Encoding scheme for API identifiers and byte arrays

The æternity node uses base58Check encoding for identifiers such as account and contract public keys. Base58c is not well suited for encoding big byte arrays, so for general byte arrays base64 is used.

The base64 encoded data is treated in the same way as the base58Check encoded data, i.e., a checksum is postfixed to the binary data before encoding. The checksum is the first four bytes of the sha256 hash of the sha256 hash of the original byte array.

The following erlang snippet shows how the checksum can be computed, and how it is added to the byte array that is to be encoded.

add_check_bytes(Bin) when is_binary(Bin) ->
  <<Check:4/binary, _/binary>> = sha256(sha256(Binary)),
  <<Bin/binary, Check/binary>>.

Both identifiers and general byte arrays are prefixed with a two letter tag and a separator (_), describing the type of data that is encoded. For example an account pubkey could look as ak_2a1j2Mk9YSmC1gioUq4PWRm3bsv887MbuRVwyv4KaUGoR1eiKi.

The tags are described below, together with the encoding method of the tag.

Prefix	Encoding	Description
ak	base58	Account pubkey
ba	base64	Byte array
bf	base58	Block Proof of Fraud hash
bs	base58	Block State hash
bx	base58	Block transaction hash
cb	base64	Contract byte array
ch	base58	Channel
cm	base58	Commitment
ct	base58	Contract pubkey
kh	base58	Key block hash
mh	base58	Micro block hash
nm	base58	Name
ok	base58	Oracle pubkey
oq	base58	Oracle query id
or	base64	Oracle response
ov	base64	Oracle query
pi	base64	Proof of Inclusion
pp	base58	Peer pubkey
sg	base58	Signature
ss	base64	State trees
cs	base64	Contract calls state tree
ck	base64	Contract state key
cv	base64	Contract state value
st	base64	State
th	base58	Transaction hash
tx	base64	Transaction