Conversation
b17feb4 to
5af1075
Compare
|
|
07c233e to
2e2a02b
Compare
|
|
c2f5d24 to
ff30e56
Compare
|
Rebased on #394 |
a2fc9dc to
552df8a
Compare
|
Manually tested with revault/revault-gui#350, seems to be fine |
552df8a to
e764c4a
Compare
|
Rebased on master |
4d68ab4 to
cf176ac
Compare
The macro made it less readable for no big win, and it would become a mess with the Cancel batch.
cf176ac to
f5ece6f
Compare
|
Yay, this finally passed CI, i'll re-kick it to be sure it's not flaky. |
It's not that useful as listpresignedtxs is already exercised under all conditions by the functional tests. And it would be a MAJOR pain to move it to the multi-cancel logic. As the usefulness / time spent ratio is very low, just drop it. I'd still like to add it back if possible, but definitely not with hardcoded PSBTs.
It was specific to single-Cancel logic
f5ece6f to
ae94b9a
Compare
|
Updated the |
| } | ||
|
|
||
| /// Get the Unvault transaction for a given vault | ||
| pub fn unvault_tx( |
There was a problem hiding this comment.
I think, in the future we should move these helpers in a scripts or consensus module with a struct handling the descriptors and the secp_ctx, encapsulating the derivation etc. Revaultd would store this struct and we use it like this:
revaultd.consensus.unvault_tx(&vault.derivation_index) or revaultd.generator.unvault_tx(&derivation_index)
(I dont know yet which name)
edouardparis
left a comment
There was a problem hiding this comment.
ACK ae94b9a
Manually tested with the GUI
…nature 5314664 dummysigner: bump revault_tx and change api (edouard) 712a102 app: handle multiple cancel txs (edouard) ae1cccd bump revaultd (edouard) Pull request description: based on revault/revaultd#393 ACKs for top commit: edouardparis: ACK 5314664 Tree-SHA512: 30adc9198a8faf7ff0c13e9b0ca714a23252c847c93b5718cb916366eed1745ad1366575ba2624a1833ee0029d1e79f83d3ba937f1d9756fc234e19986de2930
This implements revault/practical-revault#119.
We adapt all components to use multi-Cancel:
Feedback on the new RPC API would be welcome: i preferred a list of Cancel instead of a mapping from feerate to Cancel.
We still don't chose the Cancel to broadcast depending on the fee estimation in
revault.