Test files › Email › unicode-headers.eml

Sample EML file (unicode headers)

Subject and display name encoded as RFC 2047 encoded-words (=?UTF-8?B?...?=) carrying CJK, Arabic and emoji.

↓ Download unicode-headers.eml
Filenameunicode-headers.eml
FormatEML
Size427 bytes (427 bytes)
MIME typemessage/rfc822
SHA-256b1fdce2f073aca773af37f6e5cfdac7e9c654199d488ed6ce264f0505a943ef7
What this file catchesHeader decoding. A client that prints the raw =?UTF-8?B?...?= instead of the decoded text, or mangles the display name in a reply.

Handle it in code

Each snippet below was run against this exact file with Python 3.14.4 and Node 24.15.0 when this page was built, and the output shown is what it printed.

Python

Parse with the modern email API (policy=default): it decodes encoded headers and bodies, and get_body picks the readable part. Treat attachment names as hostile: keep only the base name, replace unusual characters, and refuse names that run as programs. Compare the display name with the real address too, the oldest spoofing trick there is.

import email
import re
from email import policy
from email.utils import parseaddr

RUNS = (".exe", ".scr", ".bat", ".cmd", ".com", ".js", ".vbs", ".ps1", ".msi", ".jar", ".lnk")

with open("unicode-headers.eml", "rb") as f:
    msg = email.message_from_binary_file(f, policy=policy.default)

name, addr = parseaddr(str(msg["From"] or ""))
print("from:", addr or "(missing)", "| subject:", msg["Subject"] or "(missing)")
for header in ("Date", "Message-ID"):
    if msg[header] is None:
        print(f"warning: no {header} header")
if "@" in name and parseaddr(name)[1].lower() != addr.lower():
    print(f"warning: display name {name!r} is not the real sender")

body = msg.get_body(preferencelist=("plain", "html"))
text = body.get_content().strip() if body else ""
print(f"body: {body.get_content_type() if body else '(none)'} {text[:50]!r}")

for part in msg.iter_attachments():
    kind = part.get_content_type()
    if kind == "message/rfc822":
        print("attached message:", part.get_content()["Subject"])
        continue
    raw = part.get_filename() or ""
    base = raw.replace("\\", "/").rsplit("/", 1)[-1]
    safe = re.sub(r"[^\w.\- ]", "_", base).strip(". ") or "attachment"
    verdict = "blocked, it runs as a program" if safe.lower().endswith(RUNS) else "save as " + repr(safe)
    print(f"attachment: {kind} {raw!r}: {verdict}")
    if kind == "application/ms-tnef":
        print("  Outlook's TNEF wrapper: the real attachments are inside it")

Run against this file, it prints: from: ava.stone@example.test | subject: 請求書 — مرحبا — 🦊 quarterly body: text/plain 'Unicode lives in the headers, not the body.'

Related guides and tools

Sample EML and MSG email filesReserved domains for test emailCharacter encodingsCheck a file's real typeVerify its SHA-256

Download & verify

curl -fsSL -o "unicode-headers.eml" "https://files.hexaqa.com/email/unicode-headers.eml"
curl -fsSL "https://files.hexaqa.com/email/unicode-headers.eml" | sha256sum # expect: b1fdce2f073aca773af37f6e5cfdac7e9c654199d488ed6ce264f0505a943ef7

More email test files

plain-text.emlquoted-printable.emlspoofed-display-name.emlthreaded-reply.emlweb-archive.mhtwinmail-tnef.emlwith-attachment.emlapple-mail.emlx

← All Email files