예제 #1
0
        //**********************************************************************
        //**********************************************************************
        //**********************************************************************
        // Receive message handlers

        public override void processRxMsg(ByteContent aMsg)
        {
            BaseMsg tRxMsg = (BaseMsg)aMsg;

            // Message jump table based on message type.
            // Calls corresponding specfic message handler method.
            switch (tRxMsg.mMessageType)
            {
            case MsgIdT.cTestMsg:
                processRxMsg((TestMsg)tRxMsg);
                break;

            case MsgIdT.cStatusRequestMsg:
                processRxMsg((StatusRequestMsg)tRxMsg);
                break;

            case MsgIdT.cStatusResponseMsg:
                processRxMsg((StatusResponseMsg)tRxMsg);
                break;

            case MsgIdT.cDataMsg:
                processRxMsg((DataMsg)tRxMsg);
                break;

            default:
                Prn.print(Prn.ThreadRun1, "NetworkThread.processRxMsg UNKNOWN");
                break;
            }
        }
예제 #2
0
        //--------------------------------------------------------------------------
        // For variable content messages, the message length cannot be known until
        // the entire message has been written to a byte buffer. Therefore, the
        // message header cannot be written to a byte buffer until the entire
        // message has been written and the length is known.
        //
        // The procedure to write a message to a byte buffer is to skip over the
        // buffer segment where the header is located, write the message payload
        // to the buffer, set the header message length based on the now known
        // payload length, and write the header to the buffer.
        //
        // These are called explicitly by inheriting messages at the
        // beginning and end of their copyToFrom's to manage the headers.
        // For "get" operations, headerCopyToFrom "gets" the header and
        // headerReCopyToFrom does nothing. For "put" operations,
        // headerCopyToFrom stores the buffer pointer and advances past where the
        // header will be written and headerReCopyToFrom "puts" the header at the
        // stored position. Both functions are passed a byte buffer pointer to
        // where the copy is to take place. Both are also passed a MessageContent
        // pointer to where they can get and mMessageType
        // which they transfer into and out of the headers.
        //--------------------------------------------------------------------------

        public void headerCopyToFrom(Ris.ByteBuffer aBuffer, BaseMsg aParent)
        {
            //---------------------------------------------------------------------
            // Instances of this class are members of parent message classes.
            // A call to this function should be the first line of code in a
            // containing parent message class's copyToFrom. It performs pre-copyToFrom
            // operations. It's purpose is to copy headers to/from byte buffers. The
            // corresponding function headerReCopyToFrom should be called as the last
            // line of code in the containing message class' copyToFrom. Lines of code
            // in between should copy individual data elements into/out of the buffer.

            //---------------------------------------------------------------------
            // for a "copy to" put
            //
            // If this is a "copy to" put operation then the header copy will actually
            // be done by the headerReCopyToFrom, after the rest of the message has been
            // copied into the buffer. This is because some of the fields in the header
            // cannot be set until after the rest of the message has been put into the
            // buffer. (You don't know the length of the message until you put all of
            // the data into it, you also can't compute a checksum). This call stores
            // the original buffer position that is passed to it when it is called for
            // later use by the headerReCopyToFrom. The original buffer position points
            // to where the header should be copied. This call then forward advances
            // the buffer to point past the header. Forward advancing the buffer
            // position to point just after where the header should be is the same as
            // doing a pretend copy of the header. After this pretend copy of the
            // header, the buffer position points to where the data should be put into
            // the buffer.
            //
            // Store the original buffer position for later use by the
            // headerReCopyToFrom and advance the buffer position forward
            // to point past the header.

            if (aBuffer.isCopyTo())
            {
                // Store the buffer parameters for later use by the
                // headerReCopyToFrom
                mInitialPosition = aBuffer.getPosition();
                mInitialLength   = aBuffer.getLength();

                // Advance the buffer position to point past the header.
                aBuffer.forward(Header.cLength);
            }

            //---------------------------------------------------------------------
            // for a "copy from" get
            //
            // If this is a "copy from" get operation then copy the header from the
            // buffer into the header member. Also set the message content type from
            // the variable datum id

            else
            {
                // Copy the buffer content into the header object.
                copyToFrom(aBuffer);
                // Set the message content type.
                aParent.mMessageType = mMessageIdentifier;
            }
        }
예제 #3
0
        //--------------------------------------------------------------------------
        // Preprocess a message before it is sent

        public override void processBeforeSend(Ris.ByteContent aMsg)
        {
            BaseMsg tMsg = (BaseMsg)aMsg;

            if (tMsg.mHeader.mSourceId == 0)
            {
                tMsg.mHeader.mSourceId = mSourceId;
            }
        }
예제 #4
0
        public void headerReCopyToFrom(Ris.ByteBuffer aBuffer, BaseMsg aParent)
        {
            // If this is a put operation then this actually copies the header into
            // the buffer.
            // This sets some header length parameters and copies the header into the
            // buffer position that was stored when headerCopyToFrom was called.

            if (aBuffer.isCopyTo())
            {
                // Store the buffer parameters for later use by the
                // headerReCopyToFrom
                int tFinalPosition = aBuffer.getPosition();
                int tFinalLength   = aBuffer.getLength();

                // Get message parameters from parent
                mMessageIdentifier = aParent.mMessageType;
                mMessageLength     = aBuffer.getLength();

                // Restore buffer parameters
                // to the initial position
                aBuffer.setPosition(mInitialPosition);
                aBuffer.setLength(mInitialPosition);

                // Copy the adjusted header into the buffer'
                // at the original position
                copyToFrom(aBuffer);

                // Restore buffer parameters
                // to the final position
                aBuffer.setPosition(tFinalPosition);
                aBuffer.setLength(tFinalPosition);
            }
            else
            {
            }
        }