1996-04-02 - Re: gnutar + pgp filter mode

Header Data

From: Derek Atkins <warlord@MIT.EDU>
To: danisch@ira.uka.de (Hadmut Danisch)
Message Hash: 37f7d15091e6d52209f410c04297c0f09a2da58e88cd6f0961faafd5565a23c9
Message ID: <199604011916.OAA18585@toxicwaste.media.mit.edu>
Reply To: <199604011522.RAA09571@elysion.eiss.ira.uka.de>
UTC Datetime: 1996-04-02 07:23:11 UTC
Raw Date: Tue, 2 Apr 1996 15:23:11 +0800

Raw message

From: Derek Atkins <warlord@MIT.EDU>
Date: Tue, 2 Apr 1996 15:23:11 +0800
To: danisch@ira.uka.de (Hadmut Danisch)
Subject: Re: gnutar + pgp filter mode
In-Reply-To: <199604011522.RAA09571@elysion.eiss.ira.uka.de>
Message-ID: <199604011916.OAA18585@toxicwaste.media.mit.edu>
MIME-Version: 1.0
Content-Type: text/plain


> The only problem is that this doesn't work well for
> larger amounts of data, because pgp reads in all
> data before starting with its work.
> 
> Perhaps a future release of pgp will have a real filter
> mode.

The problem is not with PGP but with the PGP message formatsa.  With
the current message formats you need to know the size of data before
you can output the header that precedes it.  As a result, whole
messages must somehow be buffered before PGP can output them.

The fix is to create new, one-pass packets, which will allow PGP to
generate messages in a single pass.  These new packet formats are
still in the design stage and will probably not be supported in PGP
3.0.

-derek





Thread