general clarity of data on the wire.

1 post / 0 new
Offline
Last seen: 5 years 3 months ago
Joined: 01/21/2014
Posts: 6
general clarity of data on the wire.

Hello, I am still building my understanding of DDS- RTPS 2.1 data. I thought I had a general conception but an input I received on it -> made me want to ask for clarification.

My understanding is that the first part of the datagram is going to be the Message header.

Directly after that will be the first sub-message header which will have the ID byte, Flags Byte and 2 sub-message length bytes.

So

ID will always be a hex value 0x01 to 0xFF

Flags will always be 0x00 to 0xFF??

Length I guess could be 0x0000 but probably not? ____(can it?)

The input I received suggest that the ID might be zeroed.

Can a valid sub-message ID be zero on a real valid packet??

If the sub-message length is zero that means that it would be directly followed by another sub-message or the end of the datagram. I am not sure why I would ever see that.

The next point of confusion I wanted to clarify in my mind is value of "a standard practice" of looking at a single sub-message out side of the other sub-messages. From the OMG standard its my understanding that the previous sub-messages can affect the meaning of subsequent ones - therefore, you need all of them to make sure you are ALWAYS getting the correct meaning in the sub-messages.

The last clarification I wanted was about byte order, If the flags bit 0x01 for endianness is set then the length and all the data of just that sub-message is going to be byte swapped. The next sub-message will be determined by its flags.

Thanks