P2SH related RPC improvements #2830
Conversation
|
ACK |
|
Note: should be merged after sipa's watch-only address stuff is merged, and modified to work with it. |
|
Related to that, is addredeemscript useful, if bitcoind has no logic to actually spend coins sent to anything but what is creatable through createmultisigaddress? |
|
ACK on the other changes. |
|
@sipa Actually bitcoind can spend any P2SH-encoded scriptPubKey that it knows how to spend, so you can (currently) put a pay-to-pubkey or a pay-to-pubkey-hash in redeemScript and things will work fine. In any case I think the feature of adding arbitrary redeemScripts makes the most sense in the context of watch-only-wallets where the actual mechanism to spend the coins can be handled elsewhere. |
|
@petertodd Can you remove addredeemscript for now since it seems like the watch stuff is currently delayed? |
|
Automatic sanity-testing: PASSED, see http://jenkins.bluematt.me/pull-tester/22536422878d05a7229f74726148dd50051c883f for binaries and test log. |
|
ACK, changes look good and I tested all 0,1, and 2 byte scripts in decodescript and tried out validateaddress on some p2sh and non-p2sh addresses. |
P2SH related RPC improvements
This is a hackish version of dashpay#2824, meant for 0.13.x only. The reason for this hackish version is that the code has diverged quite a bit and its not worth the effort to backport the hardening code. Even though 0.13.x included a lot of fixes for sporks handling and syncing, I still feel more safe with hardening the spork15 block height. If something goes wrong with spork syncing (e.g. its slower then the first DIP2/3 block arrives), the whole sync process will fail otherwise.
Some currently missing features like decoding a script and adding a specific P2SH redeemScript to the wallet.