Thiago Macieira

I will accept a patch that allows you to choose your allocator at compile time. I will not accept a patch that stores anything.

I don't want to add it to TinyCBOR. You don't have to obey the sorting order specified by the CBOR RFC. And, to be honest, what's in the RFC is...

Zero-copy API is already implemented in https://github.com/intel/tinycbor/compare/dev...thiagomacieira:dev. I've just never published it because it doesn't work really well for microcontrollers. Since I haven't got the necessary help from them for...

The problem is that TinyCBOR requires a single buffer with all of your data in it. On many small OSes (I know Zephyr), memory allocation is preferably done in chunks...

Thank you for the contribution. Let me see if I understand your use-case: you want to use the buffer like a circular buffer, so you can discard what has already...

If you're somewhere down in the code and don't have the original pointer, how do you know it's ok to reset? The only point in resetting would be after you've...

Your change resets the buffer but does not reset the state. It doesn't make sense to discard the data previously written, since you need it o make sense of what...

This function is fine. We just need to document that it does not reset the state. If you want a clean state, you can just use `cbor_encoder_init` again.

Do you just want me to update the tag listing?

The problem will be that we still need to decode and re-encode in order to keep the state in CborEncoder: it tries to keep the count of elements added, so...