Closed qingchuwudi closed 8 years ago
I believe the behavior is correct. The RFC 7252 says: The absence of the Payload Marker denotes a zero-length payload.
Is the AES encryption you are applying described somewhere?
Yes, you are right. I'm using coap in our product with a little adjustment. And the options no longer the way it is.
If you can share some details I'd be interested to implement provisions for your extension into gen-coap, but it must not violate the standard and should be useful also to other people. For example, it could automatically add random padding.
If payload is empty, it will give me an unexpected result which is not fllowed by 16#FF.
There is a problem. When encypt CoAP with AES , pads are necessary, But there is no tag to split them after decryption.
I think this is better: