Well, we have a spell checker, but not an intelligence checker... > The problem here is that the player is a command that, as the command is how about "the player enters a command and, as the command" :-)
Tony ----- Original Message ----- From: "Tony Kuiper" <[EMAIL PROTECTED]> To: <[email protected]> Sent: Tuesday, January 21, 2003 12:44 PM Subject: Re: question on ch->wait > First of all, isn't this kinda like saying "I shot my foot off so I want > to make a gun that won't hurt me if I shoot it?" Anyways, I'm not here to > judge morality or intelligence, but to help. > The problem here is that the player is a command that, as the command is > executed, it adds to ch->wait. Before ch->wait expires, they send several > more commands. So their ch->desc->inbuf is filling up with commands. The > game_loop_* function is responsible for pulling the commands out of the > player's inbuf, but only when ch->wait has expired. So if they spam the > same command, then each command will add to ch->wait, the game will wait > until ch->wait expires, process the next command which adds to ch->wait, > etc... > There are several solutions - > 1. Tell your players "Tough luck. You shoot yourself in the foot, you > deal with it." This will probably decrease your player count. :-) > 2. Remove ch->wait. Who want to wait, anyways? (Just kidding - this > would seriously imbalance the game!) > 3. Add a hack to game_loop_* to check through ch->descriptor->inbuf for > the "clear" command, and if it's found, clear the inbuf. This is a genuine > hack, and as such can greatly add to the complexity of the code. > 4. Implement a command queue system. Every time the player sends a > command, add it to the queue. Then, each time through game_loop_*, look > through the command queue, and if there are any commands that don't rely on > ch->wait, execute them. Also, get the next command and check to see if it > requires ch->wait. (Or possibly ch->wait[WAIT_FIGHTING] and > ch->wait[WAIT_GLOBAL_COMMINUCATION], etc...) If ch->wait has expired, > perform it, otherwise, don't perform any player commands. > A word of warning - this kind of clear command can greatly unbalance the > game - for example, what happens then the player sets a client based alias > to "kick scout" and "clear" and "kick scout" and "clear", etc.? (Hint, > do_kick will deal damage, do_clear clears ch->wait and the inbuf, do_kick > will deal damage, etc.. all in the space of time it takes the game engine to > loop, effectively giving the player the power to deal major damage above and > beyond their normal capability.) > The functions that control this are all stored in comm.c: game_loop_* > and read_from_buffer(). d->inbuf[] contains all text the player types. > d->incomm[] contains the current command moved out of d->inbuf[]. > > Questions? Comments? > > Tony > > ----- Original Message ----- > From: <[EMAIL PROTECTED]> > To: <[email protected]> > Sent: Tuesday, January 21, 2003 11:33 AM > Subject: question on ch->wait > > > > A few of players would like to a command to clear everything they typed > > previously, an example: > > kick scout > > kick scout > > kick scout > > kick scout > > kick scout > > > > Whoops 3 more scouts walked in, I wanna run now! But that is impossible > > because I will be kicking the first scout for the next minute or so. So I > > wanted to create a command called 'clear' or something that would delete > > everything that they previously typed in. I was looking at the code and I > > have no idea where the commands are stored, or how it determines if the > > character is in a wait state or not. Can anyone point me in the right > > direction? I would search the archives but they aren't working right now. > > THANKS > > > > -- > > ROM mailing list > > [email protected] > > http://www.rom.org/cgi-bin/mailman/listinfo/rom > > > > -- > ROM mailing list > [email protected] > http://www.rom.org/cgi-bin/mailman/listinfo/rom >
