On Sun, 6 May 2001, Adam Langley wrote:

> Notes from an IRC conversation between Tavin and myself:
> 
> * partSize =  MIN (len (data + padding), MAX (len (data + padding) >>
>   7, 16384) )
> * to pad data to a power of 2, you repeat the data
No.  This lends an attack as well, since the attacker can try and assume
that several of the last blocks of the file are the same. 

> * encryption key = make_key ( hash (data + padding) )
> * Storable.Document-header = DocHeadData zero padded to power of 2
> * DocHeadData = <length of key: 2 bytes><key><length of data: number>
>   <length of metadata: number>
> * The length of the key MUST equal the keysize of the cipher (e.g. 16
>   bytes for a 128-bit cipher)
> * DocHeadData number = <length of number: 2 bytes><number: big endian>
>   where length is the smallest length that the number will fit into
>   subject to: The minumum length is 2 AND The length may not be odd.
  Why not make DocHeadData number an MPI so that people don't have to code
yet another number representation?  



_______________________________________________
Devl mailing list
Devl at freenetproject.org
http://lists.freenetproject.org/mailman/listinfo/devl

Reply via email to