Hi,

I'm trying to use netty to implement a multiplexer/demultiplexer. Think 
multiple streams travelling together over a single socket, then getting 
spread out into individual sockets in my code.

I'm basing it on the example socks proxy code, which seems to have the 
basics in place (server and client together). However, I'm trying to figure 
out how best to deal with a couple of requirements:

The incoming traffic (call it downstream) has a stream number, which is 
used to identify which stream it belongs to, followed by some flags, 
sequence numbers, a length byte, and 60 bytes of data (of which "length" 
are usable). Depending on the flags, I will have to open a new client 
socket, and link it to the relevant stream number.

It seems to me that the server pipeline will have a simple handler that 
essentially looks up the channel/pipeline associated with a particular 
stream, or, if it does not exist, creates it. Each "client pipeline" (call 
it upstream) will have a first handler that knows how to process the flags, 
verify the sequence numbers, and extract the actual data, then write that 
data into the pipeline. It seems like this could be implemented as a 
MessageToMessageCodec, right? Packet comes in, ByteBuf of actual data goes 
out, to be written to the stream.

In the reverse direction, any data coming from upstream will have to be 
broken up into chunks of max 60 bytes, then put into a "Packet" that simply 
gets written to the downstream channel.

Does this make sense? Does it seem like a reasonable way of structuring the 
pipelines?

Thanks for your help.

Rogan

-- 
You received this message because you are subscribed to the Google Groups 
"Netty discussions" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/netty/7e3124d3-9791-4d4c-8d6e-462735a473b4%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to