Learn how easy it is to sync an existing GitHub or Google Code repo to a SourceForge project! See Demo

Close

#7 Allow encryption/signing of vm mail buffers with attachments

open
nobody
None
5
2003-12-02
2003-12-02
Jon Schewe
No

When a file is attached to a mail buffer in mail-mode with
vm-mime-attach-file the file isn't actually attached, instead
something is attached that is expanded on send to the
mime encoded version. This expansion is done by
vm-mime-encode-composition. It'd be really nice if
mailcrypt could call this before messages are signed or
encrypted so that they can properly be sent without the
user needing to manually do this.

Discussion

  • Jon Schewe
    Jon Schewe
    2003-12-02

    Logged In: YES
    user_id=209980

    A workaround for this is to add pre-encryption and pre-signature
    hooks like so:
    (add-hook 'mc-pre-encryption-hook
    'vm-mime-encode-composition)
    (add-hook 'mc-pre-signature-hook
    'vm-mime-encode-composition)

     
  • Jon Schewe
    Jon Schewe
    2003-12-02

    Logged In: YES
    user_id=209980

    Oops, that workaround doesn't work like I hoped. The MIME
    headers get munged that way. Sorry.

     
  • Logged In: YES
    user_id=275595

    I don't have a mailcrypt solution to this problem, but I
    *do* have a workaround. I was having problems where my
    FCC's weren't being done until after the mc-encrypt, by
    which time the buffer had been turned into unreadable
    gibberish. Here's the advice I used to fix this problem:

    (defadvice mc-encrypt-region (before fcc-first activate)
    "Make sure I don't forget to FCC the buffer before munging
    it with encryption."
    (fcc-by-hand))

    You could make an advice form like this, but with
    vm-mime-encode-composition. You'd have to advise both the
    signing and the encrypting functions, though.

     
  • Jon Schewe
    Jon Schewe
    2003-12-15

    Logged In: YES
    user_id=209980

    Actually that workaround suffers from the same problem as my
    previous one. The MIME headers are wrong.