VmPipe Usage Question

[boB Gage <bg...@ekosystems.com>]

I have searched far and wide for an answer on this before resorting to 
bothering anyone...
but I can't quite work this out on my own.

We're using MINA classes to implement serial and socket connections to 
external devices and decode the data streams from these devices.   I'm 
trying to use VmPipe classes to create a test mode that fakes an external 
device, using data files to define specific device behaviors.

So far it's working fairly well, with one pretty major glitch.    The 
server-side (the fake hardware device) of the VmPipe is receiving 
byte-by-byte data from the client-side (our normal device monitor) and can 
deal with it accordingly.

Problem is in the other direction.   The server-side uses 
"session.write(b)" to write a byte at a time to the stream (at properly 
faked baud rate even).    But instead of those accumulated bytes feeding 
into our MessageDecoderAdapter descendants as incoming data would from the 
hardware, they are getting picked up by our IoHandlerAdapter descendant as 
individual Byte objects (messageReceived() method).

How would I get these bytes into the proper decode chain to fully test the 
low-level device data decoding (as per the DemuxingProtocolCodecFactory 
definition from the monitor under test)???

Thanks in advance!!
boB Gage
eko systems, Inc.

Re: VmPipe Usage Question

[Mark Webb <el...@gmail.com>]

Did you try CumulativeProtocolDecoder ?

I had a very similar problem once and once I switched over to this class
everything worked great.


On Tue, Jul 29, 2008 at 2:08 PM, boB Gage <bg...@ekosystems.com> wrote:

> I have searched far and wide for an answer on this before resorting to
> bothering anyone...
> but I can't quite work this out on my own.
>
> We're using MINA classes to implement serial and socket connections to
> external devices and decode the data streams from these devices.   I'm
> trying to use VmPipe classes to create a test mode that fakes an external
> device, using data files to define specific device behaviors.
>
> So far it's working fairly well, with one pretty major glitch.    The
> server-side (the fake hardware device) of the VmPipe is receiving
> byte-by-byte data from the client-side (our normal device monitor) and can
> deal with it accordingly.
>
> Problem is in the other direction.   The server-side uses
> "session.write(b)" to write a byte at a time to the stream (at properly
> faked baud rate even).    But instead of those accumulated bytes feeding
> into our MessageDecoderAdapter descendants as incoming data would from the
> hardware, they are getting picked up by our IoHandlerAdapter descendant as
> individual Byte objects (messageReceived() method).
>
> How would I get these bytes into the proper decode chain to fully test the
> low-level device data decoding (as per the DemuxingProtocolCodecFactory
> definition from the monitor under test)???
>
> Thanks in advance!!
> boB Gage
> eko systems, Inc.
>