Ergo Addresses Details

Development Research and Development
1

An address is a short string corresponding to some script used to protect a box. Unlike (hex-encoded) binary representation of a script, an address has some useful characteristics:

  • Integrity of an address could be checked, as it is incorporating a checksum.
  • A prefix of address is showing network and an address type. In particular, the network prefix prevents you from mistakenly sending mainnet tokens to the testnet address.
  • An address is using an encoding (namely, Base58) which is avoiding similarly l0Oking characters, friendly to double-clicking and line-breaking in emails.

An address is encoding network type , address type, checksum, and enough information to watch for a particular scripts.

Possible network types are:
Mainnet - 0x00
Testnet - 0x10

Address types are (semantics described below):

  • 0x01 - Pay-to-PublicKey(P2PK) address
  • 0x02 - Pay-to-Script-Hash(P2SH)
  • 0x03 - Pay-to-Script(P2S)

For an address type, we form content bytes as follows:

P2PK - serialized (compressed) public key
P2SH - first 192 bits of the Blake2b256 hash of serialized script bytes
P2S - serialized script

Address examples for testnet:

  • 3 - P2PK (3WvsT2Gm4EpsM9Pg18PdY6XyhNNMqXDsvJTbbf6ihLvAmSb7u5RN)
  • 8 - P2SH (8UmyuJuQ3FS9ts7j72fn3fKChXSGzbL9WC, 8LnSX95GAWdbDZWJZQ73Uth4uE8HqN3emJ)
  • ? - P2S (imdaM2NzX, z4hAmfvfSnQJPChMWzfBzJjpB8ei2HoLCZ2RHTaNArMNHFirdJTc7E)

for mainnet:

  • 9 - P2PK (9fRAWhdxEsTcdb8PhGNrZfwqa65zfkuYHAMmkQLcic1gdLSV5vA)
  • 2 - P2SH (25qGdVWg2yyYho8uC1pLtc7KxFn4nEEAwD, 23NL9a8ngN28ovtLiKLgHexcdTKBbUMLhH)
  • ? - P2S (7bwdkU5V8, BxKBaHkvrTvLZrDcZjcsxsF7aSsrN73ijeFZXtbj4CXZHHcvBtqSxQ)

Prefix byte = network type + address type
checksum = leftmost_4_bytes(blake2b256(prefix byte || content bytes))
address = prefix byte || content bytes || checksum

3 Likes
2

Good to see a lot of the technical details are now coming out in more detail :+1:

I think it’d be useful to have an explanation for why both P2SH and a P2S exist as separate address schemes and what are the use cases for each.

3

Thanks!

For example, sending 10 Ergs to a P2PK address usually means that a corresponding transaction
will contain a box in which 10 Ergs are locked by a public key encoded in the P2PK address. Similarly,
in case of a P2S address the box will be locked by a script encoded in the address. In the most complicated case of a P2SH script, the box will be protected by a special predefined script which is taking first 192 bits of Blake2b256 hash value for a script which should be shown by an input spending the box. P2PK and P2SH in Ergo are pretty similar to Bitcoin, while P2S in unique.

1 Like
4

Updated version is published on the blog https://ergoplatform.org/en/blog/2019_07_24_ergo_address/

1 Like
5

I think it will be good to give some use-cases where P2S addresses are necessary and P2SH won’t be sufficient (to motivate the need for such addressing scheme).

1 Like
6

For P2SH, some off-chain knowledge of a script which is the P2SH preimage is needed, for P2S not so. P2S could be useful for non-interactive protocols and contracts where you do not quite know a recipient. Maybe mixing could be a good case for that.