I don't think we are consistent about this. For example we define utf8 as accepting only ascii strings and we don't handle non ascii strings in encode_for_hash, even though we have an utf8 encoding. Also, I think "ASCII character sequence" is what we meant as "ascii string" but is not everywhere clear.
Not sure what is best here. Should we assume that all the messages are already an octet string (and just mention some ways for transforming a char sequence to octets)?? So the only think that is left is DST's that we can encode with utf8??
I think this is what most crypto specs do, so i would go with that.
I don't think we are consistent about this. For example we define utf8 as accepting only ascii strings and we don't handle non ascii strings in
encode_for_hash
, even though we have an utf8 encoding. Also, I think "ASCII character sequence" is what we meant as "ascii string" but is not everywhere clear.Not sure what is best here. Should we assume that all the messages are already an octet string (and just mention some ways for transforming a char sequence to octets)?? So the only think that is left is DST's that we can encode with utf8??
I think this is what most crypto specs do, so i would go with that.