That, sir, is a great idea. About the sleep though, doesn't the server send out a response that the client can wait for? If this is all TCP/IP, I'm fairly certain it does. We could look at an open-source IRC client and look how they do it, and I don't think it will be done using sleep().
On 7 August 2010 14:56, ProgVal <[email protected]> wrote: > We also can replace the ident by "glob2-<version>", it'll make easier to > everybody to know someone's version > > 2010/8/7 ProgVal <[email protected]> > > Yes, it can be more short (0.1 sec may be enough). The essential is that >> the server must understand the client will quit, before the client close the >> socket. >> >> 2010/8/6 Stéphane Magnenat <[email protected]> >> >>> On 06/08/10 15:54, ProgVal wrote: >>> >>> Hello, >>>> >>>> When disconnecting, glob2 IRC client just closed the socket, but it is >>>> not the properest method given by the IRC protocol. >>>> >>>> The better is to use the QUIT command, fellowed by a quit message. This >>>> quit message is display on the most of the networks, but not on OFTC, >>>> so, we'll just see >>>> >>> >>> The sleep() delay is bad. Is the sleep necessary? If so, is it required >>> to be so long? If so, we should do it later in IRC::step(). >>> >>> Have a nice day, >>> >>> Stéphane >>> >>> -- >>> http://stephane.magnenat.net >>> >>> _______________________________________________ >>> glob2-devel mailing list >>> [email protected] >>> http://lists.nongnu.org/mailman/listinfo/glob2-devel >>> >> >> > > _______________________________________________ > glob2-devel mailing list > [email protected] > http://lists.nongnu.org/mailman/listinfo/glob2-devel > >
_______________________________________________ glob2-devel mailing list [email protected] http://lists.nongnu.org/mailman/listinfo/glob2-devel
