[ENet-discuss] Tracking Connections

Lee Salzman lsalzman1 at cox.net
Thu Aug 26 16:35:29 PDT 2010


ENET_PACKET_FLAG_RELIABLE is for reliable and sequenced, there is no way 
to get reliability without sequencing in ENet.
ENET_PACKET_FLAG_UNSEQUENCED is for unreliable and unsequenced 
currently. It gets ignored if reliable is used.
If you simply pass in neither of these flags, you get UNreliable and 
unsequenced, i.e. pass in 0 for the flags.

These flags are passed in to enet_packet_create(), which determines the 
type of packet.

Lee

On 08/26/2010 02:26 PM, Nicholas J Ingrassellino wrote:
> I am pretty new to the *ENet* library (and network programming in 
> general). I have done a few test games with *HawkNL* but that is about 
> all and was at least one year ago.
>
> With the help of the *ENet* tutorial on the site I have built a server 
> and client that can pass data back and forth (I was so happy). The 
> tutorial, however, has left at least two things unclear to me:
>
>    1. How can I send data reliably but unsequenced? How can I send
>       data unreliably but sequenced?
>    2. How can I track connections when more than one client is connected?
>
> I know there are /ENET_PACKET_FLAG_RELIABLE/ and 
> /ENET_PACKET_FLAG_UNSEQUENCED/ via the headers. The tutorial does not 
> when mention /ENET_PACKET_FLAG_UNSEQUENCED/ nor its expected behavior 
> so I am left to assume. Judging by the variable names I think these 
> are meant to be used in bit flags? I have also noticed that 
> /ENetEvent->peer/ has a few members that might be what I am looking 
> for when tracking connections (/outgoingPeerID/, /incomingPeerID/, 
> /connectID/, /outgoingSessionID/, /incomingSessionID/) but I am 
> unclear on which of these to use if any.
>
> ------------------------------------------------------------------------
>
> Nicholas J Ingrassellino
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cubik.org/pipermail/enet-discuss/attachments/20100826/879f7bc4/attachment.html>


More information about the ENet-discuss mailing list