = "<root@x>\r\n <mid@x>"
+    parent = "<leaf@x>"
+    assert mailer._reply_references(prior, parent) == "<root@x> <mid@x> <leaf@x>"
+    assert mailer._reply_references(parent, parent) == "<leaf@x>"
+    assert mailer._reply_references(None, parent) == "<leaf@x>"
+    assert mailer._reply_references(None, None) == ""
+
+
+def test_send_accepts_folded_references_without_raising(config, monkeypatch):
+    """Regression: EmailMessage rejects CR/LF in header values."""
+    settings.dry_run = False
+    settings.enabled = True
+
+    captured = {}
+
+    class FakeSMTP:
+        def send_message(self, msg):
+            captured["msg"] = msg
+
+        def quit(self):
+            pass
+
+    monkeypatch.setattr(mailer, "_connect", lambda sender: FakeSMTP())
+
+    folded = "<root@example.com>\r\n <parent@example.com>"
+    result = mailer.send(
+        config,
+        sender_key="alerts",
+        recipient_key="friend",
+        subject="Re: hello\n there",
+        body="hi",
+        in_reply_to="<parent@example.com>",
+        references=folded,
+    )
+    assert result.dry_run is False
+    assert "\n" not in captured["msg"]["Subject"]
+    assert "\n" not in captured["msg"]["References"]
+    assert "\r" not in captured["msg"]["References"]
+    assert captured["msg"]["In-Reply-To"] == "<parent@example.com>"
+    assert captured["msg"]["References"] == "<root@example.com> <parent@example.com>"