Author: ZmnSCPxj 2018-05-02 12:16:29
Published on: 2018-05-02T12:16:29+00:00
The email conversation is regarding the accurate summary of a paper on 'SIGHASH_NOINPUT' which removes all inputs of the transaction copy before signing/verifying. It sets them to a known constant and removing it could entail more costly serialization depending on the implementation. The spec of 'SIGHASH_NOINPUT' reveals that 'hashPrevouts', which normally commits to the other inputs, is blanked so that there is no commitment to them either. Therefore, 'SIGHASH_NOINPUT' implicitly has a 'SIGHASH_ANYONECANPAY'. It is suggested that using 'SIGHASH_ANYONECANPAY' on the update UTXO path will allow new inputs to be added to the transaction that will pay for the fee.
Updated on: 2023-05-24T23:50:23.308772+00:00