You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is your feature request related to a problem? Please describe.
Zebra has its own signature hash (sighash) implementation for V4 transactions and earlier. For V5, it will use librustzcash (#2051) . Therefore, we'll have two different code paths.
Describe the solution you'd like
Also use librustzcash for V4 and earlier.
Describe alternatives you've considered
We could do nothing.
Additional context
The downside of this idea is additional overhead of reserializing and deserializing with librustzcash. However, we already accepted that for V5 transactions, and it could be improved in this issue or in a later one (by e.g. converting to librustzcash while deserializing, and not when computing the signature hash, so it will only happen once).
#2051 already adds a test for comparing librustzcash sighashes with zebra sighashes, and it passes (with a small bug fix in librustzcash that is being discussed with its devs). Therefore we already have some assurance that this change won't break anything.
The text was updated successfully, but these errors were encountered:
dconnolly
changed the title
Also use librustzcash for signature hashes in V4 transactions and earlier
Also use librustzcash for signature hashes (sighash) in V4 transactions and earlier
May 25, 2021
Is your feature request related to a problem? Please describe.
Zebra has its own signature hash (sighash) implementation for V4 transactions and earlier. For V5, it will use librustzcash (#2051) . Therefore, we'll have two different code paths.
Describe the solution you'd like
Also use librustzcash for V4 and earlier.
Describe alternatives you've considered
We could do nothing.
Additional context
The downside of this idea is additional overhead of reserializing and deserializing with librustzcash. However, we already accepted that for V5 transactions, and it could be improved in this issue or in a later one (by e.g. converting to librustzcash while deserializing, and not when computing the signature hash, so it will only happen once).
#2051 already adds a test for comparing librustzcash sighashes with zebra sighashes, and it passes (with a small bug fix in librustzcash that is being discussed with its devs). Therefore we already have some assurance that this change won't break anything.
The text was updated successfully, but these errors were encountered: