-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
> I see that you fixed this? I am worried about writing null characters on a
> flush() command to outputstream, since it is not within the contract of
> flush()
> in java.io.OutputStream. Perhaps you should add another flush method to the
> connection for this (fullFlush() or something).
Its not working, actually... The problem is that we have to flush after
the name of the trailing field, or the server stalls, but if we flush
after the trailing field name, and it has a trailing field, and we write
nulls, then we write more than DataLength bytes, and the synchronization
is lost. This is that argument you and I had about CBC/ECB w/out pad
handling vs stream-cipher mode.
> I don't talk to governments.
I'd rather not, but I may have no choice.
Code is commited, if you want a look. The problem (2) is in RawMessage,
writeMessage(). Look for appropriate comments.
Scott
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.1 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE5D5EdpXyM95IyRhURAr/YAJ45oPb/6vQodBtEoQAq2TrmJ4IgvQCgqp5F
9FnPKiwL+CrBbFQe/D+mT0M=
=7NHF
-----END PGP SIGNATURE-----
_______________________________________________
Freenet-dev mailing list
Freenet-dev at lists.sourceforge.net
http://lists.sourceforge.net/mailman/listinfo/freenet-dev
