Zum Inhalt springen
uencoder

Analysieren a multipart MIME message and extract each part

Break ein multipart/form-Daten oder multipart/mixed MIME Nachricht in its constituent headers und bodies für Inspektion oder reprocessing.

  1. 1

    Multipart MIME bundles several parts (text, attachments, form fields) separated by a boundary string. Each part has its own headers and body. Parsing requires splitting on the boundary and decoding each part's Content-Transfer-Encoding.

  2. 2
    MIME-Dekodierer →

    Paste the full multipart message. The tool splits on the boundary, extracts each part's headers and body, and applies the transfer encoding (base64, quoted-printable, 7bit).

  3. 3
    MIME-Typen-Referenz →

    Look up the Content-Type of each extracted part to understand how to render or save it (e.g. image/jpeg → save as binary, text/plain → display).

  4. 4
    Base64-Dekodierer →

    If a part's body is still Base64 after MIME decode (nested encoding), decode it here to recover the original bytes.

Häufige Fragen

Where do I find the boundary string?
It is declared in the top-level Content-Type header as boundary="...". The actual separators in the body are prefixed with two dashes (--boundary), and the final separator is followed by two dashes.
Why are some parts empty?
Multipart/form-data often includes empty parts for unset form fields, or a leading/trailing empty segment from the boundary split. Filter parts with empty bodies after parsing.

Workflows, die diese Aufgabe lösen

Verwandte Tools