Wallet: Support disabling implicit Segwit operation #12146
Conversation
|
I see little reason to do this. While we shouldn't make a habit of recognizing outputs to mutations of addresses we've given out, if someone sends you funds to a SegWit address, and you wish to refuse it, you're welcome to not honor the payment. Instead, we should be moving towards longer-term goals like splitting the HD chain across address types and recognizing payment only to addresses we've given out in a non-backup-breaking way. |
|
Without this, there is currently no way to even detect they were sent to a Segwit mutation of an address. |
|
Much discussion of this PR here: https://botbot.me/freenode/bitcoin-core-dev/2018-01-11/?msg=95652211&page=5 I'd support fixing the general case of address malleation. I don't think this is the correct approach, and don't think it should hold up v0.16, for the reasons enumerated in IRC. Concept NACK from me. |
ca2751e to
4ed8341
Compare
4ed8341 to
cc78a98
Compare
|
Rebased (which simplified it quite a bit). |
|
Closing for now. Let me know when you want to continue working on this. |
This optional feature (disabled by default) leaves Segwit wallet support mostly as-is, except:
OUTPUT_TYPE_DEFAULTis changed to an actualOutputTypevalue, in expectation of reuse by #12119 as well as enabling the possibility of having the implicit segwit flag be set per-wallet once we have runtime wallet loading.