ietf-smime
[Top] [All Lists]

RE: CAdES implementation. The algorithm to hash attributes

2007-07-10 07:08:58

Nick,

    This would be great. But why do we not add a requirement for
these
attributes to be transmitted in DER form like it was done in CMS for
signed and authenticated attributes?

In the case of the Archive Time-stamp, which may be added by the
recipient,
the data being time-stamped will include unsigned attributes and
content
which may not be encoded in DER when transmitted.  Also, in the cases
of
ES-C timestamps the data being time-stamped is also unsigned.  It might
be
feasible to re-encode the ES-C attributes in DER when applying a time-
stamp,
but it could be a major burden re-encoding data for a archive time-
stamp.

        So this is what I suggested. I certainly had not have in mind an
Archive Time-stamp attribute. Let's add another note (may be just
recommended, not required) for ES-C timestamps to reencode data being
time-stamped in DER before calculating a hash.

Pavel Smirnov
Crypto-Pro
Tel./Fax: +7 495 933-1168
WWW: http://www.CryptoPro.ru
e-mail: spv(_at_)CryptoPro(_dot_)ru


<Prev in Thread] Current Thread [Next in Thread>