If a current tarball found its way onto a webserver I could try to integrate ratelimiting into it.
But an implimentation question: currently I did ratelimiting by making the SNAC sending call sleep() until sending the packet would not go over the limit. Is there a better way to do this? Would it be better to span a new thread (or five, one for each rateclass) to send packets out of a queue? I suppose the real question is whether the calls to sendSNAC() are themselves multithreaded, as causing the whole thing to sleep() is a bad idea. Does a delay in sendSNAC() block other things happening internally? ~Chris From [EMAIL PROTECTED] Sat Jan 29 07:58:59 2005 From: [EMAIL PROTECTED] (Daniel Henninger) Date: Sat Jan 29 07:58:59 2005 Subject: [py-transports] Ratelimiting implementation In-Reply-To: <[EMAIL PROTECTED]> References: <[EMAIL PROTECTED]> Message-ID: <[EMAIL PROTECTED]> I have put up "current tarballs" at: http://www.vorpalcloud.org/tarballs/ As for the multithreaded sendSNAC question, I'm not 100% sure. I tried throwing a sleep in there to see how it went and it seems to behave in a reasonable manner. I really don't have a good answer for you though. =/ Daniel -- "The most addictive drug in the world is music." - The Lost Boyz > If a current tarball found its way onto a webserver I could try to > integrate ratelimiting into it. > > But an implimentation question: currently I did ratelimiting by making the > SNAC sending call sleep() until sending the packet would not go over the > limit. Is there a better way to do this? Would it be better to span a new > thread (or five, one for each rateclass) to send packets out of a queue? > > I suppose the real question is whether the calls to sendSNAC() are > themselves multithreaded, as causing the whole thing to sleep() is a bad > idea. Does a delay in sendSNAC() block other things happening internally? > > ~Chris > _______________________________________________ > py-transports mailing list > [EMAIL PROTECTED] > http://mail.jabber.org/mailman/listinfo/py-transports >
