Implement burn

[onchainguy-btc]

After my first burn PR (#2766) got a bit outdated, I created a new one. Tried to make as less changes as possible. This PR does not support burn messages yet. Hope we can get the general burning merged first and then I'm happy to add support for an optional burn message via data pushes.

[raphjaph]

Finally having a look at this. There's just two things but they probably require a bit of refactoring. Especially the new command ord wallet burn I think is important because burning is quite drastic and we want it to be a very distinct action.

[onchainguy-btc]

@raphjaph thanks for the feedback, will get that done asap!

[onchainguy-btc]

@raphjaph just implemented the changes. Let me know if I can improve on anything else. Can also write some more tests if you have ideas for some good ones.

[raphjaph]

In OP_RETURNs there is no dust_limit and I see that you correctly disable the check for that. One thing you could try doing is only burn the exact sat that the inscription is on instead of the whole UTXO. This will require some careful changes to the TransactionBuilder, which should then be tested as well in the unit tests of the transaction_builder.rs file.

[onchainguy-btc]

In OP_RETURNs there is no dust_limit and I see that you correctly disable the check for that. One thing you could try doing is only burn the exact sat that the inscription is on instead of the whole UTXO. This will require some careful changes to the TransactionBuilder, which should then be tested as well in the unit tests of the transaction_builder.rs file.

@raphjaph Do you think that's worth it? If I'm not mistaking doing so would add additional inputs to the transaction for extracting the sat. The fee for that could potentially be higher than what the inscription sats are worth.

As an alternative idea: What if we exit the ord wallet burn command with an error if the inscription is on a utxo with more than 10_000 sats. This way we can prevent burning large bitcoin amounts by mistake. If people still want to burn the inscriptions, they can first transfer it to a different, smaller output.

[raphjaph]

@raphjaph Do you think that's worth it? If I'm not mistaking doing so would add additional inputs to the transaction for extracting the sat. The fee for that could potentially be higher than what the inscription sats are worth.

That's true, then let's just leave it how it is for now.

As an alternative idea: What if we exit the ord wallet burn command with an error if the inscription is on a utxo with more than 10_000 sats. This way we can prevent burning large bitcoin amounts by mistake. If people still want to burn the inscriptions, they can first transfer it to a different, smaller output.

Sounds reasonble

[onchainguy-btc]

@raphjaph Do you think that's worth it? If I'm not mistaking doing so would add additional inputs to the transaction for extracting the sat. The fee for that could potentially be higher than what the inscription sats are worth.

That's true, then let's just leave it how it is for now.

As an alternative idea: What if we exit the ord wallet burn command with an error if the inscription is on a utxo with more than 10_000 sats. This way we can prevent burning large bitcoin amounts by mistake. If people still want to burn the inscriptions, they can first transfer it to a different, smaller output.

Sounds reasonble

Cool, will adapt to that!

[raphjaph]

Let me know when this is ready for review :)

[onchainguy-btc]

@raphjaph Ready for review ✅

[onchainguy-btc]

@raphjaph just saw you merged master? Was that on purpose? I can also just rebase if you like.

[raphjaph]

@onchainguy-btc don't worry about rebasing, I'm going to squash merge this anyway. Gonna have a look at this today or tomorrow.

The tests and linting also don't seem to pass yet. If you could fix, that would be great :)

You can run our CI if you install just and run just ci.

[raphjaph]

See comments above

[onchainguy-btc]

In my last commit I also protected last postage from being burned. Not 100% sure if that makes sense though.

[raphjaph]

@onchainguy-btc This looks almost ready to go! I just made a couple small changes and also ran the fuzzing tests on the transaction builder.

Could you add some more tests:

1. Multiple inscriptions in the same utxo and then try to burn and make sure it either fails or only burns the correct inscription.
2. An inscription and a rune in the same utxo and then it should fail.
3. Try ord wallet burn <RUNE_ID> and it should fail.

[onchainguy-btc]

@raphjaph just implemented the tests you requested ✅