What’s the opcode sequence for doing the reverse of Multisig?

The following scriptPubKey describes the desired contract:

OP_IF
    OP_DUP OP_HASH160 <Bob's pubKeyHash>
OP_ELSE
    <now+30 days> OP_CHECKLOCKTIMEVERIFY OP_DROP
    OP_DUP OP_HASH160 <Victor's pubKeyHash>
OP_ENDIF
OP_EQUALVERIFY OP_CHECKSIG

Bob can spend the output using the scriptSig <Bob's signature> <Bob's pubKey> OP_TRUE
Victor can spend the output after 30 days using the scriptSig <Victor's signature> <Victor's pubKey> OP_FALSE.

Fraggle

Posted 2019-03-28T00:06:08.073

Reputation: 288

http://bitcoin.sipa.be/miniscript/miniscript.html. Wouldn’t it rather be https://pastebin.com/pzQKN0ud ? – user2284570 – 2019-03-28T02:38:11.627

no I don't think you need that 1 and 2 there. In my version, Bob could pay by with <signature> <pubkey bob> 1 , and victor with <signature> <pubkey victor> 0 , but I don't doubt there is a better way – Fraggle – 2019-03-28T02:45:51.723

This is after https://github.com/bitcoin/bips/blob/master/bip-0065.mediawiki#Escrow but which is for MultiSig in the case of Victor.

– user2284570 – 2019-03-28T02:47:32.587

1You don't need 1 and 2. Those are only required for OP_CHECKMULTISIG. This is just an OP_CHECKSIG which does not need a threshold number. – Andrew Chow – 2019-03-28T02:50:08.557

Yeah you don't need multi-sig right. Just bob can spend funds at any time, victor can spend funds after some time as long as bob hasn't already spent them. So no 1,2, multi-sig thing. Just Bob can spend, or Victor can spend after a time has past. – Fraggle – 2019-03-28T02:51:33.303

@AndrewChow yes, I want to avoid MultiSig. Now just a question : how can I compile this to something ? And how to test the resulting script on bitcoin testnet ? – user2284570 – 2019-03-28T02:55:09.147

Check out https://github.com/kallewoof/btcdeb. That project produces two tools. btcdeb lets you give a script and execute it in order to debug the script and learn how it works. btcc can take a script written out in english and compile it to the actual bytes for the script.

– Andrew Chow – 2019-03-28T03:09:47.153

@AndrewChow I mean something working online. I want to see what this gives in a block explorer, so I definitely need to broadcast on testnet. Like if Bob or Victor address would be changed as result (I mean the transaction would create virtual addresses). – user2284570 – 2019-03-28T13:56:37.683

There is no such thing as "virtual addresses". Bitcoin does not actually use addresses, they are just abstractions for humans. Such a script does not encode addresses and block explorers would not show that a transaction is sending to two addresses with this script. All they would show is a single address corresponding to the P2SH address that has this script as its redeemScript. – Andrew Chow – 2019-03-28T14:28:30.943

@AndrewChow ok… I found out what I was really meaning… I mean is it possible to use this in a pay‑to‑public‑key‑hash instead of a pay‑to‑public‑script‑hash transaction ? I definitely need the payment to appear to be made to at least Victor address or Bob address in the block explorer instead of script hash… – user2284570 – 2019-04-04T23:54:40.143

No, you can't. It is a script, a script is not a key. You cannot make a script be a key. – Andrew Chow – 2019-04-05T01:33:22.080

@AndrewChow I was thinking this because checksig is performed after each address in the similar case of https://github.com/bitcoin/bips/blob/master/bip-0065.mediawiki#trustless-payments-for-publishing-data. Of course, this means there’s no way Bob or Victor Balance would be changed in a block explorer.

– user2284570 – 2019-04-05T01:53:17.597

@user2284570 so perhaps you want the output to not be a p2sh format, but rather have this script visible (when decoded from hex form) in the output? Anyway I think I answered the original question. Consider opening a new one for further refinements and mark this one as answered. – Fraggle – 2019-04-16T22:20:39.870

What’s the opcode sequence for doing the reverse of Multisig?

Answers