On Mon, Jun 10, 2002 at 07:17:05PM -0400, Theo Van Dinter wrote: | On Mon, Jun 10, 2002 at 04:02:24PM -0500, Derrick 'dman' Hudson wrote: | > spamd would be the one giving the output, but it doesn't _have_ to | > output the whole message, just the headers that have changed. spamc | > wouldn't change in that respect. | | You're assuming the body isn't getting mangled.
Yes. In fact, it isn't on my server. As I already mentioned, those using Marc's sa-exim patch only mangle/add certain headers. All other output from spamc is ignored. In fact, the body of the message is never read() from the pipe. With that in mind, the proposal would fit quite nicely there. | What about the body report? Either use the headers report, or read back just enough to catch that. It would be in the beginning of the return output anyways. | What about making the spam a rfc822 attachment when score > | required? (that one is not only a complete header change, it's a | top and bottom body change...) Pay your money and take your choice. (as a certain prof. of mine likes to say) It all depends on which is more important to you -- reduced data shuffling vs. total control over data modification. Maybe these tradeoffs could be turned into mutually exclusive options? Some people might prefer one style over the other. However, I'm not coding it so I'm not the one you need to worry about ;-). | It comes down to: Stupid client/Smart server, Smart client/Stupid server, | something in between. Right now, we're in the first stage. spamc is | pretty stupid, and spamd is pretty smart. I personally would like to keep | it that way as much as possible. I agree. (though spamd is kinda stupid right now too -- it doesn't do any smart _changing_ of the data as it passes through, apart from addition of some RFC822 headers) Actually, sa-exim is a rather smart client, as is necessary for intelligent processing of the answer SA gives. -D -- "Don't use C; In my opinion, C is a library programming language not an app programming language." - Owen Taylor (GTK+ developer) GnuPG key : http://dman.ddts.net/~dman/public_key.gpg
msg06178/pgp00000.pgp
Description: PGP signature
