Add ELIP-deterministic-descriptor-blinding-key module #61
Conversation
LeoComandini
force-pushed
the
2023-11-17-deterministic-descriptor-blinding-key
branch
4 times, most recently
from
bf28d9b to
e3b775c
Compare
LeoComandini
changed the title
|
Implementing
Anyway independently from the choice this will be a separate commit, and perhaps a separate PR. |
LeoComandini
force-pushed
the
2023-11-17-deterministic-descriptor-blinding-key
branch
from
e3b775c to
358f663
Compare
|
Is there any way we can change the viewkey format from WIF (which includes a network) to hex? Or to bech32 with a custom |
I think that implies amending ELIP-150. Which I'm supportive of, and I think we're still in time to do it. |
|
I think we should. Having network bytes inside descriptors is a huge pain and I definitely didn't intend to cause that to happen. It is Thanksgiving here and I should not be at work :) but happy to review/merge such a change next week, if it makes sense on your end. |
LeoComandini
force-pushed
the
2023-11-17-deterministic-descriptor-blinding-key
branch
from
358f663 to
d79708b
Compare
ELIP changes: ElementsProject/ELIPs#9 Reference impl changes: #64 |
LeoComandini
force-pushed
the
2023-11-17-deterministic-descriptor-blinding-key
branch
from
d79708b to
7a32381
Compare
LeoComandini
force-pushed
the
2023-11-17-deterministic-descriptor-blinding-key
branch
from
7a32381 to
d2e8b26
Compare
Implementation for deterministic descriptor blinding keys.
LeoComandini
force-pushed
the
2023-11-17-deterministic-descriptor-blinding-key
branch
from
d2e8b26 to
34005bc
Compare
| Ok(Key::View(DescriptorSecretKey::Single(SinglePriv { | ||
| origin: None, | ||
| key: bitcoin::key::PrivateKey::new(secret_key, network), | ||
| }))) |
There was a problem hiding this comment.
In 34005bc:
Can you remind me why we would've had trouble parsing WIF with a network here? (I don't think we should support WIF anyway for various reasons, but I'm curious now.)
Otherwise ACK.
There was a problem hiding this comment.
The main issue is converting ct(elip151,desc) to ct(key,desc) where key is the actual view key.
If we use WIF, we should deduce the network from desc which however it might not be possible (eg if desc has only single public keys).
There are other options to handle the conversion, but we decided to go for the "hex" solution.
These were discussed here #61 (comment)
Other issues with WIF are that we might have mismatching network between the descriptor blinding key and the ordinary descriptor, and that single pub keys have no network, while single priv keys in WIF have a network so it's somehow inconsistent.
There was a problem hiding this comment.
Ok, the conversion thing makes sense, thanks!
The other issues, about mismatching networks etc., we have to deal with anyway with xprivs and xpubs. I think that most software "deals with it" by just ignoring the network bytes :)
Reference implementation for ElementsProject/ELIPs#8