Relax the AreInputsStandard() tests for P2SH transactions – allow any Script in a P2SH transaction to be relayed/mined, as long as it has 15 or fewer signature operations.
Rationale: https://gist.github.com/gavinandresen/88be40c141bc67acb247
I don’t have an easy way to test this, but the code changes are straightforward and I’ve updated the AreInputsStandard unit tests.
638- nArgsExpected += tmpExpected;
639+ if (Solver(subscript, whichType2, vSolutions2))
640+ {
641+ int tmpExpected;
642+ if (whichType2 == TX_SCRIPTHASH) // HASH160 <hash> OP_EQUAL
643+ tmpExpected = 1;
I worry that this will have the same effect as the fee drop - users and merchants will upgrade, miners won’t, and the resulting lack of mempool consensus makes double spending easier. When we did the fee drop we didn’t worry about this much because we’d done it before and it worked out OK, but this time seems to be different and miners aren’t bothering to upgrade, perhaps because 0.9 had nothing for them.
I don’t want to make things unnecessarily complicated, but I wonder if the change should be marked in the coinbase. Then IsStandard changes once a sufficient percentage of hashing power has upgraded. Alternatively maybe if @dgenr8 gets his double spend relaying code in, that might also help.
299+ oneAndTwo << OP_1 << key[0].GetPubKey() << key[1].GetPubKey() << key[2].GetPubKey();
300+ oneAndTwo << OP_3 << OP_CHECKMULTISIG;
301+ oneAndTwo << OP_2 << key[3].GetPubKey() << key[4].GetPubKey() << key[5].GetPubKey();
302+ oneAndTwo << OP_3 << OP_CHECKMULTISIG;
303+ oneAndTwo << OP_AND;
304+ keystore.AddCScript(oneAndTwo);
ACK modulo CHECKMULTISIG nit.
I had been working on my own version of this patch and came up with a pretty similar design. Tested this with real-world non-std P2SH transactions on mainnet, no issues found. I also can’t see any way that this could be a DoS issue - the ’limit sigops’ approach taken is exactly how my version worked as well.
Relax the AreInputsStandard() tests for P2SH transactions --
allow any Script in a P2SH transaction to be relayed/mined,
as long as it has 15 or fewer signature operations.
Rationale: https://gist.github.com/gavinandresen/88be40c141bc67acb247
I don't have an easy way to test this, but the code changes are
straightforward and I've updated the AreInputsStandard unit tests.
