TypeError in ported_string when parsing headers as email.header.Header objects

[AndreasLF]

Describe the bug
The ported_string utility function raises a TypeError when it receives an instance of email.header.Header.

In certain edge cases (specifically involving "dirty" Outlook conversions or specific email library configurations), the parser extracts an email.header.Header object instead of a string. Because ported_string does not check for this object type, the parser crashes when it attempts to sanitize these complex headers.

To Reproduce

1. Create a minimal Python script that simulates the edge case by creating a Header object manually.
2. Pass this object to ported_string.

from email.header import Header
from mailparser.utils import ported_string

# This simulates the return value of p.get() in complex parsing scenarios
h = Header("Test Filename.pdf", charset="utf-8")

# This causes a TypeError because ported_string expects str or bytes
ported_string(h)

This causes the followingTypeError because ported_string expects str or bytes:
TypeError: decoding to str: need a bytes-like object, Header found

Expected behavior
ported_string should detect that the input is an email.header.Header object and convert it to a string (using str(obj) or six.text_type(obj)) before returning.

Raw mail
I cannot provide the full original .eml file due to GDPR restrictions. Additionally, creating a synthetic file that forces the Python standard email library to return a Header object is non-deterministic, as it often depends on specific system locales and Python versions.

Relevant Header Context: The crash happens specifically when parsing the Content-Disposition header (see traceback). In my environment, the header contained mixed encoding/dirty bytes (likely from an Outlook conversion) similar to: Content-Disposition: attachment; filename="Report \r\n\t\x96\x96\x96 Final.pdf"

While mailparser usually receives a string here, the traceback confirms that in this specific case, the email library returned an email.header.Header object, which mailparser then failed to handle.

Looking at the docstring for decode_header from the Python standard email library, it seems that the library explicitly supports Header objects in this flow:

header may be a string that may or may not contain RFC2047 encoded words,
or it may be a Header object.

This indicates that mailparser (specifically ported_string) needs to handle Header objects defensively, as they are a valid state within the email library ecosystem.

Environment:

• OS: MacOS 15.7.3 (Python 3.12.8)
• Docker: no
• mail-parser version 4.1.4

Traceback

return mailparser.parse_from_bytes(inner_email_bytes)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/user/.virtualenvs/project/lib/python3.11/site-packages/mailparser/core.py", line 113, in parse_from_bytes
    return MailParser.from_bytes(bt)
           ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/user/.virtualenvs/venv/lib/python3.11/site-packages/mailparser/core.py", line 236, in from_bytes
    return cls(message)
           ^^^^^^^^^^^^
  File "/Users/user/.virtualenvs/venv/lib/python3.11/site-packages/mailparser/core.py", line 132, in _init_
    self.parse()
  File "/Users/user/.virtualenvs/venv/lib/python3.11/site-packages/mailparser/core.py", line 395, in parse
    content_disposition = ported_string(p.get("content-disposition"))
                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/user/.virtualenvs/venv/lib/python3.11/site-packages/mailparser/utils.py", line 88, in wrapper
    return normalize("NFC", func(*args, **kwargs))
                            ^^^^^^^^^^^^^^^^^^^^^
  File "/Users/user/.virtualenvs/venv/lib/python3.11/site-packages/mailparser/utils.py", line 122, in ported_string
    return six.text_type(raw_data, encoding)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
TypeError: decoding to str: need a bytes-like object, Header found