[ENet-discuss] Incomplete/concatenated packets?

Sebastian Ahlman sebastian.ahlman at mbnet.fi
Thu Aug 18 08:09:40 PDT 2011


No, I have not seen it happen. But better to be safe than sorry, right?

Anyway, that's all I wanted to know, thanks! This lib just keeps getting
better and better.

//Sebastian

> Eh, I meant 'like with TCP' (HTTP for example).
> Ruud
>
> On Thu, Aug 18, 2011 at 4:08 PM, Ruud van Gaal <ruud at racer.nl> wrote:
>
>> Are you saying you're seeing this happen? I'd think not.
>> One UDP packet is really a contained entity, not a stream like with UDP.
>> So
>> 2 individual packets never merge on the other side, unless you
>> explicitly
>> would do the merging in your own code.
>>
>> Cheers,
>> Ruud
>>
>>
>> On Thu, Aug 18, 2011 at 3:14 PM, Sebastian Ahlman <
>> sebastian.ahlman at mbnet.fi> wrote:
>>
>>> Hey guys, first time poster here. Wonderful lib you have created.
>>>
>>> One (possibly stupid) question though. I could not find any answer to
>>> this.
>>>
>>> Is it possible that the receiving peer only receives part of a packet
>>> or
>>> multiple packets at the same time? Eg. if I send a packet with
>>>
>>> ENetPacket *packet = enet_packet_create ("hello", strlen ("hello") + 1,
>>> ENET_PACKET_FLAG_RELIABLE);
>>>
>>> ...and then send another one from the same sender to the same receiver
>>> with...
>>>
>>> ENetPacket *packet = enet_packet_create ("world", strlen ("world") + 1,
>>> ENET_PACKET_FLAG_RELIABLE);
>>>
>>> ...is it possible that the receiving peer only receives "hel"? Or
>>> "helloworld"?
>>> What about "hellowo"?
>>>
>>> Also, does the ENET_PACKET_FLAG_RELIABLE affect this? I know that if I
>>> don't
>>> specify the flag there is no guarantee that the packet will make it's
>>> way
>>> through at all, but what about partial or "concatenated" packets?
>>>
>>> The reason I am asking is because I have implemented a simple RPC
>>> system
>>> on top
>>> of enet, and currently I just blindly expect one ENetPacket to be
>>> exactly
>>> one
>>> RPC command. Is this safe to do?
>>>
>>> Thanks!
>>>
>>> //Sebastian
>>>
>>> _______________________________________________
>>> ENet-discuss mailing list
>>> ENet-discuss at cubik.org
>>> http://lists.cubik.org/mailman/listinfo/enet-discuss
>>>
>>
>>
> _______________________________________________
> ENet-discuss mailing list
> ENet-discuss at cubik.org
> http://lists.cubik.org/mailman/listinfo/enet-discuss
>




More information about the ENet-discuss mailing list