How do you want this to work? Request bodies can be many hundreds of MBs or GBs large. They can be a binary encoding. They can have malicious content if pasted into the wrong places. It’s inefficient and expensive to read and buffer the body. I’ve never seen a good solution for generalized request body logging. Nor have I ever found it a good idea for the proxy to do this; usually the application should do it if it is needed.
Bytes or characters? I see you’re from a locale with multi-byte characters, which means we now have to consider the encoding of the body and decode it. Any suggestions how that would be done, and how we can read the body efficiently? It’ll essentially require buffering all the response bodies.